FAQ
Here's a status update on the git conversion, as well as a call for some help
mainly in testing.

After testing a bunch of tools, I've found that using cvs2git is by far the
best option when keeping keywords. It's the one that gives only the issues
that I posted about a couple of days ago.

So I've proceeded based off this one to something that would be how we create
the real repository once we go. This means I've scripted the removal of the
$PostgreSQL$ tags from the tip of the active branches as one big commit after
the migration.

I've also set up the git server and the scripts around it, that we can
eventually use. This includes commit email sending, commit policy enforcement
(no merge commits, correct author/committer tag etc) and proper access control
(a modified version of the one on git.postgresql.org - since we definitely
don't want any external dependencies for the main repository).

This is all available for testing now.

Marc has set up a mailinglist at pgsql-committers-test@postgresql.org where
commit messages from the new system is sent. If you care about what they look
like, subscribe there and wait for one to show up :-) Subscription is done
the usual way.

Anonymous users can view the repository at git.postgresql.org using gitweb
or the git:// protocol, under the name postgresql-migration.
DISCLAIMER: DO NOT BASE ANY WORK OFF THIS REPOSITORY. IT *WILL* BE RECREATED
SEVERAL TIMES AND MAY CHANGE COMPLETELY!

Existing committers have been set up to access the new repository at
ssh://git@gitmaster.postgresql.org/postgresql.git.
Robert Haas has written some instructoins for how to use this - please read
and review those.

And in general, a call to committers: please test this! Now is the time, not
after we've migrated ;) Just throw in some random commits, and some non-random
ones, both to get yourself familiar with the workflow and to iron out the bugs
in the scripts (I'm sure they're there).


For those interested in what's done, the scripts running this are all up on
github at http://github.com/mhagander/pg_githooks. The root contains the
scripts for commit messages, policy enforcement and access control.

There's also a temporary directory called "migration" that contains the
scripts and configuration files that are in use for the version of the
repository that is up there now. There's some minor plumbing around these
that isn't up there yet, but in general it is all that's used. And note that
if you want to play with it, the script uses around 8Gb of temp disk space
when running, so make sure you have enough space if you do it in a VM...

Search Discussions

  • Heikki Linnakangas at Aug 11, 2010 at 9:10 pm

    On 21/07/10 23:40, Magnus Hagander wrote:
    I've also set up the git server and the scripts around it, that we can
    eventually use. This includes commit email sending, commit policy enforcement
    (no merge commits, correct author/committer tag etc) and proper access control
    (a modified version of the one on git.postgresql.org - since we definitely
    don't want any external dependencies for the main repository).

    This is all available for testing now.
    Seems to work great. I've got a local clone of that set up now, with
    workdirs for backbranches (per the "Committing Using a Single Clone and
    multiple workdirs" instructions I added to the Committing_with_Git wiki
    page).

    I also tested with a bunch of funny commits, including merge commits,
    non-fast-forward commits and a commit with bogus Committer, and the
    post-commit hook correctly rejected all those.
    Marc has set up a mailinglist at pgsql-committers-test@postgresql.org where
    commit messages from the new system is sent. If you care about what they look
    like, subscribe there and wait for one to show up :-) Subscription is done
    the usual way.
    My posts to that lists are been stalled, but looking at the commit mail
    in the "stalled post" reply it looks OK.

    Good work! Thanks for spending the time on this, I very much prefer
    working with git nowadays.

    --
    Heikki Linnakangas
    EnterpriseDB http://www.enterprisedb.com
  • Magnus Hagander at Aug 16, 2010 at 8:13 am

    On Wed, Aug 11, 2010 at 23:10, Heikki Linnakangas wrote:
    On 21/07/10 23:40, Magnus Hagander wrote:

    I've also set up the git server and the scripts around it, that we can
    eventually use. This includes commit email sending, commit policy
    enforcement
    (no merge commits, correct author/committer tag etc) and proper access
    control
    (a modified version of the one on git.postgresql.org - since we definitely
    don't want any external dependencies for the main repository).

    This is all available for testing now.
    Seems to work great. I've got a local clone of that set up now, with
    workdirs for backbranches (per the "Committing Using a Single Clone and
    multiple workdirs" instructions I added to the Committing_with_Git wiki
    page).

    I also tested with a bunch of funny commits, including merge commits,
    non-fast-forward commits and a commit with bogus Committer, and the
    post-commit hook correctly rejected all those.
    Great, thanks for testing that. I assume this means that the commits
    that *should* go through did so? ;)

    Marc has set up a mailinglist at pgsql-committers-test@postgresql.org
    where
    commit messages from the new system is sent. If you care about what they
    look
    like, subscribe there and wait for one to show up :-) Subscription is done
    the usual way.
    My posts to that lists are been stalled, but looking at the commit mail in
    the "stalled post" reply it looks OK.
    Hmm. Were you subscribed to the list? With the same address as your
    commits are coming in from?

    If not, that would explain it. If you did, I need to figure out why ;)
  • Tom Lane at Aug 27, 2010 at 6:30 pm

    Magnus Hagander writes:
    ... This is all available for testing now.
    Marc has set up a mailinglist at pgsql-committers-test@postgresql.org where
    commit messages from the new system is sent. If you care about what they look
    like, subscribe there and wait for one to show up :-) Subscription is done
    the usual way.
    Hm, is the pgsql-committers-test thing actually working? I did a test
    push to ssh://git@gitmaster.postgresql.org/postgresql.git, and I haven't
    seen any resulting email.

    regards, tom lane
  • Magnus Hagander at Aug 27, 2010 at 6:53 pm

    On Fri, Aug 27, 2010 at 20:29, Tom Lane wrote:
    Magnus Hagander <magnus@hagander.net> writes:
    ... This is all available for testing now.
    Marc has set up a mailinglist at pgsql-committers-test@postgresql.org where
    commit messages from the new system is sent. If you care about what they look
    like, subscribe there and wait for one to show up :-) Subscription is done
    the usual way.
    Hm, is the pgsql-committers-test thing actually working?  I did a test
    push to ssh://git@gitmaster.postgresql.org/postgresql.git, and I haven't
    seen any resulting email.
    Um, it seems the list is broken somehow. I see an attempt to deliver
    them, but then:

    2010-08-27 18:04:57 1Op3IN-00084v-KZ **
    pgsql-committers-test@postgresql.org R=dnslookup T=remote_smtp: SMTP
    error from remote mail server after RCPT
    TO:<pgsql-committers-test@postgresql.org>: host mx1.hub.org
    [200.46.208.106]: 550 5.1.1 <pgsql-committers-test@postgresql.org>:
    Recipient address rejected: User unknown in relay recipient table


    Marc?
  • Marc G. Fournier at Aug 27, 2010 at 7:04 pm
    looking into it ...
    On Fri, 27 Aug 2010, Magnus Hagander wrote:
    On Fri, Aug 27, 2010 at 20:29, Tom Lane wrote:
    Magnus Hagander <magnus@hagander.net> writes:
    ... This is all available for testing now.
    Marc has set up a mailinglist at pgsql-committers-test@postgresql.org where
    commit messages from the new system is sent. If you care about what they look
    like, subscribe there and wait for one to show up :-) Subscription is done
    the usual way.
    Hm, is the pgsql-committers-test thing actually working?  I did a test
    push to ssh://git@gitmaster.postgresql.org/postgresql.git, and I haven't
    seen any resulting email.
    Um, it seems the list is broken somehow. I see an attempt to deliver
    them, but then:

    2010-08-27 18:04:57 1Op3IN-00084v-KZ **
    pgsql-committers-test@postgresql.org R=dnslookup T=remote_smtp: SMTP
    error from remote mail server after RCPT
    TO:<pgsql-committers-test@postgresql.org>: host mx1.hub.org
    [200.46.208.106]: 550 5.1.1 <pgsql-committers-test@postgresql.org>:
    Recipient address rejected: User unknown in relay recipient table


    Marc?


    --
    Magnus Hagander
    Me: http://www.hagander.net/
    Work: http://www.redpill-linpro.com/

    --
    Sent via pgsql-hackers mailing list (pgsql-hackers@postgresql.org)
    To make changes to your subscription:
    http://www.postgresql.org/mailpref/pgsql-hackers
    ----
    Marc G. Fournier Hub.Org Hosting Solutions S.A.
    scrappy@hub.org http://www.hub.org

    Yahoo:yscrappy Skype: hub.org ICQ:7615664 MSN:scrappy@hub.org
  • Marc G. Fournier at Aug 27, 2010 at 7:06 pm
    should be fixed ...
    On Fri, 27 Aug 2010, Magnus Hagander wrote:
    On Fri, Aug 27, 2010 at 20:29, Tom Lane wrote:
    Magnus Hagander <magnus@hagander.net> writes:
    ... This is all available for testing now.
    Marc has set up a mailinglist at pgsql-committers-test@postgresql.org where
    commit messages from the new system is sent. If you care about what they look
    like, subscribe there and wait for one to show up :-) Subscription is done
    the usual way.
    Hm, is the pgsql-committers-test thing actually working?  I did a test
    push to ssh://git@gitmaster.postgresql.org/postgresql.git, and I haven't
    seen any resulting email.
    Um, it seems the list is broken somehow. I see an attempt to deliver
    them, but then:

    2010-08-27 18:04:57 1Op3IN-00084v-KZ **
    pgsql-committers-test@postgresql.org R=dnslookup T=remote_smtp: SMTP
    error from remote mail server after RCPT
    TO:<pgsql-committers-test@postgresql.org>: host mx1.hub.org
    [200.46.208.106]: 550 5.1.1 <pgsql-committers-test@postgresql.org>:
    Recipient address rejected: User unknown in relay recipient table


    Marc?


    --
    Magnus Hagander
    Me: http://www.hagander.net/
    Work: http://www.redpill-linpro.com/

    --
    Sent via pgsql-hackers mailing list (pgsql-hackers@postgresql.org)
    To make changes to your subscription:
    http://www.postgresql.org/mailpref/pgsql-hackers
    ----
    Marc G. Fournier Hub.Org Hosting Solutions S.A.
    scrappy@hub.org http://www.hub.org

    Yahoo:yscrappy Skype: hub.org ICQ:7615664 MSN:scrappy@hub.org
  • Tom Lane at Aug 27, 2010 at 7:27 pm

    "Marc G. Fournier" <scrappy@hub.org> writes:
    should be fixed ...
    Ah, now I got something. Thanks.

    regards, tom lane
  • Tom Lane at Aug 27, 2010 at 7:30 pm

    Magnus Hagander writes:
    On Fri, Aug 27, 2010 at 20:29, Tom Lane wrote:
    Hm, is the pgsql-committers-test thing actually working?
    Um, it seems the list is broken somehow.
    OK, the list works now, but the commitdiff links in the messages do
    not. Looks like they are pointing at the wrong repository.

    regards, tom lane
  • Magnus Hagander at Aug 27, 2010 at 7:39 pm

    On Fri, Aug 27, 2010 at 21:30, Tom Lane wrote:
    Magnus Hagander <magnus@hagander.net> writes:
    On Fri, Aug 27, 2010 at 20:29, Tom Lane wrote:
    Hm, is the pgsql-committers-test thing actually working?
    Um, it seems the list is broken somehow.
    OK, the list works now, but the commitdiff links in the messages do
    not.  Looks like they are pointing at the wrong repository.
    Well, actually, they point to the right one, but the mirroring off to
    that one is turned off. There are just too many different versions
    right now to keep that thing pointing the right way :-)

    That part has been verified and is working, so I didn't bother keeping
    it up to date.
  • Tom Lane at Aug 27, 2010 at 8:03 pm

    Magnus Hagander writes:
    On Fri, Aug 27, 2010 at 21:30, Tom Lane wrote:
    OK, the list works now, but the commitdiff links in the messages do
    not.  Looks like they are pointing at the wrong repository.
    Well, actually, they point to the right one, but the mirroring off to
    that one is turned off. There are just too many different versions
    right now to keep that thing pointing the right way :-)
    Oh, OK.

    In any case, you asked for feedback on the message format, so here is
    some: I don't like having the commitdiff link first. I think the log
    message is the most important thing and should be first. Possible
    format (based on the example I just received):


    Log Message
    -----------
    Another experimental commit.
    Undo hacking other people previously did on README.

    Branch
    ------
    REL9_0_STABLE

    Details
    -------
    http://git.postgresql.org/gitweb?p=postgresql-migration.git;a=commitdiff;h=1add27be08978f20edb67ca5854f6701f0f6ea86

    Summary
    -------
    README | 4 ----
    1 files changed, 0 insertions(+), 4 deletions(-)


    You could argue it either way about whether to put the commitdiff link
    before or after the change summary, but offhand I think before will be
    more useful. If there are a lot of files touched, people won't want
    to have to scroll to the bottom to find the link.

    regards, tom lane
  • Magnus Hagander at Aug 28, 2010 at 10:16 am

    On Fri, Aug 27, 2010 at 22:02, Tom Lane wrote:
    Magnus Hagander <magnus@hagander.net> writes:
    On Fri, Aug 27, 2010 at 21:30, Tom Lane wrote:
    OK, the list works now, but the commitdiff links in the messages do
    not.  Looks like they are pointing at the wrong repository.
    Well, actually, they point to the right one, but the mirroring off to
    that one is turned off. There are just too many different versions
    right now to keep that thing pointing the right way :-)
    Oh, OK.

    In any case, you asked for feedback on the message format, so here is
    some: I don't like having the commitdiff link first.  I think the log
    message is the most important thing and should be first.  Possible
    format (based on the example I just received):
    Ok. I see the point of that - particularly for MUAs that show a short
    preview and don't filter the link out of it.

    Should we consider actually removing the Log Message header and just
    put the message right at the start? If there's no link first, that
    would be fairly obvious, I think..


    Log Message
    -----------
    Another experimental commit.
    Undo hacking other people previously did on README.

    Branch
    ------
    REL9_0_STABLE

    Details
    -------
    http://git.postgresql.org/gitweb?p=postgresql-migration.git;a=commitdiff;h=1add27be08978f20edb67ca5854f6701f0f6ea86

    Summary
    -------
    README |    4 ----
    1 files changed, 0 insertions(+), 4 deletions(-)


    You could argue it either way about whether to put the commitdiff link
    before or after the change summary, but offhand I think before will be
    more useful.  If there are a lot of files touched, people won't want
    to have to scroll to the bottom to find the link.
    I definitely think it should stay before - that way it will almost
    always stay "above the fold" for readers. Keeps it very easy for those
    who prefer to just click that link to get the full diff right away.
  • Tom Lane at Aug 28, 2010 at 2:30 pm

    Magnus Hagander writes:
    Should we consider actually removing the Log Message header and just
    put the message right at the start? If there's no link first, that
    would be fairly obvious, I think..
    Either way (header or not) is OK by me.

    regards, tom lane
  • Magnus Hagander at Aug 31, 2010 at 11:21 am

    On Sat, Aug 28, 2010 at 16:30, Tom Lane wrote:
    Magnus Hagander <magnus@hagander.net> writes:
    Should we consider actually removing the Log Message header and just
    put the message right at the start? If there's no link first, that
    would be fairly obvious, I think..
    Either way (header or not) is OK by me.
    Updated, with the removal of the header to help MUA previews.

Related Discussions

Discussion Navigation
viewthread | post
Discussion Overview
grouppgsql-hackers @
categoriespostgresql
postedJul 21, '10 at 8:40p
activeAug 31, '10 at 11:21a
posts14
users4
websitepostgresql.org...
irc#postgresql

People

Translate

site design / logo © 2021 Grokbase