disinto-admin/disinto
1
0
Fork 0
Code Issues 16 Pull requests 1 Projects Releases Packages Wiki Activity Actions

fix: bug: deploy.sh 360s still too tight for chat cold-start + cascade-skip masks edge/vault-runner (#1070) #1071

Merged
dev-bot merged 1 commit from fix/issue-1070 into main 2026-04-20 08:04:28 +00:00
Conversation 2 Commits 1 Files changed 1 +22 -2
1 changed files with 22 additions and 2 deletions
Download patch file Download diff file Expand all files Collapse all files

24
lib/init/nomad/deploy.sh
View file

@ -19,10 +19,12 @@
# JOB_READY_TIMEOUT_SECS — poll timeout in seconds (default: 360) # JOB_READY_TIMEOUT_SECS — poll timeout in seconds (default: 360)
# JOB_READY_TIMEOUT_<JOBNAME> — per-job timeout override (e.g., # JOB_READY_TIMEOUT_<JOBNAME> — per-job timeout override (e.g.,
# JOB_READY_TIMEOUT_FORGEJO=300) # JOB_READY_TIMEOUT_FORGEJO=300)
# Built-in: JOB_READY_TIMEOUT_CHAT=600
# #
# Exit codes: # Exit codes:
# 0 success (all jobs deployed and healthy, or dry-run completed) # 0 success (all jobs deployed and healthy, or dry-run completed)
# 1 failure (validation error, timeout, or nomad command failure) # 1 failure (validation error, or one or more jobs unhealthy after all
# jobs submitted — deploy does NOT cascade-skip on timeout)
# #
# Idempotency: # Idempotency:
# Running twice back-to-back on a healthy cluster is a no-op. Jobs that are # Running twice back-to-back on a healthy cluster is a no-op. Jobs that are
@ -35,7 +37,11 @@ SCRIPT_ROOT="$(cd "$(dirname "${BASH_SOURCE[0]}")" && pwd)"
REPO_ROOT="${REPO_ROOT:-$(cd "${SCRIPT_ROOT}/../../.." && pwd)}" REPO_ROOT="${REPO_ROOT:-$(cd "${SCRIPT_ROOT}/../../.." && pwd)}"
JOB_READY_TIMEOUT_SECS="${JOB_READY_TIMEOUT_SECS:-360}" JOB_READY_TIMEOUT_SECS="${JOB_READY_TIMEOUT_SECS:-360}"
# Per-job built-in defaults (override with JOB_READY_TIMEOUT_<JOBNAME> env var)
JOB_READY_TIMEOUT_CHAT="${JOB_READY_TIMEOUT_CHAT:-600}"
DRY_RUN=0 DRY_RUN=0
FAILED_JOBS=() # jobs that timed out or failed deployment
log() { printf '[deploy] %s\n' "$*" >&2; } log() { printf '[deploy] %s\n' "$*" >&2; }
die() { printf '[deploy] ERROR: %s\n' "$*" >&2; exit 1; } die() { printf '[deploy] ERROR: %s\n' "$*" >&2; exit 1; }
@ -215,7 +221,8 @@ for job_name in "${JOBS[@]}"; do
# 4. Wait for healthy state # 4. Wait for healthy state
if ! _wait_job_running "$job_name" "$job_timeout"; then if ! _wait_job_running "$job_name" "$job_timeout"; then
die "deployment for job '${job_name}' did not reach successful state" log "WARNING: deployment for job '${job_name}' did not reach successful state — continuing with remaining jobs"
FAILED_JOBS+=("$job_name")
fi fi
done done
@ -223,4 +230,17 @@ if [ "$DRY_RUN" -eq 1 ]; then
log "dry-run complete" log "dry-run complete"
fi fi
# ── Final health summary ─────────────────────────────────────────────────────
if [ "${#FAILED_JOBS[@]}" -gt 0 ]; then
log ""
log "=== DEPLOY SUMMARY ==="
log "The following jobs did NOT reach healthy state:"
for failed in "${FAILED_JOBS[@]}"; do
log " - ${failed}"
done
log "All other jobs were submitted and healthy."
log "======================"
exit 1
fi
exit 0 exit 0