4 ms·

This strikes me as odd: "After a complaint from one of our users [...], we realized that the very same thing could happen to someone else" One user complains a

by rrrene 14y ago

This strikes me as odd: "After a complaint from one of our users [...], we realized that the very same thing could happen to someone else" One user complains about something that might have happened by his own doing and the OP starts implementing a feature, that not circumvents the end result (can't save the content) but notifies the user that the software is broken somehow. Anyway, I'm genuinely interested in the following: What are these "universe glitches", when one gets logged out by the browser? Session Cookie vanishes? Server restart invalidates some kind of access token?

terinjokes 14y ago
That's what I was wondering, they solved user's problem, but not their problem. To me, it sounds like they might be expiring the session due to inactivity, while the user is actively writing something. EDIT: Typo
- relix 14y ago
  Yep, that's probably exactly what's happening, I've had it happen too. And the solution is the ping that he proposes. So if that was the problem, the ping is resetting the session cookie every 60 seconds and the problem should go away. One thing I would add is a logging feature that tracks how many times this actually happens (is happening) so they have metrics to back up any decision in the future.
  - terinjokes 14y ago
    Then why isn't this ping already part of the auto-save feature, and not an extra request.
glassx 14y ago
The sessions are probably stored on the server, and they expire after some time. It happens a lot here where I work: our legacy ASP.NET WebForms use Sessions (because it used to be easier this way) and once in a while people complain that the app kicks them after some 20 minutes.
mootothemax 14y ago
What are these "universe glitches", when one gets logged out by the browser I've had a situation in the past where a session has regenerated - giving the impression of expiring - due to lots of AJAX calls over a period in time. Basically, you got X number of pageviews with the current session token before a new token was generated. IIRC, the browser's session cookie wasn't getting updated due to the calls being via AJAX, and hence the user was logged out. It's been a long time since this happened though, so I may have a detail wrong.

by rrrene 14y ago

terinjokes 14y ago

That's what I was wondering, they solved user's problem, but not their problem. To me, it sounds like they might be expiring the session due to inactivity, while the user is actively writing something. EDIT: Typo

relix 14y ago
Yep, that's probably exactly what's happening, I've had it happen too. And the solution is the ping that he proposes. So if that was the problem, the ping is resetting the session cookie every 60 seconds and the problem should go away. One thing I would add is a logging feature that tracks how many times this actually happens (is happening) so they have metrics to back up any decision in the future.
- terinjokes 14y ago
  Then why isn't this ping already part of the auto-save feature, and not an extra request.

glassx 14y ago

The sessions are probably stored on the server, and they expire after some time. It happens a lot here where I work: our legacy ASP.NET WebForms use Sessions (because it used to be easier this way) and once in a while people complain that the app kicks them after some 20 minutes.

mootothemax 14y ago

What are these "universe glitches", when one gets logged out by the browser I've had a situation in the past where a session has regenerated - giving the impression of expiring - due to lots of AJAX calls over a period in time. Basically, you got X number of pageviews with the current session token before a new token was generated. IIRC, the browser's session cookie wasn't getting updated due to the calls being via AJAX, and hence the user was logged out. It's been a long time since this happened though, so I may have a detail wrong.