The following are new open issues for supported features found in ExtremeXOS 22.3.
Open Issues, Platform-Specific, and Feature Change Requests (CRs)
| CR Number | Description |
|---|---|
| ExtremeSwitching X690 | |
| xos0067489 | Link flaps occur when optics is removed, and then re-inserted, and after reboots. |
| xos0067675, xos0068380 | EEE does not work on ExtremeSwitching X690-48t switches. |
| xos0068917 | On ExtremeSwitching X690-48x switches, port 8 with 1G optics does not link up with autonegotiation turned on. |
| xos0067933 | On ExtremeSwitching X690 series switches, the following behavior occurs when
setting autonegotiation:
|
| xos0067380 | On the ExtremeSwitching X690 series switches, when a VPLS service VMAN-tagged access port is configured with any tagged VLAN, packets with a VLAN tag that equals the VMAN tag are incorrectly transmitted over the VMAN VPLS instance. |
| xos0068904 | On ExtremeSwitching X690-48t-2q-4c switches, ports 1–48 do not link at 100 Mbps when auto-negotiation is enabled. |
| xos0068971 |
On ExtremeSwitching X690 series switches, port 72 configured for 25G does not
link up with autonegotiation turned on.
Workaround: Disable autonegotiation on port 72 if configuring for 25G, which is the default
setting.
|
| ExtremeSwitching X440-G2 | |
| xos0067772 | Link flap occurs when 1m and 3m SFP+ DAC cables are inserted in ports 49–52 of ExtremeSwitching X440G2-48p-10G4 switches. This problem also occurs when ports are disabled, and then enabled. |
| xos0062773 | After switch boot up or restart of process dot1ag, 95% CPU utilization occurs with 32 UP MEPs (maximum value). |
| Summit X460-G2 Series Switches | |
| xos0067840 |
OSFPv3 routes are not redistributed into BGP after a second or
third failover with BGP peer-group configuration.
Workaround: Reboot entire stack to redistribute
the routes.
|
| xos0067622 | On Summit X460G2-24t-24ht switches, counter for undersized packets (less than 64 byte) does not increment as shown in the output of the command show ports port rxerror; however, the counter does increment when oversized packets (greater than 1518 bytes) are sent to the switch. |
| xos0066212 | VRRP flap occurs during state transition when a single VRRP group contains 2,000 IPv6 instances. |
| Summit X770 Series Switches | |
| xos0068355 | When scaling to 131,000 VLAN/port combinations,
rebooting Summit X770 series switches produces the following error
message:<Crit:vlan.err.criticalInfo> Critical Info: Forcing vlan mgr to READY - Timer expired - 1 appls did not send load cfg ack.s |
| xos0054125 | On Summit X770 series switches, VRRP traffic convergence time is greater (10 seconds) after rebooting the VRRP master compared to Summit X460 and X670 (3 seconds) series switches. The switch has VRRP and OSPF for forwarding traffic to uplink. |
| xos0069068 |
On Summit X770 and X670-G2 series switches with more than 32 IPv4 ARPs, IP unicast traffic destined to IPv4 addresses residing on port number 64 or higher may not be forwarded. Workaround: Either use port
numbers 1–63, or execute configure
forwarding internal-tables more l2, and then save and reboot.
|
| ExtremeSwitching X870 Series Switches | |
| xos0065687 | ExtremeXOS allows configuration of slot type to models that cannot be stacked with the ExtremeSwitching X870 series switches. |
| Summit X450-G2 | |
| xos0066473 | FDB manager is not responding to Configuration manager in presence of a loop in the network. |
| ExtremeSwitching X620 Series Switches | |
| xos0068344 |
On ExtremeSwitching X620 series switches, fiber combo ports do not come up when the preferred medium copper link is down. Workaround: If the preferred medium is fiber, fiber
combo port comes up when copper link is down.
|
| SummitStack | |
| xos0067789 | On Summit X460-G2 homogeneous stacks with greater than five BGP peer groups, some IPv6 neighbors are not established after a save and reboot of the stack. |
| xos0068388 | Unable to query objects in the extremeStackMemberTable and extremeStackingPortTable. |
| xos0068431 | On a stack with port partitioning, after rebooting
standby node, the following spurious error messages occasionally
appear:<Erro:HAL.VXLAN.InitSlotFail> Slot-1: Fail to initialize slot 5: Set VXLAN default tunnel enable to 0 failed: Conduit failure <Erro:HAL.VXLAN.InitSlotFail> Slot-1: Fail to initialize slot 5: Set VXLAN termination to 1 failed: Conduit failure <Erro:HAL.VXLAN.InitSlotFail> Slot-1: Fail to initialize slot 5: Set VXLAN tunnel based VNID to 0 failed: Conduit failure <Erro:HAL.VXLAN.InitSlotFail> Slot-1: Fail to initialize slot 5: Set VXLAN default tunnel enable to 0 failed: Conduit failure <Erro:HAL.Port.Error> Slot-1: Failed to configure static mac move behavior on port 5:51. The operation of the stack/switches is not affected. |
| xos0068597 | If you partition the ports on a Summit X770 stack, save
the configuration, and then reboot, occasionally the following spurious error
messages appear while the switch is shutting
down:<Erro:HAL.Card.Error> Slot-2: ledLinkscanSummitV3: failed to resolve LED port for unit 0 port 2, rv=-34 <Erro:HAL.Card.Error> Slot-2: ledLinkscanSummitV3: failed to resolve LED port for unit 0 port 3, rv=-34 The operation of the stack/switches is not affected. |
| xos0068396 | After removing a fan, fan status appears as "Operational at 0 RPM". If switch is then rebooted, the status changes to "Failed at 0 RPM". |
| xos0068592, xos0067767 | In ExtremeSwitching X690 stacks, receive packet counter is incrementing at two to three times over the correct rate causing incorrect statistics information to appear for MPLS and VXLAN. |
| xos0058419 |
After rebooting a stack, error messages similar to the following appear for ports
belonging to LAGs:
<Erro:cm.sys.actionErr> Slot-2: Error while loading "ports": Speed change is not allowed on port 2:6 as it is a trunk member port. |
| xos0060893 | On SummitStacks, master and backup nodes display PSU detail incorrectly of each other as 200W, instead of the actual value. |
| General | |
| xos0067478 | Ethernet OAM link fault management is only supported on non-combo fiber ports. Only 1G and 10G are supported. However, 40 and 100G ports are allowing OAM to be configured on them. |
| xos0056433 | In a stack with Summit X770 series switches as the master and backup with broadcast traffic running from master/backup to standby slots, standby slots may go to failed state when a stack failover occurs. |
| xos0060490 | The SMON MIB (RFC 2613) which was used to configure mirroring using SNMP is not available in ExtremeXOS. |
| xos0066067 | When a configuration file (.xsf) saved in ExtremeXOS 21.1 is run on ExtremeXOS 22.1, all the configured MAC lists, primary and secondary shared-secrets in RADIUS and TACAS, and NMS shared-secrets are replaced with empty strings. |
| xos0067659 | After clearing license info, enabling Direct Attach Feature Pack, and then
enabling Core license, crash occurs. Workaround: Reboot
after clearing license info, and apply Core license before applying Direct Attach
Feature Pack.
|
| xos0068902 | On ExtremeSwitching X440-G2, X620, Summit X46-0G2 series switches, and SummitStack,configure port preferred medium copper on a disabled combo port makes its peer end active. |
| xos0069087 |
Operational diagnostics sometimes do not run on some ExtremeSwitching X870 and
X690 series switches. When the switch is rebooted to run operational diagnostics,
the diagnostics do not run due to a timeout of the hardware watchdog timer, and
the switch reboots and starts ExtremeXOS instead.
Workaround:
|
| BFD | |
| xos0058364 | BFD and ELSM sessions go down momentarily while installing image or executing show tech. |
| BGP | |
| xos0067703 | In VPN-VRF, some ECMP routes are missing in route table. Issue does not occur in default or user VR. |
| xos0067519 | BGP graceful restart is not supported on the MBGP/VXLAN feature. |
| xos0067757 | After disabling BGP, and then executing disable bgp neighbor all, switch becomes unresponsive. |
| Clocking | |
| xos0068341 | In ExtremeXOS 22.2 and 22.3, only 12 slave clocks may be configured when configuring 10 PTP master clocks. Previous releases allowed 40 slave clocks. |
| DHCP/BOOTP | |
| xos0054230 | CLI counters are not available for DHCPv6 relay agent remote ID option feature. |
| Identity Management | |
| xos0066783 | Using IDMgr, UPM, and Policy combination makes client not reachable untill IDMgr settles down. |
| MLAG | |
| xos0063247 | Duplicate packets occur during MLAG link recovery when LACP is used for load sharing. |
| MPLS | |
| xos0059159 | An error message (EMS) is required when H-VPLS is down due to no pseudowire status support. |
| xos0061374 | With an L2VPN session between two Label Edge Routers (LERs), broadcast packets egressing the LERs are corrupted. |
| xos0066778 | With a LAG port configured on a VPLS service VLAN/VMAN, changing the tag of the service VLAN/VMAN results in traffic loss. |
| xos0067616 | Kernel Ref count messages occur when changing pseudowire path in VPLS. |
| Optics | |
| xos0068874 | On ExtremeSwitching X620 series switches, when using optic "SPG-DR-LX-IDFC-EX" from Source Photonics, if auto-negotiation is turned "on" there is a traffic issue. |
| xos0067870 |
Removing or inserting optics, system initialization, or peer rebooting, will possibly result in extra link events being logged. |
| xos0068910 | On ExtremeSwitching X620 and X440-G2 series switches, when using optic "RSPX0SER-ST5" from OE SOLUTIONS, link flap occurs while rebooting and removing/inserting optic. |
| OSPF | |
| xos0067974 |
With IGMP snooping disabled, IP DAD on, and IPv6 addresses configured on VLANs, after deleting, and then adding back ports, OSPF is not established. Workaround: Do any of the
following:
|
| OVSDB | |
| xos0064656 |
port_fault_status should be set in the hardware_vtep schema. In ExtremeXOS 22.1 for port_fault_status (PSTAG-like scenarios), EMS receives a log message. |
| xos0064657 |
port_fault_status should be set in the hardware_vtep schema. In ExtremeXOS 22.1 for port_fault_status (PSTAG-like scenarios), EMS receives a log message. |
| xos0065008 |
OVSDB schema is not cleared when changing configuration and rebooting. Workaround: In setups where OVSDB is used to manage
and configure the ExtremeXOS switch as a VTEP, the VTEP schema database must be
manually cleared when the active configuration of an ExtremeXOS switch is changed
using the CLI command use configuration. Failure to do so might
result in OVSDB using a stale VTEP schema database.
|
| xos0065504 | The command show ovsdb schema can become unresponsive when either ExtremeXOS or OVSDB are synchronizing data. |
| Policy | |
| xos0067946 |
Authentication override does not work after failing over from
master on a stack, and then restarting process netlogin.
Workaround: Disable, and then re-enable
policy.
|
| Security | |
| xos0067219 | With certificate signing request (CSR), even when private key and certificate match, HTTP access is not granted. |
| SNMP | |
| xos0062882 | Whole MIB compilation gets stuck at EXTREME-V2-TRAP MIB. |
| xos0068767 | Trap receiver configuration is not saved in ExtremeXOS when configured from Extreme Management Center. |
| SSH | |
| xos0063343 | Using SFTP to transfer a file with a name that is greater than 92 characters causes a Signal 6 crash. |
| VXLAN | |
| xos0067755 | When transit node is VTEP, metering does not work as expected. |
| xos0066776 | On SummitStacks, L2 broadcast traffic sent to 2,048 remote VTEPs is dropped after failover. |
Print
this page
Email this topic
Feedback
View PDF
Download EPUB