Repository State Error

Question

We've been running crucible on multiple repositories. However on one the repositories indexing started to fail a couple of weeks ago. It fails at 70% indexing, scanning is successful. It returns the following text

"Fri Mar 02 09:33:22 CET 2018: Repository paused due to error com.cenqua.fisheye.rep.RepositoryClientException: java.lang.IllegalStateException: Can't overwrite cause with org.tmatesoft.svn.core.SVNException: svn: E175002: Connection reset svn: E175002: REPORT request failed on '/svn/RenDArchief/!svn/vcc/default' java.lang.IllegalStateException: Can't overwrite cause with org.tmatesoft.svn.core.SVNException: svn: E175002: Connection reset svn: E175002: REPORT request failed on '/svn/RenDArchief/!svn/vcc/default' org.apache.subversion.javahl.ClientException: svn: E175002: Connection reset svn: E175002: REPORT request failed on '/svn/RenDArchief/!svn/vcc/default'"

What is the actual issue, and how can I fix it?

Andy Heinzer · Accepted Answer

From reviewing this error and looking at similar cases, it looks like the system is reaching a timeout before it can complete the indexing process.  Please check out this KB: Indexing SVN Repository fails due to svn: E175002: CRLF expected at end of chunk: -1/-1

The error is similar in nature to yours, if not exactly the same.   I also found similar posts where the solution in that KB appears to help other related cases like

https://community.atlassian.com/t5/Fisheye-Crucible-questions/Fisheye-can-t-fully-index-SVN-repo/qaq-p/387036

And the related KB: FishEye/Crucible svn: E175002: SSL peer shut down incorrectly

In all of these, it appears that if you have a reverse proxy serving up this connection the solution is to extend the timeout of that proxy.