Resolved issuesfor 8400 Switchesin AOS-CX 10.10.0002

Symptom: Routes missing after moving from AFC to NetEdit as a configuration tool.

Scenario: This issue occurs in a deployment with AFCand Net edit where config might be modifying/updating a NULL default Port VRF UUID to the actual UUID of the Port VRF.

Workaround: Perform ashut/no shuton the interfaces.

Symptom: The log messageRange Checker uninstallation has failedis seen in the event log.

Scenario: When removing a line card, the switch may report these messages.

Workaround: There is no functional impact.

Symptom: User information for a 2nd locked out user or any subsequent locked out user will not be shown.

Scenario: This issue occurs if more than one user is locked out at the same time.

Symptom: The switch will reboot/crash when a user issues the明确艺术多联机<多联机名称> ipcommand.

Scenario: When theclear arp vrf ipcommand is issued memory can be accessed incorrectly.

Symptom: From the CLI, one of the following errors will be seen:

• Could not copy image.
• Could not access primary switch image.
• Could not access secondary switch image.

Scenario: After updating the software image via the WebUI, copying or uploading a firmware image off the switch or to USB may fail when using a different user account than used to update via the WebUI.

Workaround:

Option 1) Copy or upload the switch firmware image to a remote source using the same user account as was originally used to update the firmware via the WebUI.

Option 2) Re-download the switch firmware image using TFTP, SFTP, SCP, or USB using any administrator user account.

Symptom: When there are multiple IPv6 prefix lists configured with different length of prefixes, the CLI stops responding while trying to execute commands likeshow run,copy support-filesor when trying to save the configuration.

Scenario: This issue occurs when there is more than one IPv6 prefix-list configured and the length of a prefix configured in a prefix list is more than the length of the prefix configured in the first prefix-list.

Symptom: If there are two or more LAGs attached to the same hardware ID, software will be configured with two different LAGs but they both will share the hardware instance. Hence, one of the LAG ports does not pass any traffic because of the missing hardware ID.

Scenario: This issue is observed in the following scenario:

1. Create a dynamic LAG 10int lag 10 lacp mode active no shut

2. Create a dynamic LAG 20 int lag 20 lacp mode active no shut. At this point, lag10 will have hardware ID as1and lag20 has hardware ID as "2".

3. Convert lag10 as a static LAG.int lag 10 no lacp mode active step

4. Create another LAG 30int lag 30 lacp mode active no shutAt this point, lag10 and lag30 will share the same hardware ID "1".

Workaround: Delete one lag10 and create it again.

Symptom: In certain configurations, RPVST inconsistencies may be encountered.

Scenario: When a 3rd party devices, such as a WAN Optimizer is in-path between switches in the same RPVST domain, and there is also a failure path, the might be inconsistencies in RPVST domain.

Symptom:In a VSX setup, the inter-switch link (ISL) flaps continuously.

Scenario:This issue can occur when the ISL is configured as one physical port on both switches and then changed to a static LAG on just one of the VSX switches.

Workaround:Configure a dynamic LAG (lacp mode active) on both switches.

Symptom: The hpe-routing daemon may generate a core-dump.

Scenario: This issue can occur in rare conditions, in a VxLAN environment with multi-tiered VSX, when there are multiple EVPN duplicate MAC dampening events.

Workaround:切换VxLAN interfaces using "shutdown" and "no shutdown" commands.

Symptom: The switch does not list the configured DNS servers in the order they were entered.

Scenario: When the switch is configured with multiple DNS server using theip dns-server-address command, the configuration is incorrectly listed in the output of theshow ip dnscommand.

Workaround: This is a display issue. The DNS servers are correctly programmed based in the order of configuration.

Symptom: An EVPN route is selected as best route over an eBGP route.

Scenario: This issue occurs when the same prefix is learned from EVPN and eBGP. For selecting a FIB route, local preference is taken into consideration over administrative distance.

Workaround: Configure a higher local preference to eBGP routes than EVPN routes.

Symptom: CPU usage spikes.

Scenario: This issue can occur when continuous IGMP control packets from multicast clients are sent a high rate.

Workaround: Identify the groups causing the spike and apply an ACL using the command:ip igmp snooping apply access-list ip igmp apply access-list .