High latency reported in MetroCluster IP after backend port flapping
Applies to
- MetroCluster IP
- T6 Card X1146A/X91146A
Issue
- Interface flapping reported in event log:
Thu Feb 22 08:00:16 +0200 [node-02: intr: netif.linkDown:info]: Ethernet e1b: Link down, check cable.
Thu Feb 22 08:00:18 +0200 [node-02: intr: netif.linkUp:info]: Ethernet e1b: Link up. - High frontend latency reported
- Command aborts and high latency to remote disks reported:
Fri Feb 22 10:05:16 +0200 [node-02: disk_latency_monitor: shm.threshold.ioLatency:debug]: Disk 0m.i2.1L38 has exceeded the expected IO latency in the current window with average latency of 60 msecs and average utilization of 36 percent. Highest average IO latency: 0m.i2.1L19: 237 msecs
2/22/2024 15:07:30 node-02 ERROR scsi.cmd.abortedByHost: Disk device 0m.i2.0L145: Command aborted by host adapter: HA status 0x13: cdb 0x28:ec234d10:0008. - Issue just reported on local, DR and AUX node, no issues reported on partner
- Toggling Port fixed the issue temporary
- Replacing SFPs and cable doesn't fix the issue
