forked frompostgres/postgres
- Notifications
You must be signed in to change notification settings - Fork6
Commit4016bde
committed
Fix assorted bugs in GIN's WAL replay logic.
The original coding was quite sloppy about handling the case whereXLogReadBuffer fails (because the page has since been deleted). Thiswould result in either "bad buffer id: 0" or an Assert failure duringreplay, if indeed the page were no longer there. In a couple of placesit also neglected to check whether the change had already been applied,which would probably result in corrupted index contents. I believe thatbug #5703 is an instance of the first problem. These issues could show upwithout replication, but only if you were unfortunate enough to crashbetween 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 parent220e45b commit4016bde
1 file changed
+181
-127
lines changed0 commit comments
Comments
(0)