[LU-14742] socklnd: detect link down and mark ni as unusable

• Type: Improvement
• Resolution: Fixed
• Priority: Minor
• Fix Version/s: Lustre 2.15.0
• Labels: ksocklnd lnet lnet-health
• Assignee: Serguei Smirnov
• Reporter: Serguei Smirnov

Description

To help avoid selecting lnet ni which corresponds to a downed ethernet link for sending, add a mechanism for detecting link events in socklnd. On link up/down events, find corresponding ni and set/clear ni_fatal_error_on flag, similar to how o2iblnd does it.