My concern with performance is not due to subversion its self. The
server is currently hosting part of a high traffic website streaming
audio and video. The owners want to move the remainder of the site
(all pages) to the same server, I have warned them that I don't think
this is a good idea as I am not sure if the server will cope too well.
This is why I want to keep subversion stress on the server to a
minimum. When you perform a subversion update and there have been no
changes I guess all subversion does it check the version number of
the repos against the working copy version number? Would this not be
about the same checking a text file?
I guess I was thinking it would check all the files for some
reason. Can someone confirm this?
On 19 May 2006, at 04:19, Alex Holst wrote:
> Quoting Jamie (jam5t3r@gmail.com):
>> I was thinking of writing a c program like this http://
>> subversion.tigris.org/faq.html#website-auto-update and running it
>> every 10 minutes with a cron job, but I am not sure if this would be
>> too much going on? I don't know how much resources running an update
>> on 3 repositories every 10 minutes would take.
>
> That depends entirely on the size of the previous check-ins. Any
> Pentium
> III with 1GB RAM or more and decent SCSI disks can do what you need
> without breaking a sweat.
>
> If you are so concerned about performance, have a post-commit hook
> write
> a semaphore file to /tmp/name-of-repo and from your 10 minute cron
> job,
> check if that file exists, before you run svn update. That way, svn
> update is only run when there have actually been updates to your
> repository.
>
> --
> I prefer the dark of the night, after midnight and before four-thirty,
> when it's more bare, more hollow. http://a.mongers.org
---------------------------------------------------------------------
To unsubscribe, e-mail: users-unsubscribe@subversion.tigris.org
For additional commands, e-mail: users-help@subversion.tigris.org
Received on Fri May 19 20:19:28 2006