fix(deploy): retry ECR login+pull on transient auth failures
During the production DB migration deploy, push.sh failed twice in a row with 'Your authorization token has expired' immediately after a successful docker login -- resolved both times on retry (manual login+pull, then the next push.sh attempt) without any code change, so the root cause is still unclear (possibly ECR token propagation lag).
Rather than leave this as an unexplained flake, retry the full login+pull cycle (not just the pull) up to 3 times with a 5s backoff -- re-fetching a fresh token each attempt addresses the failure mode regardless of exact root cause.
sha256:61a62e6048762fd28090f94b5b35c327b30054563e618f5999f707fe92a56cff
sha
+1
symbols
1 changed · 1061 in snapshot
files
sha256:d4e9423eed65865b80c781bd00af9502b4d9aa6163fadb99708e39919efa73eb
snapshot
+1
symbol added
1
file changed
1061
files in snapshot
0
dead code introduced
Semantic Changes
1 symbol
+
.vscode/
+
alembic/
+
deploy/
+
docs/
+
musehub/
+
scripts/
+
src/
+
src/ts/
+
tests/
+
tools/
Files Changed
~1
1061 in snapshot
~1
1061 in snapshot
0 comments
To add a comment, use the Muse CLI:
muse hub commit comment sha256:61a62e6048762fd28090f94b5b35c327b30054563e618f5999f707fe92a56cff --body "your comment"
No comments yet. Be the first to start the discussion.