Has anyone seen the issue of Fleet unable to connect to the API because it tries too fast? I confirmed via a curl test that my pods take a whole second before the API proxy (or networking) is available. Fleet should try a few times before failing, right? The failure from the fleet-migrate-gitrepo-helm-url-regex job (which is required to complete before Fleet is good):

time="2026-06-10T17:50:12Z" level=fatal msg="checking helm URL regex migration marker: failed to get server groups: Get \"<https://10.43.0.1:443/api>\": dial tcp 10.43.0.1:443: connect: connection refused"

Confirmed API proxy (or pod networking) slowness via another curl test (401 is good in this case):

checking kube api attempt 0
curl: (7) Failed to connect to 10.43.0.1 port 443 after 0 ms: Could not connect to server
checking kube api attempt 1
curl: (7) Failed to connect to 10.43.0.1 port 443 after 0 ms: Could not connect to server
checking kube api attempt 2
curl: (22) The requested URL returned error: 401

I also confirmed that the actual command "fleet migrate migrate gitrepo-helm-url-regex" works perfectly fine if I run in the same pod but just wait a few seconds. Having this command as the entrypoint is too fast for the API to be available.