Excellent - thanks Chris! On Wednesday, January 16, 2013 2:39:20 PM UTC-8, Christian Hammond wrote: > > It won't fix existing review requests. The data is truncated at write, so > you have to fix it manually, by hand. It will only prevent the problem for > new review requests. > > It's certainly not the content. This is an old, very well known and tested > bug. The problem is that there's only so much room in the field used to > store the JSON payload, which contains the whole of all the changes made > since the last draft. If you had a very long Testing Done, or a smaller > Testing Done with a Description and other fields, or whatever combination, > you can trigger this. It also stores the old and new versions. So going > from a large Testing Done to another large Testing Done would result in > this as well. > > The fix was to migrate to a larger field that could store much larger data > sets, more than you'll ever actually end up putting into it. That was done > at some point in the 1.6 releases. The upgrade to the new field should be > automatic when upgrading the site. > > Christian > > -- > Christian Hammond - chi...@chipx86.com <javascript:> > Review Board - http://www.reviewboard.org > VMware, Inc. - http://www.vmware.com > > On Jan 16, 2013, at 2:35 PM, Steve <seide...@gmail.com <javascript:>> > wrote: > > I have a 1.6 server staged for testing that I'm planning to roll out soon. > I'll import the latest database and see if the problem shows up there as > well. > > As for > > > Tell developers not to paste so much :) > > I thought it was the size of the text also, but I'm beginning to think > it's not. I've been unable to reproduce with text much greater than what's > being used in the broken reviews. I think it has something to do with the > content. In many cases actual test scripts are being uploaded, which means > they're filled with quotes, {} and other things that could be problematic > to a parser. > > I'll try repairing it through the admin UI. I don't suppose I could make > it any worse. > > --Steve > > > On Wednesday, January 16, 2013 11:26:18 AM UTC-8, Christian Hammond wrote: >> >> Hi Steve, >> >> This breaks on the review request page because the Testing Done text is >> being stored in a JSON structure along with other data for the "Review >> request updated" sections. These sections do not show on the diff viewer. >> >> Repairing this would require going into the database viewer in the >> administration UI, locating the Change Description table for the change (we >> don't give you the easiest ways to do that, so you'll have to try narrowing >> it down by date), and fixing the content to be valid JSON. YMMV. >> >> Tell developers not to paste so much :) >> >> Actually, the reason you're having this problem is that your release is >> so old. If you upgrade, this shouldn't happen on newer review requests. >> >> Christian >> >> -- >> Christian Hammond - chi...@chipx86.com >> Review Board - http://www.reviewboard.org >> VMware, Inc. - http://www.vmware.com >> >> >> On Wed, Jan 16, 2013 at 10:55 AM, Steve <seide...@gmail.com> wrote: >> >>> I get about 1 of these per month, but I have 2 this week and it's >>> getting to be a real problem. Developers paste large amounts of text into >>> the Testing Done field, after which going to the main page for the review >>> yields this: >>> >>> ======================== >>> Something broke! (Error 500) >>> >>> It appears something broke when you tried to go to here. This is >>> either a bug in Review Board or a server configuration error. Please >>> report this to your administrator. >>> ======================== >>> >>> The odd thing is we can go around the main page to the diff page and see >>> the text in Testing Done, which is baffling. In other words, >>> >>> http://reviewboard/r/123 >>> >>> fails with the above error, but >>> >>> http://reviewboard/r/123/diff/#index_header >>> >>> looks fine. This is really causing problems for our developers. I have >>> 3 questions in descending order of importance: >>> >>> 1. How can I repair a review when it gets in this state? >>> 2. What guidelines can I give my developers to prevent this from >>> happening? >>> 3. Why does the main page fail, when the text appears to be fine in the >>> diff_header page? >>> >>> We're using 1.5.5 on CentOS 5. >>> >>> Thanks >>> >>> --Steve >>> >>> >>> -- >>> Want to help the Review Board project? Donate today at >>> http://www.reviewboard.org/donate/ >>> Happy user? Let us know at http://www.reviewboard.org/users/ >>> -~----------~----~----~----~------~----~------~--~--- >>> To unsubscribe from this group, send email to >>> reviewboard...@googlegroups.com >>> For more options, visit this group at >>> http://groups.google.com/group/reviewboard?hl=en >>> >>> >>> >> >> > -- > Want to help the Review Board project? Donate today at > http://www.reviewboard.org/donate/ > Happy user? Let us know at http://www.reviewboard.org/users/ > -~----------~----~----~----~------~----~------~--~--- > To unsubscribe from this group, send email to > reviewboard...@googlegroups.com <javascript:> > For more options, visit this group at > http://groups.google.com/group/reviewboard?hl=en > > > > >
-- Want to help the Review Board project? Donate today at http://www.reviewboard.org/donate/ Happy user? Let us know at http://www.reviewboard.org/users/ -~----------~----~----~----~------~----~------~--~--- To unsubscribe from this group, send email to reviewboard+unsubscr...@googlegroups.com For more options, visit this group at http://groups.google.com/group/reviewboard?hl=en
