|From:||Michael Paquier <michael(at)paquier(dot)xyz>|
|To:||Masahiko Sawada <sawada(dot)mshk(at)gmail(dot)com>|
|Subject:||Re: Add progress reporting to pg_verifybackup|
|Views:||Raw Message | Whole Thread | Download mbox | Resend email|
On Sat, Feb 04, 2023 at 12:32:15PM +0900, Michael Paquier wrote:
> That seems rather OK seen from here. I'll see about getting that
> applied except if there is an objection of any kind.
Okay, I have looked at that again this morning and I've spotted one
tiny issue: specifying --progress with --skip-checksums does not
really make sense.
Ignoring entries with a bad size would lead to incorrect progress
report (for example, say an entry in the manifest has a largely
oversized size number), so your approach on this side is correct. The
application of the ignore list via -i is also correct, as a patch
matching with should_ignore_relpath() does not compute an extra size
I was also wondering for a few minutes while on it whether it would
have been cleaner to move the check for should_ignore_relpath()
directly in verify_file_checksum() and verify_backup_file(), but
nobody has complained about that as being a problem, either.
What do you think about the updated version attached?
|Next Message||Andrey Borodin||2023-02-06 00:44:53||Re: Amcheck verification of GiST and GIN|
|Previous Message||Andres Freund||2023-02-06 00:07:50||Re: Weird failure with latches in curculio on v15|