[LU-17735] Sanity sanity test 7b spurious failures due to check_node_health not waiting long enough for network health

• Type: Bug
• Resolution: Fixed
• Priority: Minor
• Fix Version/s: Lustre 2.16.0
• Labels: msft
• Severity: 3
• Assignee: Ellis Wilson
• Reporter: Ellis Wilson

Description

check_node_health only attempts pings for 5 seconds prior to failing out with "Network not available!".  This may well work for a static, stable set of hosts used for testing on-premise, but our auster test farm is created on-demand each and every time a suite of tests is fired off.  The likelihood of hitting a transient networking issue that self-resolves quickly (but longer than 5 seconds) is accordingly higher.

This just bumps that check from 5 seconds to a minute.  It shouldn't impact auster testing for anybody on-prem in any material way, but has eliminated transient failures for us.

Patch to be sent shortly.