On Wed, Sep 11, 2013 at 7:19 PM, Curt Sellmer <sellmerfud_at_gmail.com> wrote:
>>>
>>> You said you dumped and loaded your repository. You also have
>>> corruption shown by apache that is not shown by svnadmin. When you
>>> dump/load are you also putting the new repository in the same place as
>>> the old repository? If so, are you restarting apache? It's perfectly
>>> acceptable to replace a repository but you must restart apache if the
>>> repository is altered but the UUID is not changed.
>>>
>>> --
>>> Philip Martin | Subversion Committer
>>> WANdisco // *Non-Stop Data*
>>
>> In this case I created a new repo using a different name. I left the
>> old one intact. Then just referenced the new repo's url.
>>
>> In the last few days, thought I have moved the old repo to A.orig then
>> moved the new repo to A. I did not know of the requirement to restart
>> apache. Will do that going forward. Thanks.
>
> I just bounced the apache server and it did indeed clear up the
> problem. I can now
> access the repo with both clients. Thank you.
>
> I apologize that I didn't realize that the web server had to be
> bounced. Not sure that was in the docs anywhere?
>
> This may well be the cause of the previous problems as I had a script
> that would create a
> new repo, dump | load to it, then rename the new repo with the
> original repo's name.
>
> I'll keep testing with the new found knowledge.
Since bouncing the apache server I am not seeing any errors with any
of the repos. Classic
case of user error.
I really appreciate all of your help and quick responses.
Thanks,
Curt
Received on 2013-09-12 05:20:43 CEST