Intel ice driver NETDEV WATCHDOG timeouts
Issue
- Depending on the severity of the TX timeout there may be a cluster interruption that requires intervention like node reboots.
- The following messages will appear in dmesg and journalctl:
kernel: NETDEV WATCHDOG: ens6 (ice): transmit queue 1 timed out
kernel: ice 0000:11:00.0 ens6: tx_timeout recovery level 1, txqueue 1
Environment
- Red Hat Enterprise Linux 8.6
- LACP bonding
- Intel E810 adapters
- IBM GPFS cluster
Subscriber exclusive content
A Red Hat subscription provides unlimited access to our knowledgebase, tools, and much more.
