Re: backup manifests and contemporaneous buildfarm failures
Tom Lane <tgl@sss.pgh.pa.us>
Commits
GET /api/v1/messages/:b64id/commits
the thread's linked commits as JSON, with link sources.
API reference →
-
Try to avoid compiler warnings in optimized builds.
- 05021a2c0cd2 13.0 landed
-
Fix option related issues in pg_verifybackup.
- 0a89e93bfaa6 13.0 landed
-
Add index term for backup manifest in documentation.
- 4db819ba4039 13.0 landed
-
Code review for backup manifest.
- a2ac73e7be7a 13.0 landed
-
Document the backup manifest file format.
- 149f2ae88ab0 13.0 landed
-
Fix typo in pg_validatebackup documentation.
- c4f82a779d26 13.0 landed
-
Exclude backup_manifest file that existed in database, from BASE_BACKUP.
- 1ec50a81ec0a 13.0 landed
-
Msys2 tweaks for pg_validatebackup corruption test
- c3e4cbaab936 13.0 landed
-
Fix resource management bug with replication=database.
- 3e0d80fd8d3d 13.0 cited
-
Be more careful about time_t vs. pg_time_t in basebackup.c.
- db1531cae009 13.0 cited
-
pg_validatebackup: Fix 'make clean' to remove tmp_check.
- 9f8f881caa0f 13.0 landed
-
pg_validatebackup: Also use perl2host in TAP tests.
- 460314db08e8 13.0 landed
-
Generate backup manifests for base backups, and validate them.
- 0d8c9c1210c4 13.0 landed
-
Add checksum helper functions.
- c12e43a2e0d4 13.0 landed
-
pg_waldump: Add a --quiet option.
- ac44367efbef 13.0 landed
-
Catversion bump for b9b408c48724
- afb5465e0cfc 13.0 cited
-
pg_basebackup: Refactor code for reading COPY and tar data.
- 431ba7bebf13 13.0 landed
-
Use a ResourceOwner to track buffer pins in all cases.
- 3cb646264e8c 12.0 cited
-
Use ARMv8 CRC instructions where available.
- f044d71e331d 11.0 cited
-
Logical replication support for initial data copy
- 7c4f52409a8c 10.0 cited
-
Use Intel SSE 4.2 CRC instructions where available.
- 3dc2d62d0486 9.5.0 cited
-
Switch to CRC-32C in WAL and other places.
- 5028f22f6eb0 9.5.0 cited
-
Remove support for 64-bit CRC.
- 404bc51cde9d 9.5.0 cited
-
Change CRCs in WAL records from 64bit to 32bit for performance reasons.
- 21fda22ec46d 8.1.0 cited
Thomas Munro <thomas.munro@gmail.com> writes: > Same here, on elver. I see pg_subtrans has been chmod(0)'d, > presumably by the perl subroutine mutilate_open_directory_fails. I > see this in my inbox (the build farm wrote it to stderr or stdout > rather than the log file): > cannot chdir to child for > pgsql.build/src/bin/pg_validatebackup/tmp_check/t_003_corruption_master_data/backup/open_directory_fails/pg_subtrans: > Permission denied at ./run_build.pl line 1013. > cannot remove directory for > pgsql.build/src/bin/pg_validatebackup/tmp_check/t_003_corruption_master_data/backup/open_directory_fails: > Directory not empty at ./run_build.pl line 1013. I'm guessing that we're looking at a platform-specific difference in whether "rm -rf" fails outright on an unreadable subdirectory, or just tries to carry on by unlinking it anyway. A partial fix would be to have the test script put back normal permissions on that directory before it exits ... but any failure partway through the script would leave a time bomb requiring manual cleanup. On the whole, I'd argue that testing that behavior is not valuable enough to take risks of periodically breaking buildfarm members in a way that will require manual recovery --- to say nothing of annoying developers who trip over it. So my vote is to remove that part of the test and be satisfied with checking the behavior for an unreadable file. This doesn't directly explain the failure-at-next-configure behavior that we're seeing in the buildfarm, but it wouldn't be too surprising if it ends up being that the buildfarm client script doesn't manage to fully recover from the situation. regards, tom lane