Skip to content

chore(db): clarify full-compaction sentinel logging#238

Open
EddieHouston wants to merge 1 commit into
Blockstream:new-indexfrom
EddieHouston:fix/compaction-sentinel-logging
Open

chore(db): clarify full-compaction sentinel logging#238
EddieHouston wants to merge 1 commit into
Blockstream:new-indexfrom
EddieHouston:fix/compaction-sentinel-logging

Conversation

@EddieHouston

Copy link
Copy Markdown
Collaborator

Summary

Improve RocksDB full-compaction sentinel logging:

  • include the database identity in initial-sync compaction messages
  • move the routine F-sentinel-found message from INFO to TRACE

Motivation

Each index update checks the full-compaction sentinel for txstore, history, and
cache. Once initial sync has completed, this produced three INFO messages for
every new block.

The initial-sync messages also did not identify which database was being processed,
making the sequential compaction of the three databases look like conflicting
sentinel or trigger transitions.

Behavior

State-changing initial-sync events remain at INFO:

  • sentinel missing
  • full compaction completed
  • triggers tightened
  • sentinel set

The steady-state “sentinel found, skipping compaction” path is now logged at
TRACE.

This is a logging-only change. Compaction, trigger, and sentinel behavior is
unchanged.

Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment

Labels

None yet

Projects

None yet

Development

Successfully merging this pull request may close these issues.

2 participants