Fisheye svn repository indexing hangs

Hi Community:

I recently upgraded to Fisheye 3.2.4 and I was able to index all svn repositories except one. I have to admnit that this repository is too big. When I wanted to start automatically the repository, it didn't start, so I had (and still having) to make the initial index. While performing the initial reindex, it got stuck in one point (which I will provide later one.)

Indexing content: /file/path/at/the/subversion/repository.file (33873 of 33874)

I performed the following actions

1. Repository Indexing stuck

2. Use the fille:/// protocol, instead of using the http. (Note: The rest of the svn repositories are working fine with the http protocol).

Please, I need your help ASAP!!!!!


6 answers

1 accepted

3 votes
Accepted answer

The problem was that at the repository, I left the Start Revision value as 1. In order to fix it, I needed to empty the field. Then, shutdown Fisheye and clear the repository cache. After start up Fisheye and it will be reindexing as normal

Lukasz Pater Atlassian Team Feb 20, 2014

Tracked as FE-5091

Do you get any messages in FishEye's log? Any Timeouts? Or OutOfMemory errors?

Or is the message "Indexing content: ..." the last one in your log file?

Logfile "fisheye.out" can be found at: <fisheye-data-directory>/var/log/

There is nothing. While I was trying to stopthe repository the error at the log tells me the following : Timed out ro stop the (svn) repository. The approach that I ended up was shutdown Fisheye, delete the cache at $FISHEYE_INST/var/cache/repo and restart Fisheye. Still indexing hangs

We had a similar issue with a certain SVN repository (all other repos works fine), but actually we found a TimeoutException in our logs.

To understand which command of the indexing process runs too long to raise a TimeoutException, we looked into the "Status" view of the corresponding repository. This can be found in FishEye's Administration page at "Repository Settings" -> "Repositories". Search for the "evil" repository and click on cog-wheel icon. Select menu "view ...".

In page "Summary" we saw that one SCM activity took more than 1 hour:

svn proplist -R -r 18889 file:///svn/repos/MYPROJECT/tags/release-1.0@18889

With this information we could continue analyzing the error, because we run the same svn command in systems console. This shows us the actual problem and we could fix it (in our case with configuring a "path excluding pattern" in FE wich excludes this tags/release-1.0 path from scanning).

Maybe this helps you, too?

If the problem is a "simple" Timeout issue you can increase the timeout "Svn Operation Timeout" value at "SCM Details" page of the repo. I think the default is "1h".

Is that revision the latest of your repository?

Is that the latest revision of your repository?

Suggest an answer

Log in or Sign up to answer
Community showcase
Published Wednesday in Jira

Make your Atlassian Cloud products more secure: our NEW admin security guide

Hey admins! I’m Dave, Principal Product Manager here at Atlassian working on our cloud platform and security products. Cloud security is a moving target. As you adopt more products, employees consta...

387 views 0 7
Read article

Atlassian User Groups

Connect with like-minded Atlassian users at free events near you!

Find a group

Connect with like-minded Atlassian users at free events near you!

Find my local user group

Unfortunately there are no AUG chapters near you at the moment.

Start an AUG

You're one step closer to meeting fellow Atlassian users at your local meet up. Learn more about AUGs

Groups near you