> On Aug. 8, 2012, 7:15 p.m., Ali Lown wrote: > > Patch LGTM. > > I am applying to my server to see if it works. (Poke me again in a weeks > > time) > > Ali Lown wrote: > Edit: this patch seems to break the testDeltaHistoryRequiresCorrectHash > test for me? > > Yuri Zelikov wrote: > Didn't run the test, I ll look into it. Thanks. > > Vicente J. Ruiz Jurado wrote: > The test also fails to me. Anyway with this new patch I can't reproduce > the error again.
I am not seeing much (if any) improvement in the rate of crashes due to editions in waves. I was able to have 2 users editing the same wave with each in a different blip for a short period of time. The server was unable to cope with 2 users editing the same blip at the same time. I am getting large quantities of 'WARNING: Failed to transmit message on socket, sequence number 216, message wavelet_name: "eezysys.co.uk/w+2yMFSNdorCA/~/conv+root"' in the logs whilst users are editing, so the failure may not be related directly to this patch. - Ali ----------------------------------------------------------- This is an automatically generated e-mail. To reply, visit: https://reviews.apache.org/r/6474/#review10030 ----------------------------------------------------------- On Aug. 8, 2012, 6:49 p.m., Yuri Zelikov wrote: > > ----------------------------------------------------------- > This is an automatically generated e-mail. To reply, visit: > https://reviews.apache.org/r/6474/ > ----------------------------------------------------------- > > (Updated Aug. 8, 2012, 6:49 p.m.) > > > Review request for wave, Michael MacFadden, Vicente J. Ruiz Jurado, and Ali > Lown. > > > Description > ------- > > WAVE-312 - Not saving editions when concurrent editions in same wave > > > This addresses bug WAVE-312. > https://issues.apache.org/jira/browse/WAVE-312 > > > Diffs > ----- > > src/org/waveprotocol/box/server/waveserver/DeltaStoreBasedWaveletState.java > 98bed26 > > Diff: https://reviews.apache.org/r/6474/diff/ > > > Testing > ------- > > Couldn't reproduce the issue after applying the patch. > > > Thanks, > > Yuri Zelikov > >
