We have a SVN server here that over the years has accumulated hundreds of thousands of revisions and reached a size of >40 GB. We’d like to do a continuous backup on the repository, but it just takes too long to dump or copy the whole repository to our backup server each time.
Is there a way to add the latest revisions of the repository to an existing dump file automatically? I know with the --incremental option this can be done manually, but I was wondering if there was a command that would essentially work like so:
svnadmin dump repo --revision dumpfile_latest:repo_latest --incremental >> dumpfile
Here dumpfile_latest would be the revision number of the latest revision of the backup (named dumpfile) and repo_latest would be the latest revision number of the repository I am backing up (named repo). Thanks for any suggestions!
I currently have this setup at work. We use a customer hosted SVN server that at times has had some connectivity issues. We also want to make sure we have a local copy of things as well. We setup a SVN repo locally, and then run the svnsync command that is executed as a result of a continuous integration service to sync changes from the remote repo to the local mirror. You can read more about the svnsync command here. This generates an exact replica of the repo as it basically plays back the changeset into the mirror. This is also helpful for us in that if the connectivity “issues” return, we have a readonly repo available to pull working copies from to continue work until connectivity is restored ( you can do a svn switch/relocate to work off the mirror while the master is down ). This may be more complex then you are looking for, but I love the setup we have and it has saved my bacon more than once. Figured I would share an option. Good luck