[svn.haxx.se] · SVN Dev · SVN Users · SVN Org · TSVN Dev · TSVN Users · Subclipse Dev · Subclipse Users · this month's index

Re: svn status gets 'killed'

From: Paul Lussier <pll_at_lanminds.com>
Date: 2002-10-08 22:00:06 CEST

In a message dated: Tue, 08 Oct 2002 09:45:38 +0200
Michael Wood said:

>There have been various memory usage problems in Subversion in the past.
>0.14.1 is quite old. Do you still get the same problem with the latest
>version of the Debian package?

The problem seems to have disappeared with the svn 0.14.3-1 Debian
package. However, it still seems to take quite a long time to
complete a 'status' of a semi-large repository. Here's the 'time'
output from an 'svn status -u -v' on the Linux 2.4.19 kernel tree
just imported into svn:

        $ time svn status -u -v > /tmp/foo
        real 3m19.599s
        user 2m13.660s
        sys 0m5.260s

This kernel tree is fresh from kernel.org, and immediately after an
'svn import' and 'svn co'. So, other than the svn overhead[1], this is
pristine kernel source. 3 minutes 20 seconds seems a little long to
determine the status of an unchanged repository, though, doesn't it?
Does svn cache changes anywhere? Or does it need to check diffs
between each and every file in the repo and in the .svn area?

The Subversion Handbook states:

        "The 'svn status' command is your friend; become intimate
         with it. You'll probably use 'svn status' more than any
         other command."

If it takes 3+ minutes to tell me nothing changed, how long is going
to take to tell me when things did change[2]? At that rate, I'll never
get anything done :)

Is this similar to what others have seen?

[1] The size of the svn repo is > 3x the size of just the raw
        kernel tree:

                Checked out revision 1.
                pll@tater:/usr/local/src$ du -sh linux-2.4.19/
                479M linux-2.4.19

                pll@tater:/usr/local/src/foo$ tar jxf ../tars/linux-2.4.19.tar.bz2
                pll@tater:/usr/local/src/foo$ du -sh linux-2.4.19/
                159M linux-2.4.19

        Don't know if that matters, but it might well be why the 'svn status'
        takes so long.

[2] I'm aware svn is a major work in progress, and today's
        performance is not indicative of what things will (hopefully)
        be in the future. I was just kinda surprised that it took so
        long to establish that nothing changed.

Thanks,

-- 
Seeya,
Paul
--
	It may look like I'm just sitting here doing nothing,
   but I'm really actively waiting for all my problems to go away.
	 If you're not having fun, you're not doing it right!
---------------------------------------------------------------------
To unsubscribe, e-mail: dev-unsubscribe@subversion.tigris.org
For additional commands, e-mail: dev-help@subversion.tigris.org
Received on Tue Oct 8 22:00:54 2002

This is an archived mail posted to the Subversion Dev mailing list.

This site is subject to the Apache Privacy Policy and the Apache Public Forum Archive Policy.