19263 2012-10-03 20:36:04 +0000 Unexpectedly logged out and lost all changes 2012-10-05 04:37:12 +0000 1 1 1 Unclassified webplatform.org infrastructure unspecified PC All RESOLVED DUPLICATE 19274 P1 normal --- 1 luzc schepers cmills komoroske ssweeney team-webplatform-admin public-webplatform-bugs oldest_to_newest 75255 0 luzc 2012-10-03 20:36:04 +0000 Got kicked out after not being active for aprox 10 min, and lost all changes. There was no indication I had been logged out until I tried to save changes. 75299 1 cmills 2012-10-04 09:15:52 +0000 (In reply to comment #0) > Got kicked out after not being active for aprox 10 min, and lost all changes. > > There was no indication I had been logged out until I tried to save changes. Argh - got it is annoying when that happens. I have lost so much work on Wikis due to this kind of thing. Can we set the time out to say, 2 hours, to make sure it doesn't happen? 75303 2 komoroske 2012-10-04 12:59:57 +0000 Yesterday there was a lot of server work that led to various times when edits/logins wouldn't work. They were announced on IRC and in person at the hackathon. It's _possible_ that this problem just happened to coincide with one of those times. For what it's worth, I've never seen that happen to me--although every so often it will say "could not save because of session problem" or something similar, in which case a re-submit on that page normally works. 75307 3 luzc 2012-10-04 16:35:34 +0000 (In reply to comment #2) > Yesterday there was a lot of server work that led to various times when > edits/logins wouldn't work. They were announced on IRC and in person at the > hackathon. > > It's _possible_ that this problem just happened to coincide with one of those > times. For what it's worth, I've never seen that happen to me--although every > so often it will say "could not save because of session problem" or something > similar, in which case a re-submit on that page normally works. Hey Alex, I filed the bug because I asked around and as far as we all knew it wasn't one of those times, and Peter (the Hungarian guy, don't remember his surname, sorry) said the same thing happened to him earlier that day. That's why the guy from Microsoft (? again don't remember the name, sorry) said to file it as P1 75331 4 ssweeney 2012-10-05 04:37:12 +0000 Believe this is a dup of: https://www.w3.org/Bugs/Public/show_bug.cgi?id=19274 Where test and prod servers were both tied to the same memcache namespace. *** This bug has been marked as a duplicate of bug 19274 ***