Public bug reported:
It looks like requests to the remote server time out more often than we
care to admit. When that happens, we treat the failure as a temporary
and surprising error and disable remote server access for two hours.
We need something more sophisticated so we can react to timeout errors
with a different strategy than to other temporary surprising errors. The
problem is that a dead server is indistinguishable from a slow one, so
we probably need some exponential back-off strategy, or count the number
of consecutive timeout failures and trigger the full two-hour back-off
only if some threshold is exceeded.
** Affects: thumbnailer (Ubuntu)
Importance: High
Assignee: Michi Henning (michihenning)
Status: New
** Changed in: thumbnailer (Ubuntu)
Assignee: (unassigned) => Michi Henning (michihenning)
** Changed in: thumbnailer (Ubuntu)
Importance: Undecided => High
--
You received this bug notification because you are a member of Ubuntu
Bugs, which is subscribed to Ubuntu.
https://bugs.launchpad.net/bugs/1528105
Title:
Error back-off too aggressive
To manage notifications about this bug go to:
https://bugs.launchpad.net/ubuntu/+source/thumbnailer/+bug/1528105/+subscriptions
--
ubuntu-bugs mailing list
[email protected]
https://lists.ubuntu.com/mailman/listinfo/ubuntu-bugs
