Repository navigation
Provider status checks fail on weaker machines #13326
Description
Provider status checks time out on slower machines. The version and status probes use 3-4s caps, and cold CLIs on Windows regularly need more than that. The provider then shows Unavailable even though the CLI works when run by hand. Raise the status probe timeouts to 10s so these machines get a stable reading.
Activity
juliusmarminge commentedon Sep 24, 2026
Duplicate of #11080 — same root cause: hardcoded ~3–4s version/status probe budgets flip healthy CLIs (especially cold Windows / weaker hosts) to Unavailable. That issue is already accepted; triage there recommends raising the probe timeout toward ~8–10s and avoiding a single timeout flipping a ready provider to error.
Closing this as a duplicate so the fix tracks on #11080.
juliusmarminge commentedon Sep 24, 2026
Closed as duplicate of #11080.