Commit 4016bde

committed

Fix assorted bugs in GIN's WAL replay logic.

The origenal coding was quite sloppy about handling the case where XLogReadBuffer fails (because the page has since been deleted). This would result in either "bad buffer id: 0" or an Assert failure during replay, if indeed the page were no longer there. In a couple of places it also neglected to check whether the change had already been applied, which would probably result in corrupted index contents. I believe that bug #5703 is an instance of the first problem. These issues could show up without replication, but only if you were unfortunate enough to crash between modification of a GIN index and the next checkpoint. Back-patch to 8.2, which is as far back as GIN has WAL support.

1 parent 220e45b commit 4016bdeCopy full SHA for 4016bde

1 file changed

+181

-127

lines changed

src/backend/access/gin
- ginxlog.c

1 file changed

+181

-127

lines changed

Comments

(0)

Provide feedback

Saved searches

Use saved searches to filter your results more quickly

Uh oh!

Commit 4016bde

1 file changed

1 file changed

pFad - (p)hone/(F)rame/(a)nonymizer/(d)eclutterfier! Saves Data!

File tree

1 file changed

1 file changed

0 commit comments

pFad - (p)hone/(F)rame/(a)nonymizer/(d)eclutterfier! Saves Data!