Repository navigation
Conversation
Hashed oplog ingestion ingests every bucket, so connector pipelines no longer list buckets: drop the per-bucket pipeline wait, the fixed wait for the oplog populator, and the Kafka Connect endpoint they relied on. Issue: ZENKO-5375
Hello delthas,My role is to assist you with the merge of this Available options
Available commands
Status report is not available. |
| # yq eval 'sortKeys(.)' -i deps.yaml | ||
| backbeat: | ||
| sourceRegistry: ghcr.io/scality | ||
| sourceRegistry: ghcr.io/scality/playground/delthas |
There was a problem hiding this comment.
Backbeat points at a personal playground registry and a dev build tag, not a released tag on ghcr.io/scality. This must be reverted to ghcr.io/scality with the released BB-906 tag before merge, otherwise release packaging pulls an unpublished image.
Waiting for approvalThe following approvals are needed before I can proceed with the merge:
|
972c859 to
0f513bb
Compare
| dashboard: zenko-operator/zenko-operator-dashboards | ||
| image: zenko-operator | ||
| tag: v1.9.0-preview.1 | ||
| tag: bc9f30c710c5065747be4a2927dff3864400355b |
There was a problem hiding this comment.
zenko-operator is pinned to a raw commit SHA instead of a released tag. Replace it with the released operator tag that ships listenAllBuckets (zenko-operator#656) before merge.
This bumps backbeat and moves to the new hashed oplog ingestion (that listens to all buckets), enabled through the new
listenAllBucketsZenkoVersion feature flag (https://github.com/scality/zenko-operator/pull/656). See https://github.com/scality/citadel/pull/375 scality/backbeat#2869Issue: ZENKO-5375