Sprite wedged: exec/file/console/restart all time out, metadata reads work fine

1

Sprite acuity-baker (org christian-pugliese, id sprite-6228781b-8ccf-430e-8cc6-cf589627ee4c) has been unreachable since ~2026-08-27 13:30 UTC.

What works:

  • sprite list and GET /v1/sprites/acuity-baker respond normally — status shows “warm”
  • The dashboard loads and shows the sprite

What doesn’t (everything that touches the machine):

  • sprite exec: failed to connect: read tcp [redacted]->[2a09:8280:1::99:d7b3:0]:443: i/o timeout
  • sprite console: hangs indefinitely
  • File writes via API (from GitHub Actions CI): Put ".../fs/write...": context deadline exceeded (Client.Timeout exceeded while awaiting headers)
  • Dashboard restart button: “upstream restart failed”
  • POST /v1/sprites/acuity-baker/restart via sprite api: hangs 4+ minutes, zero bytes received

Same as topics 27093, 28230, and 28387.

This sprite has real provisioned state on it, so I’d much rather have it recovered than destroy/recreate. Can someone take a look?

3

This topic was automatically closed 10 days after the last reply. New replies are no longer allowed.