Re: More issues with pg_verify_checksums and checksum verification in base backups

Hi,

first off, a bit of a meta-question: Did the whitelist approach die
completely, or are we going to tackle it again for v13 or later?

On Fri, Nov 30, 2018 at 10:44:33AM +0900, Michael Paquier wrote:
> On Thu, Nov 29, 2018 at 11:03:54AM +0900, Michael Paquier wrote:
> > Thanks David for the input.  I think that I will be able to finish
> > wrapping and commit this stuff tomorrow.
> 
> And done.  I kept the split into two commits for clarity as suggested by
> Stephen, as bisect would actually complain only when using EXEC_BACKEND,
> and only for the test suite of pg_verify_checksums.

So if I do 'touch $PGDATA/global/123.', current pg_checksums
errors out with

|pg_checksums: error: invalid segment number 0 in file name
|"<<$PGDATA>>/global/123."

This is something I still have in the test suite of my pg_checksums
fork, cause I never reverted that one from isRelFile() back to
skipfile() (so it doesn't fail on the above cause `123.' is not
considered a relation file worth checksumming).

Independently of the whitelist/blacklist question, I believe
pg_checksums should not error out as soon as it encounters a weird looking
file, but either (i) still checksum it or (ii) skip it? Or is that to be
considered a pilot error and it's fine for pg_checksums to fold?

Thoughts?


Michael

-- 
Michael Banck
Projektleiter / Senior Berater
Tel.: +49 2166 9901-171
Fax:  +49 2166 9901-100
Email: michael.banck@credativ.de

credativ GmbH, HRB Mönchengladbach 12080
USt-ID-Nummer: DE204566209
Trompeterallee 108, 41189 Mönchengladbach
Geschäftsführung: Dr. Michael Meskes, Jörg Folz, Sascha Heuer

Unser Umgang mit personenbezogenen Daten unterliegt
folgenden Bestimmungen: https://www.credativ.de/datenschutz