Re: Add -k/--link option to pg_combinebackup - Mailing list pgsql-hackers

From Israel Barth Rubio
Subject Re: Add -k/--link option to pg_combinebackup
Date
Msg-id CAO_rXXBNuVUMnMni-vJ5aafoqsmqy_QMXuSbog7KRD1KO7ZCdw@mail.gmail.com
Whole thread Raw
In response to Re: Add -k/--link option to pg_combinebackup  (Robert Haas <robertmhaas@gmail.com>)
List pgsql-hackers
Thanks for the review, Robert!
I applied all of your suggestions.

> I'm still not a fan of the changes to 010_links.pl; let's drop those.

As discussed through a side chat, 010_links.pl is the new test file
which ensures the hard links are created as expected in the output
directory.

I've removed the changes that the v2 patch had in the other test files,
which were the ones you were suggesting to drop here.

> cfbot is fairly unhappy with your patch. See
> http://cfbot.cputube.org/israel-barth.html or
> https://cirrus-ci.com/github/postgresql-cfbot/postgresql/cf%2F5508 --
> I haven't looked into what is going wrong here, and there may well be
> more than one thing, but nothing can be committed until this is all
> green.

There are some failures on compiler warnings and Windows jobs that
seem unrelated with this patch.

Besides that, there were failures when executing `010_links.pl` in the
Linux autoconf jobs. As discussed in a side chat with Euler and Robert,
and pointed out by Robert, Cirrus CI uses a very small
`--with-segsize-blocks`. My tests in the V2 patch were assuming 1GB
segments, so they were failing in Cirrus CI.

The patch V3, which I'm attaching in this reply, implements some logic
to handle any segment size.

Best regards,
Israel.
Attachment

pgsql-hackers by date:

Previous
From: Melanie Plageman
Date:
Subject: Re: BitmapHeapScan streaming read user and prelim refactoring
Next
From: Daniel Gustafsson
Date:
Subject: Re: Adding support for SSLKEYLOGFILE in the frontend