Improve baseline resync logging - #455
Conversation
|
Codecov Report❌ Patch coverage is
Additional details and impacted files@@ Coverage Diff @@
## stable/v4.x #455 +/- ##
==============================================
Coverage ? 55.02%
==============================================
Files ? 36
Lines ? 5350
Branches ? 676
==============================================
Hits ? 2944
Misses ? 2110
Partials ? 296 ☔ View full report in Codecov by Harness. 🚀 New features to boost your workflow:
|
Eliminate inconsistent terminology, context, and severity levels across existing BR read/write logs. Add structured lifecycle, payload, retry, and failure records with consistent PG, shard, batch, blob, and block identifiers. Keep key object and payload transitions at INFO while moving per-blob details to TRACE to reduce routine log noise.
|
only log changes in this PR, no logic changes, right? |
| return std::make_error_code(std::errc::io_error); | ||
| } | ||
| // Add local blob info to index & PG | ||
| bool success = | ||
| home_obj_.local_add_blob_info(ctx_->pg_id, BlobInfo{ctx_->shard_cursor, blob_id, blk_id}); | ||
| if (!success) { | ||
| LOGE("Failed to add blob info for blob_id={}", blob_id); | ||
| LOGE("Failed to index resync blob: pg={}, shard_id=0x{:x}, blob={}, blkid={}", ctx_->pg_id, | ||
| ctx_->shard_cursor, blob_id, blk_id.to_string()); | ||
| homestore::data_service().async_free_blk(blk_id).get(); | ||
| return err; | ||
| return std::make_error_code(std::errc::io_error); |
There was a problem hiding this comment.
@JacksonYao287 All the changes are solely log optimizations, except a minor bugfix here detected during the log optimization, regarding the async return value.
|
LGTM. After this PR is merged, what log mode do you expect? IIRC production |
Let's keep the production log level at 'info' as the current SM sherlock collection is already overwhelmed. For SH we can elevate the level to |
4 participants
Eliminate inconsistent terminology, context, and severity levels across existing BR read/write logs. Add structured lifecycle, payload, retry, and failure records with consistent PG, shard, batch, blob, and block identifiers. Keep key object and payload transitions at INFO while moving per-blob details to TRACE to reduce routine log noise.