k8snetworkplumbingwg / multus-cni / 28328544746

50%

master: 50%

Pull #1519

github

shim: use single-shot readiness check for CNI STATUS

STATUS is itself a health check -- if the daemon isn't responding, fail
immediately instead of polling for 60 seconds. CmdDel already uses
CheckAPIReadyNow for the same reason.

WaitUntilAPIReady polls /healthz every 100ms for up to 60s, which adds
an unnecessary HTTP round-trip on every STATUS call and causes long
stalls when the daemon is down. Since CRI-O calls STATUS periodically,
it will naturally retry on the next cycle.

Signed-off-by: Igal Tsoiref <itsoiref@redhat.com>
Co-authored-by: Cursor <cursoragent@cursor.com>

0 of 1 new or added line in 1 file covered. (0.0%)

2528 of 5031 relevant lines covered (50.25%)

6.35 hits per line