I have a subversion repository around 400MB on my debian machine (unstable) and didn't encounter any problems yet. I also have a much smaller repository on a sarge machine - don't have problems there.
about connection way: sid - svn:// sarge svn+ssh:// > CVS on steroids.... atomic commits.... we all heard of those nice > slogans... > > > So I installed one of those toys on a Debian-testing server only to > found out that after each import of the repository the Berkly data base > gets corrupted. The only way to recover it is using "svnadmin recover > /svn/repository1" for example. RTFM leads me to the conclusion that it > can happen mainliy because the connection to the SVN server gets killed > to "fast" on brutally. > > > I am quite sure it is not the case, since I was trying the SVN server > remotely from SSH (logged into the server remotely and then I SVNed to > svn+ssh://localhost...). > > > What I did now was migrating from bdb backend to fsfs, which is a new > backend in Subversion 1.1.X. I am not trusting this server so blindly > has I have before. I will wait a few days before I decide if this new > set up will work. If not, good old CVS to the rescue... > > > I have heard many disaster stories from SVN. One of them is mine, as it > seems that the bdb backend is problematic under MacOS and Debian. Does > anyone have a clue what is really hapenning there? Anyone else > experienced those symptoms? > > > > > ================================================================= > To unsubscribe, send mail to [EMAIL PROTECTED] with > the word "unsubscribe" in the message body, e.g., run the command > echo unsubscribe | mail [EMAIL PROTECTED] > > ================================================================= To unsubscribe, send mail to [EMAIL PROTECTED] with the word "unsubscribe" in the message body, e.g., run the command echo unsubscribe | mail [EMAIL PROTECTED]
