Hi Paul, reviewboard only stores the uploaded diffs in the database, but queries the VCS in order to display the file diffs for review (it gets the original file, applies the diff (well, patch), and displays the side-by-side diff.
It needs the VCS in order to display the snippets in each review in the top-level review page. This is probably the reason #2 didn't work for you. It should work fine for reviews that don't have any comments, though, as no VCS access is required. Also, I believe that reviewboard caches files it requests from the VCS, as well as the files with the diff applied. So if you're using a proper cache system, you should see less frequent VCS accesses. Eyal On Oct 1, 1:45 pm, pfee <[EMAIL PROTECTED]> wrote: > Hello all, > > I'd like to query at what times the reviewboard server interacts with > the version control system. > > I've been using reviewboard, but recently our VCS system was offline > for a while. This affected the operation of reviewboard. I think > improvements could be made in this area. > > I found that the dashboard functioned as normal, but I was unable to > progress to view a review. > > In the event of a VCS failure, reviewboard server should be able to: > 1) Show the dashboard (currently works) > 2) Show the top level review page (doesn't work) > 3) Show the diffs - I'm assuming here that they're stored within > reviewboard (doesn't work) > > I also assume that reviewboard doesn't store the entire contents of > the files for review, hence I would not expect the expanded entire > file view to work during a VCS failure. > > Perhaps reviewboard could be configurable so that it could store the > entire file contents. This would further isolate it from the VCS > making it more fault tolerant. However the cost would be increased > storage requirements. I doubt if this should be the default setting, > but it could be an option. This enhancement would be a lower priority > than the other suggestions here. > > Eliminating unnecessary VCS access will make reviewboard work better > during failure conditions and should also give speed ups even under > normal operation. > > I'm posting here first rather than raising a bug. If these ideas seem > worthwhile, I'll raise a bug/feature request. > > Thanks, > Paul --~--~---------~--~----~------------~-------~--~----~ You received this message because you are subscribed to the Google Groups "reviewboard" group. To post to this group, send email to reviewboard@googlegroups.com To unsubscribe from this group, send email to [EMAIL PROTECTED] For more options, visit this group at http://groups.google.com/group/reviewboard?hl=en -~----------~----~----~----~------~----~------~--~---
