Contents

Introduction· 1

Version information· 1

Version number 1

Version history· 1

Hardware and software compatibility matrix· 2

ISSU upgrade type matrix· 4

Upgrade restrictions and guidelines· 5

Hardware feature updates· 5

R6715· 5

E6713P01· 5

E6713· 5

E6712P01· 6

E6712· 6

E6711· 6

R6710· 6

E6706· 6

E6705· 6

E6702· 6

F6701L01· 6

Software feature and command updates· 6

MIB updates· 7

Operation changes· 8

Operation changes in R6715· 8

Operation changes in E6713P01· 9

Operation changes in E6713· 9

Operation changes in E6712P01· 12

Operation changes in E6712· 12

Operation changes in E6711· 12

Operation changes in R6710· 14

Operation changes in E6706· 15

Operation changes in E6705· 16

Operation changes in E6702· 17

Operation changes in F6701L01· 17

Restrictions and cautions· 17

Restrictions· 17

Cautions· 19

Licensing· 19

About licensing· 19

Registering and installing licenses· 19

Obtaining license server software and documentation· 19

Open problems and workarounds· 20

List of resolved problems· 23

Resolved problems in R6715· 23

Resolved problems in E6713P01· 26

Resolved problems in E6713· 27

Resolved problems in E6712P01· 34

Resolved problems in E6712· 34

Resolved problems in E6711· 36

Resolved problems in R6710· 38

Resolved problems in E6706· 45

Resolved problems in E6705· 47

Resolved problems in E6702· 50

Resolved problems in F6701L01· 52

Troubleshooting resources· 52

Related documentation· 52

Documentation set 52

Obtaining documentation· 52

Technical support 53

Appendix A Feature list 54

Hardware features· 54

Software features· 56

Appendix B Fixed security vulnerabilities· 59

Fixed security vulnerabilities in R6715· 59

Fixed security vulnerabilities in E6713P01· 60

Fixed security vulnerabilities in E6713· 62

Fixed security vulnerabilities in E6712· 62

Fixed security vulnerabilities in E6711· 62

Fixed security vulnerabilities in R6710· 62

Fixed security vulnerabilities in E6705· 63

Fixed security vulnerabilities in E6702· 63

Appendix C Upgrading software· 64

System software file types· 64

System startup process· 64

Upgrade methods· 65

Upgrading from the CLI 66

Verifying device status· 66

Preparing for the upgrade· 66

Downloading software to the master switch· 68

Upgrading the software images· 70

Installing a patch package· 71

Upgrading from the Boot menu· 72

Prerequisites· 72

Accessing the Boot menu· 73

Accessing the extended Boot menu· 74

Using TFTP to upgrade software images through the management Ethernet port 76

Using FTP to upgrade software through the management Ethernet port 77

Using XMODEM to upgrade software through the console port 79

Using TFTP to upgrade Boot ROM through the management Ethernet port 83

Using FTP to upgrade Boot ROM through the management Ethernet port 85

Using XMODEM to upgrade Boot ROM through the console port 86

Managing files from the Boot menu· 90

Displaying all files· 90

Deleting files· 91

Changing the attribute of software images· 91

Handling software upgrade failures· 93

List of tables

Table 1 Version history· 1

Table 2 Hardware and software compatibility matrix· 2

Table 3 ISSU compatibility list 4

Table 4 MIB updates· 7

Table 5 S9850_6850 series hardware features· 54

Table 6 Software features of the S6850&S9850 series· 56

Table 7 Shortcut keys· 74

Table 8 Extended Boot ROM menu options· 75

Table 9 TFTP parameter description· 76

Table 10 FTP parameter description· 78

Table 11 TFTP parameter description· 84

Table 12 FTP parameter description· 85

Introduction

This document describes the features, restrictions and guidelines, open problems, and workarounds for version R6715P01. Before you use this version on a live network, back up the configuration and test the version to avoid software upgrade affecting your live network.

Use this document in conjunction with H3C S9850_6850-CMW710-R6715P01 Release Notes (Software Feature Changes) and the documents listed in “Related documentation”.

Version information

Version number

H3C Comware Software, Version 7.1.070, Release 6715P01

Note: You can see the version number with the display version command in any view. Please see Note ②.

Version history

Table 1 Version history

Hardware and software compatibility matrix

Table 2 Hardware and software compatibility matrix

Note① LS-6850-XXXX represents a product code, which is printed on the lower right corner of the bar code label on the rear panel or upper panel of the device.

To display version information for the system software and Boot ROM of S6850-56HF:

<H3C>display version                                                         

H3C Comware Software, Version 7.1.070, Release 6715P01                ------- Note② 

Copyright (c) 2004-2024 New H3C Technologies Co., Ltd. All rights reserved.    

H3C S6850-56HF uptime is 0 weeks, 0 days, 23 hours, 2 minutes                  

Last reboot reason : User Reboot                                               

Boot image: flash:/s9850_6850-cmw710-boot-r6715p01.bin                            

Boot image version: 7.1.070, Release 6715P01                                      

  Compiled Jun 06 2024 16:00:00                                                

System image: flash:/s9850_6850-cmw710-system-r6715p01.bin                        

System image version: 7.1.070, Release 6715P01                                      

  Compiled Jun 06 2024 16:00:00                                                 

MPU(S) Slot 1:                                                                 

Uptime is 0 weeks,0 days,23 hours,2 minutes                                     

H3C S6850-56HF LPU with 1 Processor(s)                                   

BOARD TYPE:         S6850-56HF                                                 

DRAM:               8192M bytes                                                 

FLASH:              3630M bytes                                                

NVRAM:              0K bytes                                                   

PCB 1 Version:      VER.A                                                      

PCB 2 Version:      VER.A                                                      

PCB 3 Version:      VER.A                                                      

Basic    BootWare Version: 205                                     ------ Note③  

Extended BootWare Version: 205                                     ------ Note③ 

CPLD 1 Version:     002                                                        

CPLD 2 Version:     002                                                         

CPLD 3 Version:     002                                                        

CPLD 4 Version:     002                                                        

FPGA 1 Version:     001                                                        

Release Version:    H3C S6850-56HF-6715P01                                        

Patch Version:      None                                                       

Reboot Cause:       UserReboot                                                  

[SubSlot 0] 48SFP28 + 8QSFP28 + 2SFP   

ISSU upgrade type matrix

ISSU provides two upgrade types: compatible upgrade and incompatible upgrade. Table 3 provides the approved ISSU upgrade types only between the current version and the history versions within the past 18 months. This matrix does not include history versions that are 18 months earlier than the current version, for which, no ISSU upgrade verification is performed.

For more information about ISSU, see the fundamental configuration guide for the device.

Table 3 ISSU compatibility list

Upgrade restrictions and guidelines

1.       To prevent routing neighbor flapping when upgrading the software from E6702 to E6706 by using step-by-step ISSU, use the irf mac-address persistent always command to enable the IRF bridge MAC address to be permanent.

2.       To prevent routing neighbor flapping when upgrading the software from F6701L01 to E6702 by using step-by-step ISSU, use the irf mac-address persistent always command to enable the IRF bridge MAC address to be permanent.

3.       When a software version earlier than R6710 is upgraded to R6710 through an ISSU, VXLAN Layer 3 traffic loss persists for longer than 2 seconds. In a lab environment, traffic loss persists for 21 seconds during a one-click upgrade and 35 seconds during a distributed upgrade.

4.       Release 6710 and earlier versions does not support upgrading the switch to E6711 or later through ISSU.

5.       Before you upgrade a version earlier than E6705 to Release 6710, identify whether a license has been activated or installed. If yes, the upgrade will cause the license to be lost or to fail to be installed. To avoid this issue, first uninstall the license and obtain uninstall key, and then re-activate the license (the DID file must be obtained again) in Release 6710 through license transfer.

If you upgrade to R6710HS03, you can use the boot-loader file command. The license will not be lost or fail to be installed.

6.       Before you upgrade or down upgrade a version between E6705 and Release 6710 or install or uninstall R6710HS03 or a later version, identify whether a license has been activated or installed. If yes, first uninstall the license and obtain uninstall key, and then re-activate the license (the DID file must be obtained again) in the new version through license transfer.

7.       The S6850-56HF-SAN is not supported in software versions earlier than E6711.

8.       The S6850-56HF-IM is not supported in software versions earlier than E6712P01.

9.       The S6850-56HF (with product code LS-6850-56HF-H3) and S6850-56HF-CP are not supported in software versions earlier than E6713.

10.    As from E6711, password recovery capability is enabled by default in the initial configuration instead of the factory defaults on the device.

11.    As from R6715P01, lossless related features are no longer restricted to licenses and are supported by default. The display license feature command no longer displays information about lossless related features. You can still activate and install lossless licenses, without affecting the lossless features. To roll back to the R6715 version or earlier with feature-based licensing, remove the IPCC-related settings, activate lossless licenses, and then reconfigure IPCC.

12.    The S6850-2C (with product code LS-6850-2C-H1) are not supported in software versions earlier than R6715P01.

Hardware feature updates

R6715P01

Added support for the SFP-25G-CSR-MM850 transceiver module.

Added support for the S6850-2C (with product code LS-6850-2C-H1).

R6715

Added support of S6850-56HF(product code LS-6850-56HF-H3) for FAN-40B-1-D modules.

E6713P01

None.

E6713

Added support for the S6850-56HF (with product code LS-6850-56HF-H3).

Added support for the S6850-56HF-CP.

Added support for the 25G/10G autosensing SFP-XG-SX-MM850-D transceiver module.

E6712P01

Added support for the S6850-56HF-IM switch.

Added support for the QSFP-100G-SR4-MM850-Y5 liquid cooled optical transceiver module.

Added support for the SFP-25G-D-CAB-3M-Y 3 m (9.84 ft) liquid cooled DAC cable.

Added support for the SFP-25G-D-CAB-5M-Y 5 m (16.40 ft) liquid cooled DAC cable.

E6712

The transceiver module QSFP-40G-LX4-WDM1300 was supported.

E6711

Added support for S6850-56HF-SAN.

R6710

Added support for S9850-4C (LS-9850-4C-H1).

E6706

Added support for S9850-32H (LS-9850-32H-H1).

E6705

None.

E6702

None.

F6701L01

First release.

Software feature and command updates

For more information about the software feature and command update history, see H3C S9850_6850-CMW710-R6715P01(Software Feature Changes).

MIB updates

Table 4 MIB updates

Operation changes

Operation changes in R6715P01

[202402010243] Capitalized the first letter of the parameter comments for each option in hardware-resource commands

Before modification: The first letter of the parameter comments for each option in the hardware-resource commands is not capitalized.

After modification: The first letter of the parameter comments for each option in the hardware-resource commands is capitalized.

Related commands:

·         hardware-resource clock-mode

·         hardware-resource flex-mode

·         hardware-resource routing-mode

[202403270041] Lossless related features are no longer restricted to licenses and are supported by default

As from R6715P01, lossless related features are no longer restricted to licenses and are supported by default. The display license feature command no longer displays information about lossless related features. You can still activate and install lossless licenses, without affecting the lossless features. To roll back to the R6715 version or earlier with feature-based licensing, remove the IPCC-related settings, activate lossless licenses, and then reconfigure IPCC.

[202310251525] The identifier character for commands delayed to take effect is changed

Before modification: The identifier character for commands delayed to take effect is a tilde (~).

After modification: The identifier character for commands delayed to take effect is an asterisk (*).

[202406120202] In PIM-SM, traffic can still be forwarded after the ingress replication MVXLAN configuration is deleted

Before modification: After the ingress replication MVXLAN configuration is deleted, traffic cannot be forwarded in PIM-SM, but traffic can still be forwarded in PIM-SSM.

After modification: After the ingress replication MVXLAN configuration is deleted, traffic can still be forwarded in both PIM-SM and PIM-SSM.

Operation changes in R6715

[202402061133] Support for executing the lacp transparent enable command through NETCONF

Before modification: The lacp transparent enable command cannot be executed through NETCONF.

After modification: The lacp transparent enable command can be executed through NETCONF.

[202207120205]Changed the execution of the undo mac-address static source-check enable command on a Layer 2 aggregate interface configured as a peer-link interface.

Before modification: To successfully forward Layer 3 traffic over the peer link, you must manually execute the undo mac-address static source-check enable command on a Layer 2 aggregate interface configured as a peer-link interface.

After modification: After a Layer 2 aggregate interface is configured as a peer-link interface, the system automatically executes the undo mac-address static source-check enable command on it

[202402260239]Setting the maximum number of supported BFD sessions

Before modification: The device supports a maximum of 256 BFD sessions.

After modification: The device supports a maximum of 512 BFD sessions.

Remarks: In software BFD mode, set the BFD detection time to at least 300ms*5 if the number of BFD sessions is greater than 64, and set the BFD detection time to at least 400ms*5 if the number of BFD sessions is greater than 256 as a best practice. Since the BFD protocol is time-sensitive, high CPU pressure on the device can lead to BFD flapping, which affects network stability. To avoid this issue, increase the BFD detection time as much as possible to enhance network stability.

In hardware BFD mode, set the BFD detection time to at least 100ms*3 as a best practice. Additionally, the hardware BFD mode has certain usage restrictions. For more information about these restrictions, see hardware BFD configuration in the configuration guides.

[202402220271] Changed the MAC authentication specification from 8K to 2K

Before modification: The MAC authentication specification is 8K.

After modification: The MAC authentication specification is 2K.

[202210290537] ARP/ND learning  and remote MAC address learning are disabled for automatic tunnels by default in an EVPN network

Before modification: To disable ARP/ND learning for tunnels in an EVPN network, execute the vxlan tunnel arp-learning disable/vxlan tunnel nd-learning disable command. To disable remote MAC address learning for tunnels in an EVPN network, execute the vxlan tunnel mac-learning disable command.

After modification: ARP/ND learning and remote MAC address learning are disabled for automatic tunnels by default in an EVPN network.

NOTE: This change only applies to automatic tunnels in an EVPN network, and does not affect manually created VXLAN tunnels.

[202209270016] MAC address learning is disabled by default

Before modification: MAC address learning is enabled by default.

After modification: MAC address learning is disabled by default.

[202401171983] uRPF loose mode and the default route configuration

Before modification: If you enable the ip urpf loose command first, and then configure a blackhole static route or configure a default route pointing to the management interface or gateway, packets with mismatched source IPs will be forwarded through the default route.

After modification: If you enable the ip urpf loose command first, and then configure a blackhole static route or configure a default route pointing to the management interface or gateway, packets with mismatched source IPs will not be forwarded.

Operation changes in E6713P01

[202312052129]When the free-memory space decreases below 224 MB, the minor, severe, and critical alarms are triggered. When a critical alarm occurs, the device generates a log message and then restarts.

Before modification: The minor, severe, and critical free-memory thresholds are 448 MB, 224 MB, and 128 MB, respectively. When the free-memory space decreases below 224 MB, the device generates a severe alarm and then restarts. The device generates a log message only when the free-memory space decreases below the critical free-memory threshold.

After modification: All free-memory thresholds (minor, severe, critical) are set to 224 MB. When the free-memory space decreases below 224 MB, it triggers all alarms. If a critical alarm occurs, the device logs the event and then restarts.

[202310131260]gRPC sensor path buffermonitor/commbufferusages can collect the cumulative numbers of multicast and unicast packets

Before modification: The information collected by gRPC sensor path buffermonitor/commbufferusages does not contain the MulticastTransPkts (the cumulative number of multicast packets) and UnicastTransPkts (the cumulative number of unicast packets) fields.

After modification: The information collected by gRPC sensor path buffermonitor/commbufferusages has the MulticastTransPkts and UnicastTransPkts fields.

[202309151956] Interface packet rate statistics was added to the information collected by gRPC

·         Before modification: gRPC cannot collect interface packet rate statistics.

·         After modification: The interface information collected by gRPC includes interface packet rate statistics.

[202309132125] ARP fast update for MAC address moves (configured by using the mac-address mac-move fast-update command) is enabled by default

·         Before modification: ARP fast update for MAC address moves (configured by using the mac-address mac-move fast-update command) is disabled by default.

·         After modification: ARP fast update for MAC address moves (configured by using the mac-address mac-move fast-update command) is enabled by default.

Operation changes in E6713

[202303170413] hh3c-bgp4v2.mib is supported. Traps are generated upon IPv4 IPv6 or peer state changes

Before modification: hh3c-bgp4v2.mib is not supported. The system only generates traps about IPv4 peer state changes.

After modification: hh3c-bgp4v2.mib is supported. The system can generate traps about IPv4 or IPv6 peer state changes.

[202305170768] Interface traffic statistics are also displayed if the device obtains the traffic from ifInUcastPkts in the MIB

Before modification: Interface traffic statistics cannot be displayed if the device obtains the traffic from ifInUcastPkts in the MIB.

After modification: When the statistics enable command is executed in VSI view, interface traffic statistics are also displayed if the device obtains the traffic from ifInUcastPkts in the MIB.

[202306290192] Both the DRNI/Base table and MLAG/Base table of NETCONF support the ExtraVlan and ReservedVlanList fields

Before modification: The fields in the DRNI/Base table and MLAG/Base table are inconsistent in NETCONF. The DRNI/Base table contains only the ExtraVlan field. The MLAG/Base table contains only the ReservedVlanList field.

After modification: In NETCONF, both the DRNI/Base table and MLAG/Base table support the ExtraVlan and ReservedVlanList fields.

[202306290462] The MLAG/Keepalive table of NETCONF contains the TrackID attribute column

Before modification: The MLAG/Keepalive table of NETCONF does not contain the TrackID attribute column.

After modification: The MLAG/Keepalive table of NETCONF contains the TrackID attribute column.

[202307010905] Data from the device/transceivers and device/transceiverchannels sensor paths can be pushed in GPB encoding format

Before modification: When the device uses the three-layer telemetry data model architecture to push data, it does not support using GPB to encode data from the device/transceivers and device/transceiverchannels sensor paths.

After modification: When the device uses the three-layer telemetry data model architecture to push data, it supports using GPB to encode data from the device/transceivers and device/transceiverchannels sensor paths.

[202307011061] Aggregate interfaces support configuring global CRC error packet alarm parameters

Before modification: Aggregate interfaces do not support configuring global CRC error packet alarm parameters by using the ifmonitor crc-error command.

After modification: Aggregate interfaces support configuring global CRC error packet alarm parameters by using the ifmonitor crc-error command.

[202307071286] When all the specified temperature alarm thresholds are invalid, the value range for the temperature sensor number is not displayed

Before modification: When all the specified temperature alarm thresholds are invalid, the command output from the temperature-limit { hotspot | inflow | outflow } ? command displays the value range for the temperature sensor number as an invalid value.

After modification: When all the specified temperature alarm thresholds are invalid, the command output from the temperature-limit { hotspot | inflow | outflow } ? command does not display the value range for the temperature sensor number.

[202308070845] Support for statistics for all types of tunnel interfaces, including GRE, IPv4 over IPv4, and IPv4 over IPv6 tunnel interfaces

Before modification: You cannot use the display interface tunnel command to view traffic statistics for GRE, IPv4 over IPv4, and IPv4 over IPv6 tunnel interfaces.

After modification: You can use the display interface tunnel command to view traffic statistics for GRE, IPv4 over IPv4, and IPv4 over IPv6 tunnel interfaces.

[202308230421] The device restarts when a level-2 threshold is reached, and generates a log message about the restart cause

Before modification: The device does not restart when a level-2 threshold is reached.

After modification: The device restarts when a level-2 threshold is reached, and generates a log message about the restart cause.

[202309132121] The VLAN interface limit is 4K

Before modification: The VLAN interface limit is 2K.

Before modification: The VLAN interface limit is 4K.

[202309211555] Supported the [undo] snmp-agent trap enable igmp/mld commands in private-mode system view

Before modification: The [undo] snmp-agent trap enable igmp and [undo] snmp-agent trap enable mld commands cannot be executed in private-mode system view.

After modification: The [undo] snmp-agent trap enable igmp and [undo] snmp-agent trap enable mld commands can be executed in private-mode system view.

[202309211558] Supported the [undo] snmp-agent trap enable pim/pim6 commands in private-mode system view

Before modification: The [undo] snmp-agent trap enable pim and [undo] snmp-agent trap enable pim6 commands cannot be executed in private-mode system view.

After modification: The [undo] snmp-agent trap enable pim and [undo] snmp-agent trap enable pim6 commands can be executed in private-mode system view.

[202304120566] Supported a maximum of four DSCP mappings in the priority-flow-control dscp-mapping command

Before modification: A maximum of two DSCP mappings can be configured.

After modification: A maximum of four DSCP mappings can be configured.

[202303281927] The device monitors incorrect TPCE messages on ports and generates log messages and traps for notification

Before modification: The device does not monitor incorrect TPCE messages on ports.

After modification: The device monitors incorrect TPCE messages on ports, counts incorrect TPCE messages if any, and generates log messages and traps for notification.

[202307310775] Optimized health check for IRF members

Before modification: The health check feature does not include checks on MMUs of IRF members, port down, LACP selection, memory, CPLD status, and PHY status.

After modification: The health check feature adds checks for MMU of IRF members, port down, LACP selection, memory, CPLD status, and PHY status as follows:

·         MMU fault check: Checks the number of cells occupied by MMUs for all the ports on the device. When the cell count exceeds the threshold, the health value increases by 1.

·         Port down check: Checks the status of all the ports on the device. If all the ports are down during the device operation, the health value increases by 3.

·         LACP selection check: Checks the status of port LACP selection on the device. If the number of unselected ports is equal to or greater than 48, the health value increases by 1.

·         Memory check: Checks the free memory space. If the memory space is smaller than 128M, the health value increases by 1.

·         CPLD check: Checks CPLD components on the device. If an incorrect CPLD register value is incorrect, the health value increases by 1.

·         PHY check: Checks PHY components on the device. If an incorrect PHY register value is incorrect, the health value increases by 1.

[202302170010] Changed the dscp dscp-value option in the priority-flow-control dscp-mapping command from required to optional

Before modification: The dscp dscp-value option is required in priority-flow-control dscp-mapping command.

After modification: The dscp dscp-value option is optional in priority-flow-control dscp-mapping command. If this option is not specified, the DSCP value of packets is not changed.

[202302081531/202306151058] Change to the default NSR enabling status

Before modification: By default, NSR is disabled in IS-IS/OSPF/OSPFv3/RIP/RIPng/BGP/RIB/LDP view.

After modification: By default, NSR is enabled in IS-IS/OSPF/OSPFv3/RIP/RIPng/BGP/RIB/LDP view.

[202305081549] Command effect on authentication failed and fail-permit users changed

Before modification: The following commands take effect only on users passing authentication and they do not take effect on authentication failed users and fail-permit users: mac-authentication offline-detect enable, mac-authentication packet-detect enable, dot1x offline-detect enable, and dot1x packet-detect enable.

After modification: The following commands take effect on users passing authentication, authentication failed users and fail-permit users: mac-authentication offline-detect enable, mac-authentication packet-detect enable, dot1x offline-detect enable, and dot1x packet-detect enable.

Authentication failed users and fail-permit users: Users added to the critical or Auth-Fail VLAN, VSI, or microsegment after they failed the authentication on the device where dot1x critical, dot1x guest, dot1x auth-fail, mac-authentication guest, and mac-authentication critical features are configured.

[202304181578] Changes to supported algorithms in FIPS mode

Before modification: In FIPS mode, the HMAC-SHA-1 algorithm can be specified in the ntp-service authentication-keyid and sntp authentication-keyid commands for authentication.

After modification: No HMAC-SHA-1 algorithm can be specified in the ntp-service authentication-keyid and sntp authentication-keyid commands for authentication.

Operation changes in E6712P01

None.

Operation changes in E6712

[202303171926] ip binding vpn-instance command on Layer 3 interfaces with private VLANs configured

Before modification: Layer 3 interfaces with private VLAN enabled do not support the ip binding vpn-instance command.

After modification: Layer 3 interfaces with private VLAN enabled support the ip binding vpn-instance command.

[202305250499]Maximum number of supported PBR policy nodes

Before modification: The device supports configuring a maximum of 50 PBR policy nodes.

After modification: The device supports configuring a maximum of 1024 PBR policy nodes.

[202304111948] Support of switches for sampling packets according to the entered sampling rate

Before modification: A sampler supports only the sampling rate that is 2 to the nth power, where n is the entered sampling rate. One packet is sampled from every 2 to the nth power packets.

After modification: A sampler supports sampling packets according to the entered sampling rate.  For example, if you set the sampling rate to 100, one packet is sampled from every 100 packets.

[202208241350] The Instances table of MVPN NETCONF and the StaticGroups table of IGMP NETCONF changed

Before modification:

·         The PmsiTunnelType column of the Instances table cannot be configured as 5 (ingress replication MVXLAN).

·         The StaticGroups table (static group member) cannot be deployed.

After modification:

·         The PmsiTunnelType column of the Instances table can be configured as 5 (ingress replication MVXLAN).

·         The StaticGroups table (static group member) can be deployed.

Operation changes in E6711

[202208170830] Loopback testing

Before modification: Loopback testing can be enabled on an Ethernet interface by using the loopback { external | internal } command.

After modification: Loopback testing can be enabled on an Ethernet interface by using the loopback-test { external | internal } command.

[202211070131] Forwarding of unknown frames after the MAC learning limit on a Layer 2 aggregate interface is reached

Before modification: A Layer 2 aggregate interface does not support forwarding unknown frames after the MAC learning limit on the Layer 2 aggregate interface is reached.

After modification: The mac-address max-mac-count enable-forwarding command is added in Layer 2 aggregate interface view. A Layer 2 aggregate interface supports forwarding unknown frames after the MAC learning limit on the Layer 2 aggregate interface is reached.

[202208060328] Support for IP precedence marking actions in an outbound QoS policy

Before modification: The IP precedence marking action (remark ip-precedence) is not supported in a QoS policy applied to the outbound direction of an interface.

After modification: The IP precedence marking action (remark ip-precedence) is supported in a QoS policy applied to the outbound direction of an interface.

[202208081419] Support for applying a QoS policy to the outbound direction of a Layer 3 aggregate subinterface

Before modification: A QoS policy cannot be applied to the outbound direction of a Layer 3 aggregate subinterface.

After modification: A QoS policy can be applied to the outbound direction of a Layer 3 aggregate subinterface.

[202211241281] Support of NETCONF for the bgp as-path-relax ebgp/ibgp command

Before modification: In NETCONF mode, the bgp as-path-relax ebgp and bgp as-path-relax ibgp commands are not supported. You cannot use only EBGP or IBGP routes for BGP load sharing.

After modification: In NETCONF mode, the bgp as-path-relax ebgp and bgp as-path-relax ibgp commands are supported. You use only EBGP or IBGP routes for BGP load sharing.

[202302031073] Traffic statistics collection for Layer 3 Ethernet subinterfaces in inbound and outbound directions

Before modification: Layer 3 Ethernet subinterfaces do not support inbound or outbound traffic statistics collection. The counts in the input and output traffic statistics are 0 in the output from the display interface command.

After modification: Layer 3 Ethernet subinterfaces support inbound and outbound traffic statistics collection. The display interface command can display traffic statistics in both the inbound and outbound directions.

[202209240884] Support for the rewrite inbound tag remark 1-to-2 command

Before modification: The rewrite inbound tag remark 1-to-2 command is not supported in Ethernet service instance view.

After modification: The rewrite inbound tag remark 1-to-2 command is supported in Ethernet service instance view.

[202212030374] Support for enabling IPv6 PIM snooping in VSI view

Before modification: The ipv6 pim-snooping enable command is not supported in VSI view.

After modification: The ipv6 pim-snooping enable command is supported in VSI view.

[202212050036] Aging mechanism for ND flood suppression entries

Before modification: Before an ND flood suppression entry ages out, the device will send an NS message. If the IPv6 address of the involved VSI interface is not on the same subnet as the IPv6 address in the ND flood suppression entry, or if the VSI is not associated with a VSI gateway interface, the source IPv6 address in the NS message will be an all-zero IPv6 address. If the receiving end is a third-party device, it might mistakenly identify the NS message as an address conflict.

After modification: If the IPv6 address of a VSI interface is not on the same subnet as the IPv6 address in an ND flood suppression entry, or if a VSI is not associated with a VSI gateway interface, the device does not send an NS message before the entry ages out. The entry will be learned again from subsequent ND messages.

[202302070640] Support for NETCONF deployment of stp transparent enable/lldp transparent enable

Before modification: The stp transparent enable and lldp transparent enable commands do not support NETCONF deployment .

After modification: The stp transparent enable and lldp transparent enable commands support NETCONF deployment.

[202209160614] Support of DCI Layer 3 multicast for the SSM group address

Before modification: DCI Layer 3 multicast does not support SSM group address 232.0.0.0/8.

After modification: DCI Layer 3 multicast supports SSM group address 232.0.0.0/8.

[202301121136] Password recovery capability is enabled by default in the initial configuration instead of the factory defaults on the device.

Before modification: Password recovery capability is enabled by default in the factory defaults on the device. You can use display default-configuration, display current-configuration, and display current-configuration all to view the configuration of password recovery capability.

After modification: Password recovery capability is enabled by default in the initial configuration on the device. You can use only display current-configuration all to view the configuration of password recovery capability.

[202208111132] Changes to supported commands in FIPS mode

Before modification:

·         In FIPS mode, the following commands are supported:

¡  ntp-service multicast-client

¡  ntp-service multicast-server

¡  ntp-service ipv6 multicast-client

¡  ntp-service ipv6 multicast-server

¡  ntp-service broadcast-client

¡  ntp-service broadcast-server

·         In FIPS mode, the MD5 algorithm can be specified in the ntp-service authentication-keyid and sntp authentication-keyid commands for authentication.

·         In FIPS mode, the value range for the version number parameter in the ntp-service unicast-peer, ntp-service unicast-server, and sntp unicast-server commands is 1 to 4.

After modification:

·         The following commands are not supported in FIPS mode:

¡  ntp-service multicast-client

¡  ntp-service multicast-server

¡  ntp-service ipv6 multicast-client

¡  ntp-service ipv6 multicast-server

¡  ntp-service broadcast-client

¡  ntp-service broadcast-server

·         In FIPS mode, no MD5 algorithm can be specified in the ntp-service authentication-keyid and sntp authentication-keyid commands for authentication.

·         In FIPS mode, the value for the version number parameter in the ntp-service unicast-peer, ntp-service unicast-server, and sntp unicast-server commands is 3 or 4.

Operation changes in R6710

[202211111499]Added information for troubleshooting unexpected reboots due to Intel C3XXX CPU microcode

The device records Machine Check Error  (MCE) register information for troubleshooting when it reboots due to Intel C3XXX CPU microcode.

[202209280283] The S9850-32H switch added support for automatic saving of power supply monitoring/power-on sequence controller fault information

Before modification: Fault information of the power supply monitoring/power-on sequence controller cannot be obtained.

After modification: The following probe commands are available for obtaining fault information of the power supply monitoring/time sequence controller. The device will save fault information of the power supply monitoring/ power-on sequence controller to the flash once every day automatically to facilitate fault locating.

The following probe commands (executed in probe view) were added:

·         debug system blackbox-info slot slot-number current: Reads fault information of the power supply monitoring/power-on sequence controller.

·         debug system blackbox-info slot slot-number flash: Reads fault information of the power supply monitoring/power-on sequence controller stored in flash.

[202212071604] Added support for NETCONF/gRPC collection of the system power consumption in real time

Before modification: The chassis and boards data collected through NETCONF/gRPC does not include the total power, residual power, and nominal power data.

After modification: The chassis and boards data collected through NETCONF/gRPC includes the total power, residual power, and nominal power data.

[202207130194]Changed the BGP VPNv4 route advertisement behavior when the peer next-hop-vpn and advertise l3vpn route are executed

Before modification: BGP VPNv4 routes are not advertised through the BGP EVPN address family even if the peer next-hop-vpn and advertise l3vpn route are executed.

After modification: BGP VPNv4 routes are advertised through the BGP EVPN address family if the peer next-hop-vpn and advertise l3vpn route are executed.

 [202209211687] Added support for secure BootWare

R6710 added support for secure BootWare, which prevents attackers from tampering with BootWare or planting malicious software to control/damage device software. When installing R6710, the device will automatically upgrade Bootware to a version that supports secure Bootware. A message "PKT_CheckFileHead failedFailed" (indicating an illegitimate upgrade file) will be displayed during device startup. This message is displayed because a related set bit flag has not been used before and has an initial value. This message will not be generated when versions that support secure Bootware are upgraded later.

[202209051705] The displayed IPv6 peers in the display bgp peer command output are too long and so they are displayed in multiple lines

Before modification: A peer IP address exceeding 15 characters will be displayed in multiple lines.

After modification: A peer IP address is displayed in one line.

[202210171050]Restrictions for an interface to join a Layer 2 aggregation group

Before modification:

·         An interface can join a Layer 2 aggregation group only when the interface meets the following requirements:

¡  It uses default VLAN configuration.

¡  Its port isolation configuration is the same as that on the aggregate interface.

·         After leaving a Layer 2 aggregation group, an interface restores its VLAN configuration to the default.

After modification:

·         An interface cannot join a Layer 2 aggregation group if it has different attribute configurations from the aggregate interface.

·         After leaving a Layer 2 aggregation group, an interface retains the VLAN configuration inheriting from the aggregate interface.

[202406121103] Changed the conditions for the L3VNI VSI interface packet statistics feature to take effect

Before modification: The VSI interfaces of L3VNIs support interface statistics by default. You can use the display l2vpn vsi verbose and display interface vsi commands to obtain the statistics.

After modification: The VSI interfaces of L3VNIs do not support interface statistics by default. To enable the statistics feature, execute the l2vpn statistics vsi l3-vni command.

Operation changes in E6706

[202207121259] Changed the maximum MTU allowed for an interface to 9216 bytes 

Before modification: The maximum MTU allowed for an interface is 9008 bytes.

After modification: The maximum MTU allowed for an interface is 9216 bytes.

[202208151044]Changed the traffic forwarding behavior for link aggregation management VLANs

Before modification: The unicast packets, ARP packets, or DHCP packets are load shared across the Selected ports of the aggregation group even if link aggregation management VLANs are configured.

After modification: The unicast packets, ARP packets, or DHCP packets are forwarded through only the management port if link aggregation management VLANs are configured.

Operation changes in E6705

[202112171336] Modified feature: Displaying kernel information upon power cycling of a device

Before modification: After you execute the reboot command to reboot the device, the device can display kernel-related commands. After you power cycle the device, the device cannot display kernel-related commands.

After modification: After you power cycle the device, the device cannot display the following kernel-related commands:

·         display kernel deadloop: Displays kernel thread deadloop information.

·         display kernel exception: Displays kernel thread exception information.

·         display kernel reboot: Displays reboot information for the device.

·         display kernel starvation: Displays kernel thread starvation information.

[202201190744/202204191564] Modified feature: Optimizing downlink interface state switchover time for Monitor Link

Before modification: In a monitor link group, when the uplink interfaces go down, the state switchover for all downlink interfaces takes a relatively long time.

After modification: In a monitor link group, when the uplink interfaces go down, the state switchover for all downlink interfaces takes less than 5 seconds.

[202203071738] Restrictions for an interface to join a Layer 2 aggregation group

Before modification:

·         An interface cannot join a Layer 2 aggregation group if it has different attribute configurations from the aggregate interface.

·         After leaving a Layer 2 aggregation group, an interface retains the VLAN configuration inheriting from the aggregate interface.

After modification:

·         An interface can join a Layer 2 aggregation group only when the interface meets the following requirements:

¡ It uses default VLAN configuration.

¡ Its port isolation configuration is the same as that on the aggregate interface.

·         After leaving a Layer 2 aggregation group, an interface restores its VLAN configuration to the default.

[202205301755]Modified the value range for the default CPU MAC address of a device

Before modification: The default CPU MAC address of a device is 00E0-FC00-XXXX.

After modification: The default CPU MAC address of a device is a MAC address in the range of F010-90DB-7400 to F010-90DB-740F.

[202205301735] Adjusted the limit on the MAC addresses assigned to VSI interfaces

Before modification: A maximum of 1000 MAC addresses different from the default MAC address can be assigned to VSI interfaces. Two VSI interfaces are considered to use two MAC addresses even through you assign the same MAC address to them.

After modification: A maximum of 1000 MAC addresses different from the default MAC address can be assigned to VSI interfaces. Multiple VSI interfaces are considered to use one MAC address if you assign the same MAC address to them. When Layer 3 traffic is forwarded to a VXLAN network, the traffic matches MAC addresses of VSI interfaces instead of VSIs. Therefore, Layer 3 traffic might be sent to an incorrect VSI.

Operation changes in E6702

[202202280302]Sent SNMP traps when uncorrectable parity errors occur on memory management entries

Before modification: Only logs are printed when uncorrectable parity errors occur on memory management entries.

After modification: Logs are printed and SNMP traps are sent when uncorrectable parity errors occur on memory management entries.

NOTE: The node ID for SNMP traps is 1.3.6.1.4.1.25506.8.35.12.3.0.44 (hh3cMMUParityError). SNMPv2 and later versions support this node.

Operation changes in F6701L01

First release.

Restrictions and cautions

Restrictions

1.       When you use 40-GE QSFP+ cables to connect IRF physical interfaces, follow these rules:

·         Make sure ports at both ends of a link are fixed ports or ports on interface modules.

·         If you use ports on interface modules as IRF physical interfaces, make sure the ports at both ends of a link reside on interface modules of the same model.

·         If you use a fixed QSFP28 port on the S6850-2C rear panel as an IRF physical interface, make sure the port at the other end is also a fixed QSFP28 port on the S6850-2C rear panel.

2.       ISSU is not supported on an IRF fabric that uses 40-GE QSFP+ cables for IRF link connections. To perform ISSU on the IRF fabric, replace the cables with 100-GE or 40-GE transceiver modules or cables.

3.       When you enter interface range view by using the interface range command and then bulk configure interfaces, the device does not output prompt or alarm messages. Make sure you are fully aware of the impacts of the configuration.

4.       If an IRF member device receives packets from another IRF member device and then forwards them out of a VSI interface, the packet filtering or QoS policy configuration applied in the outbound direction of the VSI interface cannot match these packets.

5.       If the QoS and ACL resource hardware mode is set to outbound-vlan-enhanced, do not apply a QoS policy that matches SVLANs to the outbound direction or apply a QoS policy to a VLAN.

6.       When the clock node type of the device is configured as E2ETC or P2PTC, up to 10 synchronization devices can be attached to the device.

7.       When an SFP28 or SFP+ interface uses an SFP transceiver module to connect to a peer interface, you must disable autonegotiation on the peer interface.

8.       Executing the following commands will cause traffic interruption on interfaces BFD flapping, and LLDP flapping:

¡  buffer apply

¡  buffer egress cell queue shared

¡  qos wred apply

¡  qos wrr weight

¡  qos wrr group weight

¡  priority-flow-control no-drop dot1p

9.       After the switch is upgraded from a DRNI version (version earlier than E6705) to an M-LAG version (E6705 or later), all drni keywords are automatically replace with m-lag keywords. The switch cannot automatically replace keywords after an M-LAG version is rolled back to a DRNI version. You must prepare a configuration file with drni keywords, specify it as the startup configuration file, and reboot the switch.

An M-LAG version cannot be rolled back to a DRNI version by using the configuration replace file command. To roll back, you must prepare a configuration file with drni keywords, specify it as the startup configuration file, and reboot the switch.

The name and terms were changed for the DRNI feature as follows:

¡  The feature name was changed to M-LAG.

¡  The drni keyword was changed to m-lag or mlag.

¡  The intra-portal-port keyword was changed to peer-link.

¡  The ipp keyword was changed to peer-link.

¡  The dr keyword was changed to m-lag-interface.

For more information, see H3C S9850_6850-CMW710-E6705 Release Notes (Software Feature Changes).

10.     In version E67xx/F67xx, the user-defined ACL rules configured by using the rule command do not support matching GRE packets with the specified flags. That is, the udf-format argument does not support the following values:

¡  gre-c0r0k1: Matches GRE packets with c=0, r=0, and k=1.

¡  gre-c1r0k1: Matches GRE packets with c=1, r=0, and k=1.

¡  gre-k1r1: Matches GRE packets with k=1 and r=1.

11.     As from version F6612, when an aggregate interface is configured as an IPP (peer link interface), the following configuration in grey is automatically deployed. During the software upgrade process, if the DR system member devices (M-LAG member devices) on both ends have different software versions, the IPPs (peer-link interfaces) on both ends will have different configurations. As a result, the IPPs (peer-link interfaces) cannot come up, and the DR system (M-LAG system) cannot operate normally.  Before you upgrade a DR system running a version earlier than F6612 to F6612 or later, you must deploy the following configuration in grey or other configuration as needed to the IPPs (peer-link interfaces) to ensure that the configuration is the same on IPPs (peer-link interfaces) on both ends. In this way, you can avoid DR system exceptions during the upgrade process.

    [H3C-Bridge-Aggregation11]display this

    #

    Interface Bridge-Aggregation11

    port link-type trunk

    port trunk permit vlan all

    port drni intra-portal-port 1

    #

    return

    [H3C-Bridge-Aggregation11]

12.     A 100G port on the device might fail to come up when a 100G DAC cable is used to connect the port to an Intel network card, for example, Intel (rainbow) Eth E810-CQDA2. As a best practice, use an AOC cable or transceiver module for connection to the Intel network card.

13.     When a software version earlier than R6710 is upgraded to R6710 through an ISSU, VXLAN Layer 3 traffic loss persists for longer than 2 seconds. In a lab environment, traffic loss persists for 21 seconds during a one-click upgrade and 35 seconds during a distributed upgrade.

14.     Mirroring outgoing VXLAN packets of a port is supported only when the first mirroring group is configured to mirror outgoing packets of that port. In other cases, if the first mirroring group is configured to mirror both incoming and outgoing packets of a port, or any non-first mirroring group is configured to mirror both incoming and outgoing packets or outgoing packets of a port, mirroring outgoing VXLAN packets of that port is not supported.

15.     To issue PFC commands to an interface range, make sure all interfaces in the interface range are of the same type.

Cautions

1.       The following interfaces cannot be split:

¡  Interface numbered 31 on the S6850-56HF.

¡  All interfaces on the LSWM116Q interface module.

¡  Interface numbered 7 on the LSWM18CQ or LSWM18CQMSEC expansion card installed in expansion slot 2 of the S985-4C.

¡  Interface numbered 8 on the LSWM18QC expansion card installed in expansion slot 2 of the S9850-4C switch.

¡  Interface numbered 25 on the LSWM124XGT2Q, LSWM124TG2H, LSWM124XG2QFC, LSWM124XG2Q, or LSWM124XG2QL expansion card installed in expansion slot 2 of the S9850-4C switch.

¡  The interface numbered 31 on an S9850-32H switch cannot be split by default. To split this interface, first enable the hardware resource flex mode for the device by using the hardware-resource flex-mode enable command. .

2.       The S6850-56HF (product code LS-6850-56HF-H3) and S6850-56HF-CP switches support the SW-A-PSR550-12A-B and SW-B-PSR550-12A-B power supplies, and the two types of power supplies cannot be mixed up.

Licensing

About licensing

H3C offers licensing options for you to deploy features and expand resource capacity on an as needed basis. To use license-based features, purchase licenses from H3C and install the licenses. For more information about the license-based features and licenses available for them, see H3C Switches License Matrixes.

Registering and installing licenses

To register and transfer licenses, access H3C license services at http://www.h3c.com/en/License.

For information about registering licenses, installing activation files, and transferring licenses, see H3C Switches and Routers Licensing Guide.

Obtaining license server software and documentation

To perform remote licensing, first download and install the H3C license server software.

·         To obtain the H3C license server software package, click

H3C license server software package

·         To obtain H3C license server documentation, click

H3C license server documentation

Open problems and workarounds

202204210001

·         Symptom: Two EDs form a DR system. After the uplink or downlink on one ED fails, its traffic fails over to the other ED through the IPL. As a result, the other ED reflects the traffic back to the sender data center site and forwards multiple replicas of the traffic.

·         Condition: This symptom might occur if an uplink or downlink fails and traffic fails over between the EDs through the IPL.

·         Workaround: None.

202205091692

·         Symptom: The device erroneously deletes outgoing Layer 3 Ethernet interfaces from the multicast forwarding entries of a multicast group after an interface leaves the multicast group and joins another one.

·         Condition: This symptom might occur if the following events occur:

a.    Fast-leave processing is enabled on Layer 2 and Layer 3 Ethernet interfaces.

b.    Multiple interfaces join multicast group 1.

c.    An interface leaves multicast group 1 and then joins multicast group 2, and the report message for joining multicast group 2 and the leave message for leaving multicast group 1 are encapsulated into one packet.

·         Workaround: Do not execute the igmp fast-leave command on Layer 3 Ethernet interfaces.

202208181547

·         Symptom: The same gRPC packet type has different packet formats in different software versions (for example, R6635 and E6706).

·         Condition: This symptom occurs if two switches loaded with different software version are configured to send gRPC packets of the same type to a gRPC server.

·         Workaround: Configure the gRPC server to accept different packet formats of the same gRPC packet type.

202209212126

·         Symptom: The PVST state is abnormal. As a result, traffic cannot be forwarded.

·         Condition: This symptom occurs if Device A and Device B are interconnected and run PVST, the private VLAN feature is configured on Device A, and the port private-vlan trunk secondary command is executed on the interface connecting Device A to Device B.

·         Workaround: Do not execute the port private-vlan trunk secondary command on a PVST network.

202303250552

·         Symptom: On an M-LAG system, a small number of packets are loss during a GIR upgrade.

·         Condition: This symptom occurs if traffic is present on the AC on the M-LAG interface.

·         Workaround: None.

202305260415

·         Symptom: After a device is rebooted, the subcard configuration is lost.

·         Condition: This symptom occurs if an IRF member device with subcards is rebooted, the subcards are unplugged during the reboot process, and the subcards are plugged again after the device is rebooted.

·         Workaround: Do not unplug subcards during the reboot process.

202403111706

·         Symptom: Ports on the LSWM116FC interface module cannot be connected.

·         Condition: This symptom occurs if you switch the interface type from Ethernet to FC for ports on the LSWM116FC interface module, and then switch the interface type back to Ethernet.

·         Workaround: None.

202401040403

·         Symptom: If you execute the shutdown command on the VLAN interface, hardware BFD cannot detect that the interface state has changed to down.

·         Condition: This symptom occurs if you use an IP address to configure a hardware BFD session on a VLAN interface, and then execute the shutdown command on the VLAN interface to shut it down.

·         Workaround: Use software BFD to replace hardware BFD for detecting state changes of the VLAN interface.

202312190713

·         Symptom: Outbound flow mirroring does not take effect.

·         Condition: This symptom occurs if you configure outbound flow mirroring to mirror traffic to another member device after forming an IRF fabric by using LS-6850-56HF-H1 and LS-6850-56HF-H3 switches.

·         Workaround: Do not use the LS-6850-56HF-H1 and LS-6850-56HF-H3 switches to form an IRF fabric, or configure the flow mirroring destination interfaces on the same device.

202402030162

·         Symptom: The dhcp snooping trust command cannot take effect on a VSI.

·         Condition: This symptom occurs if you perform the following operations:

a.    Configure the dhcp snooping trust command in a VSI and on an AC port.

b.    Undo the dhcp snooping trust command on the AC port. This operation mistakenly removes the related flag from the VSI, causing the symptom.

·         Workaround: Undo the dhcp snooping trust command in the VSI, and then reconfigure the command.

202402041660

·         Symptom: On the 6850 or 9850 device, even when loose uRPF check is enabled and a default route is configured with a reachable next hop, traffic forwarding may still encounter issues.

·         Condition: This symptom occurs if you enable loose uRPF check on the 6850/9850 device and manually set the next hop of the default route to the CPU.

·         Workaround: Ensure that the manually configured default route's next hop is valid and does not point to the CPU or a blackhole address.

202401290949

·         Symptom: Duplicate traffic exists in the network.

·         Condition: This symptom occurs if FRR is enabled and it affects broadcast traffic in an ES network.

·         Workaround: Use unicast forwarding.

202310301520

·         Symptom: BFD flapping occurs with a certain probability.

·         Condition: This symptom occurs in an M-LAG network where devices that support hardware BFD are configured. If you shut down the physical interface directly connected to a downlink device to have the reachable path traverse the IPL, BFD flapping occurs.

·         Workaround: None.

202402200464

·         Symptom: The device experiences an abnormal reboot.

·         Condition: This symptom occurs when multiple interfaces on the device are configured with sFlow sampling and the interfaces receive a large amount of traffic.

·         Workaround: In interface view, configure a higher sFlow packet sampling rate to reduce the number of packets sampled by sFlow.

202403070954

·         Symptom: When both arp snooping and arp suppression are configured on a VSI, the local ARP flood suppression entries migrate from the M-LAG interface to the peer-link interface, and static ARP snooping entries are generated. As a result, ARP entries do not have outgoing interfaces. The same situation can also occur with the ND protocol.

·         Condition: This symptom occurs if arp snooping and arp suppression are configured, and endpoint migrate from the M-LAG interface to the single-homed mode on an EVPN+M-LAG network.

·         Workaround: Disable ARP snooping, or prevent endpoints from switching to the single-homed mode.

202403090085

·         Symptom: Traffic is interrupted on the M-LAG interface due to the dispute guard feature.

·         Condition: This symptom occurs if the following conditions exist:

¡  In the network configured with both the spanning tree protocol and M-LAG, upgrade the M-LAG system by using the traditional method: first upgrade the secondary member device, and then disable the downlink interfaces on the primary member device.

¡  The rlink reliability settings are different between versions before and after upgrade.

·         Workaround: Before upgrade, execute the undo stp dispute-protection command to disable the dispute guard feature on the M-LAG device.

202401051728

·         Symptom: When an FTP client copies a file from the device, the following RAS alarm is generated: EEPROM read/write failed.

·         Condition: This symptom occurs if the device acts as an FTP server and the connection is established for the first time.

·         Workaround: None.

202403181643

·         Symptom: Failed to log in to the device through Telnet.

·         Condition: This symptom might occur if you enter ASCII code 19 in decimal notation (default stop character for PTY) in the password for Telnet login, which causes login timeout and failure to close the Telnet connection.

·         Workaround: Do not enter ASCII code 19 in decimal notation when you enter a password for login.

202404130067

·         Symptom: If ECMP contains 128 next hops, and you shut down one or multiple interfaces, a number of 2 to 10 packets might be lost.

·         Condition: This symptom occurs if ECMP contains 128 next hops, and you shut down one or multiple interfaces.

·         Workaround: If ECMP contains too many next hops, do not shut down interfaces.

202405060191

·         Symptom: The Ethernet link aggregation mode LBN group feature does not take effect when the following conditions are met:

¡  Both the ECMP-mode LBN group feature and the Ethernet link aggregation mode LBN group feature are configured.

¡  LBN group member interface delete actions are performed (such as Layer 2/3 mode switching, port split and combination, and removal of LBN group member interfaces).

·         Condition: This symptom might occur when the following conditions are met:

¡  Both the ECMP-mode LBN group feature and the Ethernet link aggregation mode LBN group feature are configured.

¡  LBN group member interface delete actions are performed (such as Layer 2/3 mode switching, port split and combination, and removal of LBN group member interfaces).

·         Workaround: Do not configure both the ECMP-mode LBN group feature and the Ethernet link aggregation mode LBN group feature or perform LBN group member interface delete actions such as Layer 2/3 mode switching, port split and combination, and removal of LBN group member interfaces.

202406200682

·         Symptom: The switch reboots due to memory exhaustion after the buffer transient-capture enable command is executed on an interface.

·         Condition: This symptom occurs if you specify an ACL in the command and the captured packets are not TCP or UDP packets.

·         Workaround: Do not specify an ACL in the buffer transient-capture enable command if the captured packets are not TCP or UDP packets.

202406261321

·         Symptom: In a multicast VXLAN, the switch fails to forward multicast traffic.

·         Condition: This symptom occurs if the following operations are performed:

a.    Execute the reset multicast forwarding-table command or stop traffic forwarding after traffic is switched to the data group.

b.    Forward traffic again when multicast forwarding entries are cleared.

·         Workaround: Do not configure a data group.

202406190225

·         Symptom: In a multicast VXLAN, the switch fails to forward traffic from the source to a receiver.

·         Condition: This symptom occurs if the following conditions exist:

¡  An L3 VXLAN ID is configured for the public instance.

¡  Multicast VXLAN is not configured for the public instance.

·    Workaround: Configure Multicast VXLAN for the public instance.

List of resolved problems

Resolved problems in R6715P01

202401161819

·         Symptom: A device fails to roll back BGP configuration.

·         Condition: This symptom occurs if the AS number of a BGP peer changes after BGP configuration rollback.

202302141258

·         Symptom: On an EVPN M-LAG system, learned dynamic ND entries never age.

·         Condition: This symptom occurs after an M-LAG member device is rebooted.

202312271152

·         Symptom: The BFD session flaps once if the statistics l3-packet enable inbound command is configured on a network enabled with BFD.

·         Condition: This symptom might occur if the statistics l3-packet enable inbound command is configured on a network enabled with BFD.

202401050745

·         Symptom: The M-LAG table reports an error.

·         Condition: This symptom occurs if you read MIB node information.

202401151855

·         Symptom: After you configure an inband management VLAN by using the in-band management vlan command, packets on the corresponding VLAN interface should only pass through hardware forwarding. However, ARP packets are actually copied and sent to the controller during hardware forwarding.

·         Condition: This symptom occurs if two OpenFlow instances on the device both match packets of the specified VLAN interface, and the corresponding VLAN is configured as the inband management VLAN for one instance.

202402020349

·         Symptom: Broadcast packets passing through a real tunnel are not dropped.

·         Condition: This symptom occurs if broadcast packets pass through a real tunnel in an M-LAG VXLAN network.

202312011980

·         Symptom: The device restarts abnormally.

·         Condition: This symptom might occur if a large amount of route oscillation occurs.

202404021174

·         Symptom: No output is displayed after you execute the display transceiver power command.

·         Condition: This symptom might occur if you execute the display transceiver power command to power information for transceiver modules.

202403080772

·         Symptom: The device restarts abnormally.

·         Condition: This symptom might occur in the following conditions:

¡  The device uses gRPC to subscribe to RoCEv2 traffic events and reports flow tables containing generated RoCEv2 traffic via gRPC.

¡  The gRPC process restarts abnormally.

202403120391

·         Symptom: The if-match any command in a class-behavior association cannot match IPv4 packets in a QoS policy applied to the outbound direction.

·         Condition: This symptom occurs if the following operations are performed:

a.    Configure a QoS policy with three class-behavior associations. The first class-behavior association is used to match any packets (if-match any), the second class-behavior association is used to match IPv4 packets, and the third class-behavior association is used to match IPv6 packets.

b.    Enable statistics collection for outgoing Layer 3 packets on an interface and apply the QoS policy to the outbound direction of the interface.

c.    Delete the second class-behavior association.

202402060902

·         Symptom: Failed to deploy a static ARP entry that contains a VSI. However, when the device starts with the configuration, the configuration succeeds but does not take effect.

·         Condition: This symptom occurs if you configure a static ARP entry that contains a VSI not bound to a tunnel interface.

202402260716

·         Symptom: After you execute the arp route-direct advertise tag tag-value command on an interface and ARP advertises ARP entries to generate direct routes, ARP does not update route management information if you edit the route tag or preference for the ARP-advertised direct routes.

·         Condition: This symptom occurs if you edit the route tag or preference for the ARP-advertised direct routes after you configure ARP direct route advertisement on an interface.

202401111026

·         Symptom: If you repeatedly use the evpn edge group command to configure and delete the virtual ED address, tunnel address information might be incorrect.

·         Condition: This symptom might occur if the following conditions exist:

¡  In an EVPN VXLAN network, BGP is configured with multiple RR-oriented peers. The source addresses used for peer session establishment are not specified on those peers.

¡  The evpn edge group command is repeatedly use to configure and delete the virtual ED address.

202401311793

·         Symptom: The sysmand process abnormally quits and the device generates core files.

·         Condition: This symptom might occur is a corrupted .bin file is uploaded to the device.

202401230998

·         Symptom: MAC learning does not take effect on an interface.

·         Condition: This symptom might occur if you disable MAC learning globally, perform port splitting or merging, and then enable MAC learning globally.

202402211655

·         Symptom: Ports on the device do not recognize Alibaba or ACT 100G ZR4 modules.

·         Condition: This symptom occurs if you install an Alibaba or ACT 100G ZR4 module on a port of the device.

202403130948

·         Symptom: After running for a period of time, the PTP loses clock and fails to synchronize the time, with an offset exceeding 400,000 ns per second if the frequency offset remains below 10 ns after PTP convergence and clock source lock.

·         Condition: This symptom might occur if the frequency offset remains below 10 ns after PTP convergence and clock source lock.

·         Remarks: After the clock converges to a narrow range of accuracy, frequency offset adjustments fail to take effect. This causes abnormal accumulations of frequency values, which continuously build up and eventually trigger a frequency reset callback. This causes the clock to become unlocked and unable to recover.

202402201209

·         Symptom: The multicast packets with a group address in the 224.0.0.0 or 224.0.1.0 network are dropped in a VLAN.

·         Condition: This symptom occurs if dropped unknown multicast data packets is enabled for the VLAN.

202404130541

·         Symptom: An interface is always up and cannot switch between Layer 2 mode and Layer 3 mode.

·         Condition: This symptom might occur if a 10G fiber interface is inserted with a fiber-to-copper transceiver module and the loopback-test internal command is executed on that interface.

202403211980

·         Symptom: An interface failed to come up on the device.

·         Condition: This symptom might occur if the device is inserted with a third-party 100G LR4 long-reach transceiver module.

·         Remarks: The impacted modules include LR4, ER4L, DWDM2, ZR4, DR1, LR1, and FR1.

202208050810

·         Symptom: The protocol queue on the device is congested and has packet loss, causing frequent IS-IS flapping

·         Condition: This symptom occurs if the following conditions exist:

a.    The controller has issued an ACL to filter DHCP messages sent to the controller.

b.    A large number of packets with unreachable destination IPs are received from the VXLAN tunnel interface.

202404291067

·         Symptom: ND/IGMP protocol packets are forwarded abnormally on blocked ports.

·         Condition: This symptom occurs if the following conditions exist on the overlay network:

a.    The spanning tree protocol is enabled globally and an AC interface is created on the device.

b.    An interface or aggregate interface is blocked by STP.

202404090532

·         Symptom: The ND outgoing interface learned by the gateways on both M-LAG member devices is the peer-link interface. As a result, a traffic loop is generated and traffic fails to be forwarded.

·         Condition: This symptom occurs if the following operations are performed in a VLAN M-LAG network:

a.    The gateways on both M-LAG member devices have local ND proxy enabled.

b.    An M-LAG member device initiates ND detection to an unknown IPv6 host.

202404291194

·         Symptom: Failed to bind a VSI to a VXLAN ID, but no log indicating a failure caused by hash conflict is printed.

·         Condition: This symptom might occur when you configure a large number of ACs or VSIs.

202404100492

·         Symptom: When you configure the ip load-sharing mode per-flow command, you can specify the algorithm, tunnel, and five-tuple parameters together. However, if you delete the algorithm, tunnel, or five-tuple configuration, the other per-flow configurations will also be deleted. For example, after you execute the undo ip load-sharing mode per-flow algorithm command, the other per-flow configurations will be deleted, including the tunnel and five-tuple configurations.

·         Condition: This symptom might occur when you delete the configuration of any parameter in the ip load-sharing mode per-flow command.

202404220719

·         Symptom: Traffic forwarding based on ECMP routes fail.

·         Condition: This symptom might occur if IPv6 routes with prefixes longer than 64 bits form ECMP routes when the hardware resource mode is IPv6-64.

202405061402

·         Symptom: LBN configuration fails to take effect after upgrading the primary and backup in an IRF environment.

·         Condition: This symptom might occur if the interfaces acting as group members are Layer 3 interfaces.

·         Remarks: If this issue occurs after the device is upgraded from R6715 or an earlier version to R6715P01, you can delete the LBN configuration, and then re-configure LBN configuration.

202401041128

·         Symptom: After the external egress recovers from a fault, multicast traffic is interrupted for a long period of time when the primary border device resumes carrying multicast traffic.

·         Condition: This symptom occurs if the following operations are performed:

¡  Multicast traffic from an external source is forwarded to leaf devices through the border devices of two DCs in a cross-DC EVPN network. In normal conditions, multicast traffic is forwarded through the primary border device.

¡  When the external egress on the primary border device experiences a fault, multicast traffic is forwarded to the leaf devices through the secondary border device and the ED DCI path.

202406121282

·         Symptom: After you configure flow sampling on an interface to sample the inbound packets and configure the sampling rate, the outbound packets of the interface are also sampled.

·         Condition: This symptom occurs if you configure sFlow to sample the inbound packets of an interface.

202406121195

·         Symptom: The received VLAN-tagged packets do not carry VLAN tags after they are sampled by sFlow.

·         Condition: This symptom occurs if you configure sFlow sampling on a main Layer 3 Ethernet interface and configure IP addresses for its Layer 3 Ethernet subinterfaces.

202405060248

·         Symptom: In an M-LAG network, when the peer-link flaps, messages about configuration inconsistency are reported, but the display m-lag consistency command shows no relevant information.

·         Condition: This symptom might occur when the M-LAG interface does not exist or is not associated with a VSI and the peer-link interface flaps.

202405292207

·         Symptom: In an EVPN+IRF network, two leaf nodes use the same MAC address. As a result, the MAC address entries of the other leaf nodes point to incorrect destinations.

·         Condition: This symptom occurs if the MAC address synchronized from the EVPN remote device is overridden when the subordinate IRF member device synchronizes its MAC address to the master IRF member device. As a result, the deletion and addition of the local MAC address might not be timely reported.

202405131762

·         Symptom: Peer device Leaf2 did not correctly record the proxy status of Leaf1, leading to the erroneous forwarding of IGMP general query packets to Leaf1, resulting in superfluous packets.

·         Condition: This symptom might occur when you first set up an EVPN network, and then execute the igmp-snooping proxy enable command on Leaf1.

202405060900

·         Symptom: The information about dual-blade fans is displayed incorrectly.

·         Condition: This symptom might occur if you execute the display key-component command to view speed information about other dual-blade fans after you remove a dual-blade fan.

202406111529

·         Symptom: After you execute the restore factory-default command to restore a device to factory defaults, the system generates the PKT_CheckFileHead failedFailed message during the device restart.

·         Condition: This symptom might occur if you execute the restore factory-default command to restore a device to factory defaults, and then restart that device.

202403270738

·         Symptom: A CRC error packet occurs on an interface.

·         Condition: This symptom might occur if MACsec is enabled on an interface on the LSWM18CQMSEC interface card and the interface forwards traffic at a speed that exceeds the limit.

202401250136

·         Symptom: The device might fail to withdraw routes on a peer.

·         Condition: This symptom occurs if the following conditions exist:

a.    First-AS-number check is enabled for EBGP routes on the device by the peer-as-check enable command, and the device has multiple peers.

b.    The device updates and withdraws BGP routes simultaneously.

Resolved problems in R6715

202306160638

·         Symptom: VRRP cannot learn unicast ARP entries with the destination MAC addresses as VRRP virtual MAC addresses.

·         Condition: This symptom occurs on an M-LAG+VRRP network.

202306070923

·         Symptom: After you execute the undo ospf command on a device in a certain large-scale network, the kernel of the device gets stuck, and the device reboots unexpectedly.

·         Condition: This symptom occurs if the following operations are performed on a certain large-scale network with a large number of devices interconnected by using IGP:

a.    Establish BGP and IBGP neighbors among these devices through BGP RRs.

b.    Enable the label capability and labeled routes.

c.    Execute the undo ospf command.

202212221227

·         Symptom: A security risk exists on the switch.

·         Condition: This symptom occurs if a self-signed certificate is saved in the PKCS#12 certificate file format without an encryption password and can be parsed by a third-party tool.

202305291896

·         Symptom: During a step-by-step ISSU on an IRF fabric, the convergence time of some traffic is 2 seconds.

·         Condition: This symptom occurs if the configuration file of a member device include EVPN and VXLAN settings.

202310161033

·         Symptom: The device reboots unexpectedly.

·         Condition: This symptom might occur if the vxlan default-decapsulation source command is executed after the MPLS hardware resource entry limit is exceeded.

202309280817

·         Symptom: Aggregation management interfaces are configured. After the device reboots with saved running configuration, the aggregation management interfaces do not take effect.

·         Condition: This symptom might occur if aggregation management interfaces are configured, the running configuration is saved, and the device is rebooted.

202312270219

·         Symptom: When the physical status of a port changes to Down, the logical status of that port remains Up at the underlying layer, and the port can send and receive packets normally.

·         Condition: This symptom might occur if STP is disabled and a firmware failure interrupts physical port status detection on the device.

202312150636

·         Symptom: The function to disable MAC address learning on new ports does not take effect.

·         Condition: This symptom might occur if the following conditions exist:

a.    MAC address learning is disabled globally.

b.    Port splitting or merging is performed.

202312221597

·         Symptom: The traffic on the main interface is dropped after a QoS policy with a deny action is applied to a Layer 3 aggregate subinterface.

·         Condition: This symptom occurs if you apply a QoS policy with a deny action to a Layer 3 aggregate subinterface.

202312062097

·         Symptom: Network congestion occurs when inter-VPN traffic is repeatedly forwarded by the device-internal loopback interface.

·         Condition: This symptom might occur if traffic matching static route configuration is forwarded through repeated routing table lookups in two VPN instances on the same device. Because the TTL of packets is not decreased by 1 when they pass through a loopback interface, traffic loop forwarding occurs, resulting in congestion.

202401080838

·         Symptom: The class-based accounting action in a QoS policy cannot count IPv6 traffic.

·         Condition: This symptom occurs if the if-match any clause is configured in the traffic class.

202401101931

·         Symptom: Next hop loss occurs after a refresh of the tunnel next hop.

·         Condition: The symptom occurs if you configure link-delay on the tunnel-side interface, which causes the interface to disconnect and then triggers a tunnel next hop refresh.

202402061134

·         Symptom: Authentication-related configuration fails to deploy.

·         Condition: This symptom occurs when you use NETCONF to configure authentication settings.

202402061677

·         Symptom: IPv4 traffic fails to be forwarded correctly.

·         Condition: This symptom occurs if the following conditions exist on an MVXLAN network:

¡  Configurations exist in both MVXLAN IPv4 address family view and MVXLAN IPv6 address family view.

¡  When IPv4 and IPv6 traffic is forwarded correctly, delete the configuration in MVXLAN IPv6 address family view.

202402060953

·         Symptom: The configured static ARP entry fails the check.

·         Condition: This symptom occurs if you specify a VSI and a tunnel interface when configuring a static ARP entry but no binding exists between them.

202309110615

·         Symptom: BGP flaps on a device.

·         Condition: This symptom occurs if the following conditions exist on the device:

a.    The device receives a large number of ipv4 BGP attack packets with TTL=1.

b.    The rate of delivering BGP packets to the CPU exceeds the upper hardware limit.

202310100263

·         Symptom: After the m-lag extra-vlan command is executed, the M-LAG member devices cannot synchronize ARP or ND entries for the extra VLANs through the peer-link interface.

·         Condition: This symptom might occur if some M-LAG interfaces are not assigned to the extra VLANs, and the device is rebooted or the peer-link interface flaps.

202307201560

·         Symptom: After the IS-IS route advertisement mode is changed from single-source to multi-source, route calculation errors occur, causing incorrect traffic forwarding.

·         Condition: This symptom might occur if the following conditions exist:

¡  FRR is enabled for IS-IS on the device.

¡  IS-IS routes have multiple advertisement sources, each with equal-cost output interfaces.

202211300042

·         Symptom: With NSR disabled, BGP performs active/standby switchover on the local device after GR capacity negotiation with a peer. However, the peer does not terminate the related peer session during active/standby switchover, because it is unaware of the abnormal TCP connection with the local device. After active/standby switchover, the local device cannot establish a BGP session to the peer unless the peer session to the local device times out on the peer side.

·         Condition: This symptom might occur if the following conditions exist:

¡  NSR is disabled, but BGP performs active/standby switchover on the local device after GR capacity negotiation with a peer.

¡  The previous peer retains the BGP peer session to the local device during active/standby switchover.

202309090022

·         Symptom: When NETCONF is used for device management and configuration, no information is displayed when you perform the operation of reading all device configurations.

·         Condition: This symptom occurs if only automatic tunnels exist and you execute the get-bulk operation to obtain the tunnel configuration.

Resolved problems in E6713P01

202312260590

·         Symptom: Obtaining the full device configuration through NETCONF timed out.

·         Condition: This symptom occurs after you obtain the full device configuration through NETCONF for more than 1000 times.

202311230241

·         Symptom: The following problems might occur with a low probability:

¡  The PVID of an aggregate interface is 1. However, VLAN 1 is not permitted on the aggregate interface, and the VLANs in STG 1 are not permitted either.

¡  The state of an aggregate interface is forwarding in the display stp brief command output, but the state of the corresponding aggregation member ports is block in STG 1 in the hardware.

¡  An AC created on an aggregate interface cannot forward traffic.

·         Condition: This symptom occurs if the PVID of the aggregate interface is in STG instance 1 and the aggregate interface does not permit any VLANs in STG instance 1.

202310230059

·         Symptom: For the C2000 device, the flash file system changes to read-only with a low probability, which might cause configuration file saving failures.

·         Condition: This symptom might occur if USB2244 is faulty, which causes the flash file system to change to read-only.

202311040370

·         Symptom: The interface information on the S6850-56HF device displays GE 1/0/58, but the GE 1/0/58 interface has been deleted.

·         Condition: This symptom might occur if the S6850-56HF device in the extended operating mode is restarted.

202309111514

·         Symptom: On an EVPN network, rule ARP entries are incorrectly deleted.

·         Condition: This symptom might occur in one of the following conditions:

¡  The number of ARP entries has reached the upper limit and the device cannot learn new ARP entries.

¡  In an M-LAG overlay scenario, ARP packets are received by peer-link interfaces. ARP learning fails because the M-LAG interfaces take priority to learn ARP entries.

¡  If you have configured ARP attack protection features on the device, these features might incorrectly delete valid ARP entries.

¡  Gratuitous ARP packets are injected to the device with gratuitous ARP packet learning unconfigured.

¡  The source and destination IP addresses in the ARP packets reside on different network segments and proxy ARP is not configured.

202212010143

·         Symptom: Member devices in an EVPN M-LAG system do not synchronize ARP flood suppression entries.

·         Condition: This symptom might occur if ARP flood suppression entries move from a remote VTEP to a single-homed interface of an M-LAG member device.

202305311737

·         Symptom: The mac-address mac-learning pdu command fails to be configured if interface splitting is configured first after the device starts up and then the mac-address mac-learning pdu command is configured.

·         Condition: This symptom might occur if interface splitting is configured first after the device starts up and then the mac-address mac-learning pdu command is configured.

202309271663

·         Symptom: After the ND detection feature is enabled in VLAN view and global static IPSG bindings are configured, ND messages from the VLAN cannot be forwarded.

·         Condition: This symptom might occur if global static IPSG bindings are configured and no permit rules are configured on the related VLAN interface.

202204121359

·         Symptom: During a step-by-step ISSU, the OSPFv3 neighbor of a Layer 3 aggregate interface flaps.

·         Condition: This symptom occurs if the following conditions exist:

¡  The upgrade is from F6701L01 to E6702.

¡  The irf mac-address persistent timer command is executed.

202311171824

·         Symptom: The device restarts because the ports on the device fail to come up properly.

·         Condition: This symptom might occur if a module is inserted during the device operation.

202309260387

·         Symptom: The ovsdb-server process abnormally exits with a low probability.

·         Condition: This symptom might occur if the controller deploys the configuration.

202309132118

·         Symptom: The S6850-56HF and S9820-8C devices restart due to a power outage.

·         Condition: This symptom might occur if the fluctuation of the secondary power voltage and interference with power-on control signals can trigger the ADM1166 device to mistakenly shut down.

Resolved problems in E6713

202208050065

·         Symptom: On an LSWM124XG2Q card, the MACsec protocol is down, and the interface is down.

·         Condition: This symptom occurs if the following operations are performed:

a.    Install an SFP-GE-T or SFP-GE-T-D module into the interface on the LSWM124XG2Q card and connect it to an interface on an LSWM124XGT2Q card to establish an MKA session.

b.    Disable MACsec on the interface of the LSWM124XG2Q card.

202209230460

·         Symptom: If multiple dial-in sessions in gRPC gNMI mode issue subscriptions concurrently, a session might fail to receive data.

·         Condition: This symptom occurs if multiple dial-in sessions in gRPC gNMI mode issue subscriptions concurrently.

202211140499

·         Symptom: OSPF BFD sessions flap repeatedly.

·         Condition: This symptom occurs if you use borrowed loopback interface addresses to establish OSPF neighbor relationships, configure BFD for OSPF, and then reboot the device.

202304120566

·         Symptom: Only a maximum of two DSCP mappings in one priority-flow-control dscp-mapping command can take effect.

·         Condition: This symptom occurs when you configure more than two DSCP mappings in one priority-flow-control dscp-mapping command.

202110191607

·         Symptom: When the initialization of the CPU management port fails on the S6850, S9850, S9820, S9820-8C, or S6805 device, the output from the display interface command still shows inbound statistics and CRC errors.

·         Condition: This symptom might occur if the initialization of the CPU management port on the S6850, S9850, S9820, S9820-8C, or S6805 device has failed.

202303020270

·         Symptom: When the speed of a 25G interface is set to 10000 Mbps, the FEC mode configured for that interface does not take effect and is none by default.

·         Condition: This symptom might occur if the FEC mode is set for a 25G interface after the speed of that interface is set to 10000 Mbps.

202303250672

·         Symptom: After the vxlan vlan-based command is executed, a Layer 2 aggregate interface of the access type is moved between VLANs. As a result, ACs cannot be created on the aggregation member ports based on the new VLAN ID.

·         Condition: This symptom might occur if the following conditions exist:

¡  The vxlan vlan-based command is executed.

¡  The link type of the Layer 2 aggregate interface is access.

¡  The Layer 2 aggregate interface is configured as an edge port with the lacp edge-port command.

¡  Member ports of the Layer 2 aggregate interface are in individual state.

202305290148

·         Symptom: The device cannot forward untagged traffic correctly on a VXLAN network.

·         Condition: This symptom might occur if VXLAN is configured and the dhcp snooping trust tunnel command is executed.

202306021684

·         Symptom: The hh3cStackBoardRole node in the MIB Browser can obtain only one role from all member devices in an IRF fabric.

·         Condition: This symptom occurs when you use the MIB Browser to view the information of the hh3cStackBoardRole node for the IRF fabric.

202306082492

·         Symptom: After the configuration of the device is rolled back or the device is rebooted with configuration, ARP requests will be broadcast and Layer 3 unicast packets will be transparently transmitted.

·         Condition: This symptom occurs if a large number of Layer 3 Ethernet interfaces exist on the device when the configuration of the device is rolled back or the device is rebooted with configuration.

202306131838

·         Symptom: The device generates the following alarm message: Operation failed. The minimum interval between IRF bridge MAC address modifications is 30 seconds. Please try again later.

·         Condition: This symptom occurs when you execute the irf mac-address command to configure the IRF bridge MAC address.

202306141412

·         Symptom: The controller disconnects from the device during patch installation and the NETCONF over SOAP configuration on the device is lost.

·         Condition: This symptom occurs when you install a patch by using the controller, which restarts the xmlcfgd process.

202306161118

·         Symptom: After a Layer 2 interface is rolled back to a member port of a Layer 3 aggregation group enabled with PIM-SM, it fails to forward to multicast traffic because no VFP ACL is deployed to it.

·         Condition: This symptom occurs after a Layer 2 interface is rolled back to a member port of a Layer 3 aggregation group enabled with PIM-SM.

202306190397

·         Symptom: Traffic forwarding exceptions occur on a 1G interface or an interface negotiated to 1G.

·         Condition: This symptom might occur on a 1G interface or an interface negotiated to 1G.

202307041942

·         Symptom: No prompt message is displayed when you configure multiport ARP entries or multiport MAC entries without specifying a service loopback interface on an underlay network. The entries cannot be deployed to the hardware.

·         Condition: This symptom occurs if you do not specify a service loopback interface when you configure multiport ARP entries or multiport MAC address entries.

202307061934

·         Symptom: ARP messages are broadcasted instantaneously.

·         Condition: This symptom might occur if the device restarts or a subcard is re-installed.

202307111015

·         Symptom: A QoS policy applied to an aggregate interface does not take effect on new member ports that join the aggregation group.

·         Condition: This symptom occurs if you apply a QoS policy to an aggregate interface and then add new ports to the aggregation group.

202307111828

·         Symptom: The arp snooping enable command failed to be executed in VLAN view on some products and the system prompted an operation failure.

·         Condition: This symptom occurs when you execute the arp snooping enable in VLAN view on some products. The command has a command word that specifies RARP packet redirection to the CPU, which is not supported by some products.

202307210952

·         Symptom: On an M-LAG system with VRRP configured, the backup device in a VRRP group cannot learn ARP entries.

·         Condition: This symptom might occur if the backup device receives ARP unicast packets.

202308110863

·         Symptom: After sending ICMP redirect messages and sending ICMPv6 redirect messages are enabled synchronously, they always function even if you disable them.

·         Condition: This symptom might occur if both the ip redirects enable command and the ipv6 redirects enable command are executed synchronously to enable sending ICMP and ICMPv6 redirect messages.

202308141250

·         Symptom: When DHCPv6 relay agent features are configured on a device, the device delivers traversing unicast IPv6 DHCP packets to its CPU, causing packet loss.

·         Condition: This symptom might occur if DHCPv6 relay agent features are configured on the device.

202309041408

·         Symptom: The device reboots repeatedly and cannot start up correctly.

·         Condition: This symptom occurs if you perform the following operations:

a.    Configure PTP in INT-PTP mode and then save the configuration and reboot the device.

b.    Modify the hardware-resource firmware mode to INT-BFD.

202309042691

·         Symptom: A BFD session flapping occurs.

·         Condition: This symptom occurs with a certain probability if you bring up a device interface with the undo shutdown command to perform bulk software BFD negotiation and bring up the BFD session. The BFD session might come up after a flapping.

202309071510

·         Symptom: The primary device and the secondary device in an M-LAG system have different DHCP snooping entries within a period.

·         Condition: This symptom occurs if DHCP snooping is disabled when the peer-link interfaces are flapping and is enabled after peer-link interface flapping.

202309071523

·         Symptom: DHCPv6 commands fail to be deployed after users come online and the DHCP process starts

·         Condition: This symptom occurs if the address pool uses the prefix configured by the ipv6 prefix command and the network prefix command is configured in the address pool.

202309110535

·         Symptom: When an IRF fabric forwards traffic across member devices, it cannot obtain outbound traffic statistics for a GRE tunnel interface even if that interface is configured with the statistics l3-packet enable outbound command.

·         Condition: This symptom occurs if the IRF fabric forwards traffic across member devices and cannot match the next hop class ID of the output interface for collecting outbound statistics for tunneled traffic.

202309110575

·         Symptom: An IPv4 ACL will be deployed if a traffic class in a QoS policy applied to the outbound direction is configured to match the EtherType.

·         Condition: This symptom occurs if a traffic class in a QoS policy applied to the outbound direction is configured to match the EtherType.

202309111302

·         Symptom: The device sends ICMP error packets if only Layer 2 multicast is configured.

·         Condition: This symptom occurs if you enable only Layer 2 multicast, enable the IGMP snooping querier, and enable sending ICMP destination unreachable messages.

202309120444

·         Symptom: On an IRF fabric, a host still receives multicast data from an output interface after sending a leave message.

·         Condition: This symptom occurs if the output interface is an aggregate interface.

202309132177

·         Symptom: After an IRF master/subordinate switchover, the MAC address of a VLAN interface repeatedly moves among the physical interfaces in that VLAN if BFD MAD is enabled on that VLAN interface.

·         Condition: This symptom occurs if you enable MAC move on an IRF fabric.

202309261948

·         Symptom: Statistics about ECMP resources are incorrect.

·         Condition: This symptom might occur if ECMP flaps.

202306060568

·         Symptom: Other routers cannot learn the OSPF external routes and default routes advertised by a device.

·         Condition: This symptom might occur if no interfaces are added into the OSPF process on the device and the nssa default-route-advertise command is configured in the OSPF process.

202305091145

·         Symptom: The ARP entries learned through LLDP on the local device are refreshed, which might cause route flapping.

·         Condition: This symptom occurs when the description information of the peer device, such as the system name, is changed.

202305190212

·         Symptom: A new switch ID does not take effect after modification.

·         Condition: This symptom might occur if the mirror-to interface command is executed with the erspanv3 and sampler keywords to modify the switch ID in ERSPANv3 mirroring packets.

202306072180

·         Symptom: On a physical interface of a DCI ED, cross-subnet rate limiting does not take effect on traffic that carries an unknown inner source MAC address.

·         Condition: This symptom might occur if L2VPN is enabled, tunneled traffic with an unknown inner source MAC address is received, and cross-subnet rate limiting is configured on a physical interface.

202307131776

·         Symptom: In the IRF environment, the BFD session established on the backup card interface repeatedly flaps,

·         Condition: This symptom occurs if you reboot the device after configuring BFD session parameters in the IRF environment.

202307211459

·         Symptom: CFD packets are not forwarded correctly.

·         Condition: This symptom occurs if you execute the cfd enable command globally, and then execute the undo cfd enable command for the device.

202307260983

·         Symptom: The virtual IP address of the IPv4 VRRP group configured on a VLAN interface cannot be pinged.

·         Condition: This symptom occurs with a low probability if you repeatedly enable and disable the spanning tree protocol globally when the spanning tree operates in PVST mode in a network configured with M-LAG and VRRP.

202309121717

·         Symptom: The spanning tree status displayed in the output from the display stp command is inconsistent with the actual condition.

·         Condition: This symptom occurs if you add a port that does not belong to any STG to the M-LAG aggregation group.

202207071638

·         Symptom: When the device is enabled with BGP Graceful Restart (GR) capability and has BGP sessions to non-H3C devices, the GR feature does not take effect.

·         Condition: This symptom occurs when the following conditions exist:

¡  The device is enabled with BGP GR capability and has established BGP sessions to non-H3C devices.

¡  The value for the Forwarding State(F) field is 0 in the OPEN messages from those non-H3C devices. In this situation, if you use the display bgp peer verbose command on the local device, the Forwarding State preserved by Peer for following Address families field in the command output is empty.

202212010145

·         Symptom: Member devices in an EVPN M-LAG system do not synchronize ARP flood suppression entries.

·         Condition: This symptom might occur if ARP flood suppression entries move from a remote VTEP to a single-homed interface of an M-LAG member device.

202303131032

·         Symptom: In the output of the display ipv6 interface command, the IPv6 address, interface name, and VPN fields are displayed on different lines, which should be displayed on the same line.

·         Condition: This symptom occurs if you execute the display ipv6 interface command.

202304100930

·         Symptom: Packets matching a QoS policy applied to the outbound direction of a VSI interface are mistakenly dropped.

·         Condition: This symptom might occur if the following operations are performed:

a.    Apply QoS policies to the outbound direction of two or more VSI interfaces.

b.    Apply a QoS policy to the outbound direction of a physical interface.

202304141939

·         Symptom: A network interruption occurs on the S9820-8C-SAN or S6850-56HF-SAN device during the process of transferring a large file via FTP.

·         Condition: This symptom might occur if the device uses FTP to transfer large files.

202304191794

·         Symptom: BFD flapping occurs on the device with a low probability.

·         Condition: This symptom occurs if a VSI interface receives a large number of ARP or RARP packets in an EVPN network, and VSIs or VSI configuration is changed.

202305110201

·         Symptom: On a multicast VXLAN network, multicast traffic cannot be forwarded.

·         Condition: This symptom occurs if the device starts with the factory defaults and then you configure multicast VXLAN in the following order: first configure tunnels and VSIs, and then configure multicast.

202305200097

·         Symptom: The xmlcfgd process on the device has exceptions after a patch is loaded on the device. Then, the controller cannot be reconnected to the device or the NETCONF feature has exceptions.

·         Condition: This symptom might occur if a patch for the xmlcfgd process is loaded on the controller connected to the device.

202306152032

·         Symptom: Static MAC address entries are configured for overlay MAC addresses that have been dynamically learned by an M-LAG member device, and the static MAC address entries cannot be synchronized to the M-LAG peer.

·         Condition: This symptom might occur if static MAC address entries are configured for overlay MAC addresses that have been dynamically learned by the M-LAG member device.

202307181232

·         Symptom: The peer-link interface incorrectly learns ARP entries. As result, remote IP addresses on the same subnet might fail to be accessed.

·         Condition: This symptom occurs if proxy ARP is configured for the M-LAG dual-active VLAN gateways or VRRP, and automatic ARP scanning is enabled by using the arp scan auto enable command.

202307200892

·         Symptom: Due to hash conflicts, the creation of AC-attached interfaces fails without any log prompts on the device.

·         Condition: This symptom might occur if a large number of AC-attached interfaces are configured on the device.

202307210976

·         Symptom: After you configure the network command in OSPF view, OSPF neighbor relationship establishment fails.

·         Condition: This symptom might occur if you perform the following operations:

a.    Add or remove the ospf 1 area and ospf network-type configurations.

b.    Configure the network command in OSPF view.

202309041301

·         Symptom:On an M-LAG system, packets received on an M-LAG interface are forwarded over the peer link and then forwarded out of the M-LAG interface in the same M-LAG group.

·         Condition:This symptom might occur if an M-LAG aggregation group flaps.

202309090399

·         Symptom: After the speed or duplex mode is changed for the copper management Ethernet interface of an S6850-56HF-SAN switch, the device fails to be pinged.

·         Condition: This symptom might occur if you change the speed or duplex mode for the copper management Ethernet interface of an S6850-56HF-SAN switch.

Resolved problems in E6712P01

None.

Resolved problems in E6712

202305041124

·         Symptom: The switch cannot obtain an IPv6 address after it is rebooted, and IPv6 automatic deployment fails.

·         Condition: This symptom occurs if the controller deploys the configuration to change the hardware resource mode during automatic deployment and the controller does not assign a fixed IPv6 address.

202305101923

·         Symptom: The system sends probe packets out of the VSI interface to the IP address in an aging ARP suppression entry even if the interface has been down.

·         Condition: This symptom might occur if you enable ARP suppression for a VSI and the VSI interface is down.

202305120015

·         Symptom: Packets matching a QoS policy applied to the outbound direction of a VSI interface are mistakenly dropped.

·         Condition: This symptom might occur if the following operations are performed:

a.    Apply QoS policies to the outbound direction of two or more VSI interfaces.

b.    Apply a QoS policy to the outbound direction of a physical interface.

202305180008

·         Symptom: In an EVPN or VXLAN distributed gateway network, when the device receives a tunneled packet with a source IP address the same as a VSI interface address, the device will reply with a gratuitous ARP response, which can lead to high CPU usage.

·         Condition: This symptom might occur if the distributed gateways perform ARP probing in response to traffic.

202305120953

·         Symptom: In a VXLAN network, VXLAN tunnel interfaces are not outgoing interfaces in multicast forwarding entries, and multicast forwarding fails.

·         Condition: This symptom might occur if VXLAN tunnels and VSIs are configured prior to multicast forwarding.

202305122075

·         Symptom: In a VXLAN network, traffic loss persists for longer than 2 seconds during an ISSU load.

·         Condition: This symptom might occur if an ISSU load or active/standby MPU switchover occurs in a VXLAN network.

202305041113

·         Symptom: The BFD session cannot come up.

·         Condition: This symptom occurs if frequent flapping occurs for the physical interfaces attached to the link associated with the BFD session.

202305220010

·         Symptom: In an EVPN network, the CPU usage of a leaf device is very high.

·         Condition: This symptom occurs if attached devices move between leaf devices, and the leaf device is enabled with ARP suppression.

202305232222

·         Symptom: Packets that enter the device from an AC on a Layer 2 aggregate interface cannot be forwarded out through the Layer 2 aggregate interface.

·         Condition: This symptom occurs if the aggregate interface has multiple member ports and the Selected states the member ports change when the AC is configured.

202304070641

·         Symptom: In the display mac-address statistics command output, the Total Multicast and Multiport MAC Addresses Available field displays 4000.

·         Condition: This symptom occurs if you execute the display mac-address statistics command to display MAC address table statistics after the device is rebooted.

202304060743

·         Symptom: A MAC authentication user is still online and cannot go offline after the offline detection timer expires.

·         Condition: This symptom occurs if a MAC authentication user stops sending traffic after coming online successfully. After the offline detection timer expires, execute the display mac-authentication command to identify whether the MAC authentication user is offline.

202305041120

·         Symptom: With VLAN-based VXLAN assignment enabled, after the permitted VLAN is modified on a Layer 2 aggregate interface of the access type, the Ethernet service instance corresponding to the VLAN cannot be generated.

·         Condition: This symptom occurs if you modify the permitted VLAN on the Layer 2 aggregate interface of the access type that is configured as an edge aggregate interface.

202305200364

·         Symptom: In the configuration file, the key for BGP MD5 authentication is displayed in plain text.

·         Condition: This symptom occurs if the value for the PasswordType field in the BGP CfgSession table is 2.

202305180009

·         Symptom: Route flapping occurs when ARP entries are updated due to LLDP information modification.

·         Condition: This symptom occurs if the LLDP packet length changes on the peer. For example, the sysname of the peer is edited.

Resolved problems in E6711

202212050034

·         Symptom: In an EVPN VXLAN network, a third-party device with an AC attached reports IPv6 address conflicts.

·         Condition: This symptom might occur if the peer H3C device has ND flood suppression enabled for the related VSI and no VSI interface is assigned to the VSI.

202207011235

·         Symptom: When MVXLAN supports M-LAG, the packets sent out of an M-LAG device are more than those received.

·         Condition: This symptom occurs if a large number of entries are configured on the M-LAG device, and the DCI and source configurations repeatedly change.

202208190593

·         Symptom: A QoS policy fails to be applied to a Layer 2 aggregate interface.

·         Condition: This symptom occurs if the traffic behavior is configured with a traffic policing action.

202208050810

·         Symptom: The Effective count field is always displaying 0 in the output from the display grpc verbose command.

·         Condition: This symptom might occur if you execute the display grpc verbose command after configuring periodical sampling in gRPC dial-out mode.

202302130821

·         Symptom: When an aggregation member port comes up,  its state will transition to Selected, then Unselected, and finally back to Selected again.

·         Condition: This symptom occurs if the following operations are performed:

a.    Configure the short LACP timeout interval on the port.

b.    Execute the link-delay down 5 command to configure the link state change suppression interval as 5 seconds on the port.

c.    Bring up the port.

202302140214

·         Symptom: Packets that are flow-mirrored to an interface are encapsulated abnormally.

·         Condition: This symptom occurs if you specify the IPv6 encapsulation method and configure a sampler when you configure flow-mirroring traffic to an interface.

202302090664

·         Symptom: In an M-LAG system, the peer-link interface learns MAC addresses incorrectly, leading to generation of abnormal forwarding entries.

·         Condition: This symptom might occur if one of the following conditions exists:

¡  The telemetry ifa device-id x.x.x.x command is executed on an S6850-56HF or S6850-2C device, and the LPORT value of a physical member interface of the peer link and the port number reserved for an internal loopback interface of INT are the same.

¡  Residual IRF physical interface configurations exist on the device, and the internal port number of the IRF physical interface is the same as the LPORT value of a physical member interface of the peer link.

202302131035

·         Symptom: Data collected in the ifmgr/trafficstatistics/interfaces path is abnormal. The collected data at the NMS side displays a jagged line in the data chart.

·         Condition: This symptom might occur when you specify the ifmgr/trafficstatistics/interfaces data collection path in gRPC dial-out mode.

202301170521

·         Symptom: Lower-layer ACL entries of OpenFlow entries are lost after VMs migrate, and some traffic fails to be forwarded.

·         Condition: This symptom might occur if VMs migrate.

202302031011

·         Symptom: The buffer apply command does not take effect on interfaces on a subcard or new breakout interfaces.

·         Condition: This symptom occurs if you insert a subcard or split an interface after executing the buffer apply command.

202302200113

·         Symptom: In an M-LAG system, after an ND entry migrates between the member devices, its outgoing interface becomes incorrect, and traffic interruption occurs.

·         Condition: This symptom might occur if an M-LAG member device reboots and synchronizes the ND entries of the M-LAG peer, and then an ND entry migrates.

202302070374

·         Symptom: After an IRF member device is rebooted, the ports on the device fail to come up.

·         Condition: This symptom might occur if the following conditions exist on the rebooted IRF member device:

¡  The 25-GE ports on LSWM124TG2H interface modules in the S6850-56HF, S6850-2C, or S9850-4C switch are configured to operate at 10 Gbps. The duplex full command is executed on a port in a port group. (Every four ports in ascending order of port number are organized into one group.)

¡  A large number of ports have modules installed but do not come up or have the shutdown command executed.

202209291014

·         Symptom: Duplicate multicast packets exist after one of the two outgoing tunnels for a static multicast group is flapped.

·         Condition: This symptom occurs if the static multicast group has more than 128 VSI interfaces as the outgoing interfaces and the actual outgoing interface is a tunnel interface.

202301120577

·         Symptom: After an incremental patch is uninstalled, the display boot-loader command does not display information about a non-incremental patch.

·         Condition: This symptom occurs if both an incremental patch and a non-incremental patch are installed.

202301150100

·         Symptom: Failed to apply ERSPAN-type QoS policies due to insufficient resources.

·         Condition: This symptom occurs if you apply a ERSPAN-type QoS policy to the inbound direction globally or apply ERSPAN-type QoS policies to the inbound direction of more than four interfaces.

202302090611

·         Symptom: The log file fabric.log generated by VCF fabric exhausts the memory.

·         Condition: This symptom occurs if the automated deployment scenario of VCF fabric runs for a long period of time or interfaces flap.

Resolved problems in R6710

202208241285

·         Symptom: A QoS policy applied to a control plane cannot filter the protocol packets to the control plane

·         Condition: This symptom occurs when you apply a QoS policy to a control plane to filter protocol packets.

202209211461

·         Symptom: A port that should be down periodically, repeatedly comes up and goes down.

·         Condition: This symptom occurs if the following conditions exist:

¡  The speed 1000 command is executed on a 10-GE port.

¡  The port has an SFP-GE-T/SFP-GE-T-D transceiver module installed.

¡  The shutdown command is not executed on the port.

202302070183

·         Symptom: After the igmp-snooping router-port-discard command is executed in a VLAN, the switch cannot forward IGMP protocol packets to the router ports in the VLAN.

·         Condition: This symptom occurs if the igmp-snooping router-port-discard command is executed in a VLAN.

202210202895

·         Symptom: IPv6 autoconfiguration fails on a device.

·         Condition: This symptom might occur when the device uses a third-party DHCPv6 server for IPv6 autoconfiguration and the DHCPv6 server checks the DUID in Option 1.

202211301422

·         Symptom: The device does not respond or reboots unexpectedly because the Intel C3XXX CPU is faulty.

·         Condition: This symptom occurs with a low probability when the system runs for a long time.

202212190854

·         Symptom: Multicast traffic cannot be broadcast in a VLAN when the hardware resource operating mode of the switch is ARP or MAC.

·         Condition: This symptom occurs if the hardware resource operating mode of the switch is ARP or MAC.

202207011516

·         Symptom: The link-delay down setting is configured on both ends of a link, and the link traverses an optical transmission device. After an optical link switchover occurs on the optical transmission device, the link-delay down setting does not take effect.

·         Condition: This symptom occurs if an optical link switchover occurs on the optical transmission device.

202207060567

·         Symptom: In the outbound direction of an interface, mirroring traffic to the CPU and packet capturing do not take effect.

·         Condition: This symptom occurs if you configure mirroring traffic to the CPU and packet capturing in the outbound direction of an interface.

202209130556

·         Symptom: Memory leaks occur.

·         Condition: This symptom occurs if the following conditions exist:

¡  The device uses a large number of VLAN interfaces and Layer 2 interfaces to forward multicast traffic.

¡  The device receives multicast traffic of different source-group pairs after multicast entries age out.

202208030541

·         Symptom: The MACsec protocol flaps on a subcard, and packet loss occurs.

·         Condition: This symptom occurs if another subcard configured with MACsec is removed and reinserted.

202208031661

·         Symptom: The LED for the master's IRF physical interface is on, and the LEDs for the subordinates' IRF physical interfaces are off.

·         Condition: This symptom might occur if the switch loads BootWare in full startup mode and performs system inspection.

202208161569

·         Symptom: The shutdown configuration on an interface of a subcard in a slot is lost after a subcard in another slot is removed and reinserted.

·         Condition: This symptom occurs if a subcard in another slot is removed and reinserted after the switch is deployed automatically.

202209010367

·         Symptom: The traffic rate of the ifmgr/statistics node appears as a saw-tooth in Grafana.

·         Condition: This symptom occurs if the following conditions exist:

¡  The json row-timestamp enable command is executed to enable per-row time-stamping for JSON-encoded subscription data.

¡  The switch uses gRPC to report the ifmgr/statistics node data at 5-second intervals.

202209270925

·         Symptom: If the buffer egress/ingress command is executed on an interface and then the operating mode of the interface is switched between Layer 2 and Layer 3, the buffer egress/ingress command configuration in the configuration file is lost. However, the configuration still takes effect.

·         Condition: This symptom occurs if the buffer egress/ingress command is executed on an interface and then the operating mode of the interface is switched between Layer 2 and Layer 3.

202210211599

·         Symptom: The switch gets stuck when the NETCONF <get-config> operation is used to retrieve information about the Ifmgr/Suppression table. The switch remains stuck until the set timeout time expires.

·         Condition: This symptom occurs when the NETCONF <get-config> operation is used to retrieve information about the Ifmgr/Suppression table.

202210191104

·         Symptom: VXLAN default decapsulation can be deployed from the CLI. When this feature is deployed through NETCONF, the device reports deployment success even through the deployment fails.

·         Condition: This symptom might occur if VXLAN default decapsulation is deployed through NETCONF.

202210171204

·         Symptom: A VSI with the igmp-snooping drop-unknown setting configured still can forward unknown multicast traffic in a VXLAN Layer 2 multicast environment.

·         Condition: This symptom might occur if the igmp-snooping drop-unknown command is executed in VSI view in a VXLAN Layer 2 multicast environment.

202210141248

·         Symptom: The BGP process is abnormal.

·         Condition: This symptom might occur if FRR is enabled globally for EVPN VXLAN.

202210110425

·         Symptom: The switch reboots after you apply a QoS policy to an interface and delete all actions in the traffic behavior in the QoS policy.

·         Condition: This symptom occurs if you apply a QoS policy to an interface and delete all actions in the traffic behavior in the QoS policy.

202210100072

·         Symptom: The switch cannot ping a directly connected device.

·         Condition: This symptom occurs if you use a physical port with a port ID greater than 127 on the switch for the ping operation. The port ID refers to the value of the Port field in the debug port mapping slot command.

202209261332

·         Symptom: When a device is cold or warm rebooted, the MIB nodes hh3cSysColdStartTrap and hh3cSysWarmStartTrap do not normally report traps.

·         Condition: This symptom occurs if the device is configured with SNMP and then rebooted.

202210090116

·         Symptom: On an EVPN VXLAN network, the outgoing traffic of an AC is forwarded in queue 2 by default, which is different from the queue to which the 802.1p priority is mapped.

·         Condition: This symptom occurs if the incoming port of an AC trusts the 802.1p priority of packets and the access mode of the AC is VLAN on an EVPN VXLAN network.

202211150761

·         Symptom: The CPU usage of a device is high.

·         Condition: This symptom occurs if a port is installed with an SFP copper port module SFP-GE-T/SFP-GE-T-D and the shutdown command is executed to shut down the port.

202210210266

·         Symptom: A MAC address might fail to move between IRF member devices.

·         Condition: This symptom occurs if a device attached to an IRF fabric moves between IRF member device and sends Layer 3 traffic to the IRF fabric.

202211100661

·         Symptom: On an EVPN VXLAN network, some routes cannot guide traffic forwarding.

·         Condition: This symptom occurs if multiple VPN instances exist on an EVPN VXLAN network and the automatically created VPN routes sent for the first time are withdrawn.

202210270699

·         Symptom: On a VXLAN network, broadcast traffic cannot be forwarded.

·         Condition: This symptom occurs if VSIs are repeatedly added and deleted on the device and the VFI entry index in the hardware exceeds 4095.

202211181011

·         Symptom: On a DRNI+EVPN network, a MAC address is migrated from an IPP to a single-homed interface, and the single-homed interface is shut down by the shutdown command. However, the ARP entries learned by the single-homed interface are not deleted.

·         Condition: This symptom occurs if the device MAC address is migrated from an IPP to a single-homed interface, and the single-homed interface is shut down by the shutdown command.

202211161812

·         Symptom: ARP packets cannot be flooded to different AC ports and tunnels in the same VSI.

·         Condition: This symptom occurs if the following steps are performed:

a.    Execute the arp suppression enable command to enable ARP suppression in a VSI view.

b.    Execute the gateway vsi-interface command to specify a gateway interface for the VSI, and then delete this setting.

c.    Remove ARP suppression settings.

202211090166

·         Symptom: When a 1-Gbps transceiver module is installed into an interface of an LSWM124TG2H subcard and the interface is configured with the speed 1000 command, the interface cannot come up.

·         Condition: This symptom occurs if the following operates are performed:

a.    A 1-Gbps transceiver module is installed into an interface of an LSWM124TG2H subcard and the interface is configured with the speed 1000 command.

b.    Then the speed 25000 or speed auto command is executed.

c.    The speed 1000 command is executed at last.

202211300367

·         Symptom: After you use the ecmp mode enhanced command to enable the enhanced ECMP mode, the hash consistency of ECMP routes will be affected by the flapping of next hops of unrelated routes.

·         Condition: This symptom occurs if you use the ecmp mode enhanced command to enable the enhanced ECMP mode and the next hops of multiple unrelated routes flap.

202211110382

·         Symptom: The display mac-address command does not display the dynamic secure MAC address entries and static secure MAC address entries of an aggregate interface.

·         Condition: This symptom occurs if the display mac-address command is executed on the device to display the MAC address entries of an aggregate interface.

202211171005

·         Symptom: After the device is rebooted, the mirroring configuration is lost, but the hardware resources in the underlayer remain. As a result, the later mirroring configuration fails.

·         Condition: This symptom occurs if the following operations are performed:

a.    Configure port mirroring.

b.    Specify multiple mirroring source ports for a mirroring group. First configure the member ports of an aggregate interface as mirroring source ports, and then configure the aggregate interface as a mirroring source port. 

c.    Save the configuration and reboot the device.

202212011911

·         Symptom: On an EVPN M-LAG network, the device cannot forward ARP requests from endpoints and cannot send ARP replies with the gateway address to the endpoints.

·         Condition: This symptom occurs if the VSI interface with ARP flood suppression enabled and acting as the gateway is down. 

202209120036

·         Symptom: A QoS policy has multiple class-behavior (CB) associations (for example, CB associations 1 and 2). Typically, only CB association 1 takes effect on traffic matching both CB associations 1 and 2. If you modify the actions in CB association 1, CB association 2 might take effect on the traffic matching both CB associations with a low probability.

·         Condition: This symptom occurs if the following operations are performed:

a.    In a QoS policy applied, the accounting or CAR action exists in a behavior.

b.    Modify the actions in any behavior or modify the match criteria in any class of the QoS policy, or apply the QoS policy again.

202209070077

·         Symptom: A DCI tunnel flaps or multiple next hops of a DCI tunnel switch. As a result, traffic matching the PBR policy cannot be forwarded.

·         Condition: This symptom occurs if the next hop of the PBR policy is a DCI tunnel, and the DCI tunnel flaps or multiple next hops of the DCI tunnel switch.

202209160292

·         Symptom: The device might not send some routes to peers.

·         Condition: This symptom occurs if the device filters the route receivers based on the first AS number in EBGP routes (configurable with the peer-as-check enable command) and exchange routes with multiple EBGP peers.

202209200263

·         Symptom: On an EVPN network, another device might fail to ping the IPv4 or IPv6 address of the local device.

·         Condition: This symptom occurs if microsegments are applied on the EVPN network.

202208050116

·         Symptom: When you modify the parameters for WRR queuing or WFQ queuing, packets will be dropped transiently.

·         Condition: This symptom occurs if you modify the parameters for WFQ queuing or WRR queuing.

202208090444

·         Symptom: The switch disconnects from the BMP server frequently.

·         Condition: This symptom occurs if the number of messages sent over the TCP connection between the switch and the BMP server exceeds the limit allowed by the buffer.

·         Remarks: When the switch detects that the number of messages sent over the TCP connection exceeds the limit allowed by the buffer, it prints the following log message: BGP BGP instance name: Disconnected from BMP Server BMP server number for maximum limit of sending buffer reached. To resolve the issue, configure the BMP server to use the asynchronous receive mode. If the issue persists, configure the BMP server to receive messages faster or contact H3C Support.

202208090448

·         Symptom: BFD sessions flap after the multicast routing command is executed.

·         Condition: This symptom occurs if the multicast routing command is executed.

202206240525

·         Symptom: The flow mirroring configuration affects the PBR function.

·         Condition: This symptom occurs if both flow mirroring and PBR are configured.

202208050141

·         Symptom: In an M-LAG system configured with multicast settings, ICMPv6 packets are matched by an IPv4 ACL.

·         Condition: This symptom occurs if the ICMPv6 packets have a specific format.

202207050544

·         Symptom: In an EVPN-DCI network, an ED device fails to forward IP unicast packets at Layer 2.

·         Condition: This symptom might occur if the ED device removes VXLAN encapsulation from the IP unicast packets whose inner TTL is 1 and forwards the packets at Layer 2.

202208021466

·         Symptom: Some behaviors do not take effect in a QoS policy that contains multiple class-behavior associations.

·         Condition: This symptom occurs if the class of the first class-behavior association is configured to match an IPv4 ACL and Layer 2 attributes.

202206120009

·         Symptom: The value of the IGMPv3 specific queries field in the display igmp-snooping statistics command output is incorrect.

·         Condition: This symptom occurs if an IGMPv3 leave message with the multicast group as the Exclude mode triggers the sending of IGMPv3 last member queries.

202206060838

·         Symptom: On a multicast network that supports multiple levels of DR systems, IGMP packets form a loop between DR interfaces.

·         Condition: This symptom occurs if DR interfaces receive IGMP group-specific query packets on a multicast network that supports multiple levels of DR systems.

202207050871

·         Symptom: Executing the undo mac-address mac-learning enable command in system view or executing the undo mac-learning enable command in VSI view cannot disable MAC address learning in the corresponding view.

·         Condition: This symptom occurs when you execute the undo mac-address mac-learning enable command in system view or execute the undo mac-learning enable command in VSI view.

202208190680

·         Symptom: In an M-LAG system that use VLAN interfaces to act as dual-active gateways for the same VLAN, an M-LAG member device cannot successfully ping an endpoint.

·         Condition: This symptom occurs if the other M-LAG member device receives ICMP replay packets.

202206230765

·         Symptom: The switch prompts the Permission denied. error message.

·         Condition: This symptom occurs if the following operations are performed:

a.    Enable command authorization.

b.    Repeat a command in the command history buffer for more than 1000 times.

202208221073

·         Symptom: Loop detection is enabled and the loop protection action is set to block on an interface.  When the interface detects a loop, it stays in forwarding state, and the loop is not removed.