- Notifications
You must be signed in to change notification settings - Fork28
Commit4482c6a
committed
BRIN revmap pages are not standard pages ...
... and therefore we ought not to tell XLogRegisterBuffer the opposite,when writing XLog for a brin update that moves the index tuple to adifferent page. Otherwise, xlog insertion would try to "compress thehole" when producing a full-page image for it; but since we don't updatepd_lower/upper, the hole covers the whole page. On WAL replay, therevmap page becomes empty and so the entire portion of the index isuseless and needs to be recomputed.This is low-probability: a BRIN update only moves an index tuple to adifferent page when the summary tuple is larger than the existing one,which doesn't happen with fixed-width datatypes. Also, the revmappage must be first after a checkpoint.Report and patch: Kuntal GhoshBug is alleged to have detected by a WAL-consistency-checking tool.Discussion:https://postgr.es/m/CAGz5QCJ=00UQjScSEFbV=0qO5ShTZB9WWz_Fm7+Wd83zPs9Geg@mail.gmail.comI posted a test case demonstrating the problem, but I'm refraining fromadding it to the test suite; if the WAL consistency tool makes it in,that will be a better way to catch this from regressing. (We shoulddefinitely have someting that causes not-same-page updates, though.)1 parent8b1bf31 commit4482c6a
1 file changed
+1
-1
lines changedLines changed: 1 addition & 1 deletion
| Original file line number | Diff line number | Diff line change | |
|---|---|---|---|
| |||
289 | 289 | | |
290 | 290 | | |
291 | 291 | | |
292 | | - | |
| 292 | + | |
293 | 293 | | |
294 | 294 | | |
295 | 295 | | |
| |||
0 commit comments
Comments
(0)