[server] Continuing problems with svn operations
I'm using svn a lot as part of the run-up to 1.46.0, but am having trouble. svn operations are very slow, and often fail. A typical error message is: Error: REPORT of '/svn/boost/!svn/vcc/default': Could not read chunk size: An existing Error: connection was forcibly closed by the remote host. Error: (https://svn.boost.org) Can anything be done about svn problems? I know Rene has already done a lot of work trying to improve the situation, but there are still serious performance problems. --Beman
On 1/13/2011 6:31 AM, Beman Dawes wrote:
I'm using svn a lot as part of the run-up to 1.46.0, but am having trouble. svn operations are very slow, and often fail. A typical error message is:
Error: REPORT of '/svn/boost/!svn/vcc/default': Could not read chunk size: An existing Error: connection was forcibly closed by the remote host. Error: (https://svn.boost.org)
Can anything be done about svn problems? I know Rene has already done a lot of work trying to improve the situation, but there are still serious performance problems.
Recently I had to rebuild my own SVN server. Along the way I learned a fair amount of "things" about the SVN server setup. But two unfortunate items about this are: 1. I won't have time to really work on tweaking the server for some time. Sorry about that, but my daughter is starting daycare and it's taking much of my "free time" away :-) 2. I'm now very familiar with the above connection error. It's caused by bugs in the WebDAV module that appears only when using it over SSL (https). For my own server I had to run a reverse proxy doing the SSL (lighttpd -- which has its own problems depending on which version you use) connecting to the Apache server running only the WebDAV/SVN on HTTP. So the unfortunate part is that there doesn't seem to be any hope of the WebDAV stuff ever getting fixed when used with HTTPS. When I get back to looking at the server one of my first changes is going to be to remove the HTTPS side, and change to non-clear-text authentication to both fix the above problem and to relieve a bit more server CPU. But if one of the real sys-admins can get to this before I do I would be very grateful ;-) -- -- Grafik - Don't Assume Anything -- Redshift Software, Inc. - http://redshift-software.com -- rrivera/acm.org (msn) - grafik/redshift-software.com -- 102708583/icq - grafikrobot/aim,yahoo,skype,efnet,gmail
Rene Rivera wrote:
On 1/13/2011 6:31 AM, Beman Dawes wrote:
I'm using svn a lot as part of the run-up to 1.46.0, but am having trouble. svn operations are very slow, and often fail. A typical error message is:
Error: REPORT of '/svn/boost/!svn/vcc/default': Could not read chunk size: An existing Error: connection was forcibly closed by the remote host. Error: (https://svn.boost.org)
Can anything be done about svn problems? I know Rene has already done a lot of work trying to improve the situation, but there are still serious performance problems.
Recently I had to rebuild my own SVN server. Along the way I learned a fair amount of "things" about the SVN server setup. But two unfortunate items about this are:
1. I won't have time to really work on tweaking the server for some time. Sorry about that, but my daughter is starting daycare and it's taking much of my "free time" away :-)
2. I'm now very familiar with the above connection error. It's caused by bugs in the WebDAV module that appears only when using it over SSL (https). For my own server I had to run a reverse proxy doing the SSL (lighttpd -- which has its own problems depending on which version you use) connecting to the Apache server running only the WebDAV/SVN on HTTP. So the unfortunate part is that there doesn't seem to be any hope of the WebDAV stuff ever getting fixed when used with HTTPS. When I get back to looking at the server one of my first changes is going to be to remove the HTTPS side, and change to non-clear-text authentication to both fix the above problem and to relieve a bit more server CPU. But if one of the real sys-admins can get to this before I do I would be very grateful ;-)
Why do we use HTTP in the first place? svn:// should be less overhead. - Volodya -- Vladimir Prus Mentor Graphics +7 (812) 677-68-40
participants (3)
-
Beman Dawes -
Rene Rivera -
Vladimir Prus