Movatterモバイル変換

[0]ホーム
URL:

Skip to content

Navigation Menu

Sign in
Appearance settings

Search code, repositories, users, issues, pull requests...

Provide feedback

We read every piece of feedback, and take your input very seriously.

Saved searches

Use saved searches to filter your results more quickly

 Sign up
Appearance settings

Commit530050d

Browse files
committed
Add code comment explaining ins_since_vacuum and aborted inserts
Sami complained that there's a discrepancy between n_mod_since_analyzeand n_ins_since_vacuum, as the former only accounts for committed changesand the latter tracks committed and aborted inserts. Nobody seemedoverly concerned that this would cause any concerning issues. Therepercussions, from what I can tell, are limited to causing anautovacuum to trigger for inserts sooner than it otherwise might. Fortypical ratios of commits to aborts, it's unlikely to ever be noticed.Fixing things to make it so n_ins_since_vacuum only displays committedinserts would require an additional field in PgStat_TableCounts, whichdoes not quite seem worthwhile at this stage. This commit just adds acomment with some details to mention that we know about it, which willhopefully prevent repeat discussions.Reported-by: Sami Imseih <samimseih@gmail.com>Author: David Rowley <drowleyml@gmail.com>Reviewed-by: Sami Imseih <samimseih@gmail.com>Discussion:https://postgr.es/m/CAApHDvpgV3a-R2EGmPOh0L-x3pHbZpM3y4dySWfy+UqUazwDQA@mail.gmail.com
1 parent39729ec commit530050d

File tree

1 file changed

+9
-0
lines changed

1 file changed

+9
-0
lines changed

‎src/backend/utils/activity/pgstat_relation.c‎

Lines changed: 9 additions & 0 deletions
Original file line numberDiff line numberDiff line change
@@ -868,7 +868,16 @@ pgstat_relation_flush_cb(PgStat_EntryRef *entry_ref, bool nowait)
868868
tabentry->live_tuples+=lstats->counts.delta_live_tuples;
869869
tabentry->dead_tuples+=lstats->counts.delta_dead_tuples;
870870
tabentry->mod_since_analyze+=lstats->counts.changed_tuples;
871+
872+
/*
873+
* Using tuples_inserted to update ins_since_vacuum does mean that we'll
874+
* track aborted inserts too. This isn't ideal, but otherwise probably
875+
* not worth adding an extra field for. It may just amount to autovacuums
876+
* triggering for inserts more often than they maybe should, which is
877+
* probably not going to be common enough to be too concerned about here.
878+
*/
871879
tabentry->ins_since_vacuum+=lstats->counts.tuples_inserted;
880+
872881
tabentry->blocks_fetched+=lstats->counts.blocks_fetched;
873882
tabentry->blocks_hit+=lstats->counts.blocks_hit;
874883

0 commit comments

Comments
 (0)
[8]ページ先頭
©2009-2025 Movatter.jp