We have experienced a very poor performance of the svn log command on the new repository that had only a few revisions.

It took more than a minute(!) to get history that had only one revision - the initial import.

While blaming the server configuration and network security, we didn't find anything unusual that could affect the svn performance so much.

The reason was much simpler - initial import of the full project tree had more than 32,000 files, and it turned out to be too much for one revision.

Breaking the initial import into several commits solved the problem completely.

If you have a big project to import, instead of running

svn import myproject file:///opt/opencm/svnroot/trunk/myproject -m "Initial import"

write a short script to import smaller parts, like:

cd myproject