Constant "All indexers are unavailable due to failures" after internet failure

Help & Support
1

Version
2.0.0.3953
Mono Version
4.2.1 (Stable 4.2.1.102/6dd2d0d Thu Nov 12 09:52:44 UTC 2015)

OS: ubuntu 14.04
((Debug logs)) (posted to hastebin or similar):
Description of issue (if you think you’ve found a bug please include steps to reproduce):

My internet drops for a minute or two relatively frequently - Im working with my ISP to resolve this, but i think they are going to need to rerun cable. Due to these frequent drop outs, sonarr thinks my indexers are failing. All i ever need to do is go in a re test them and its fine again, but it kind of defeats the purpose of automation if its something i have to go in and do each time i want to check for a new episode or something.

Is there a way to extend the timeout period, or prevent it from “giving up” on an indexer if it fails too many times within that period?

2

Its not immediate failure, it escalates the more times it fails:

5, 15, 30, 60 minutes then 3, 6, 12, 24 hours. If the request succeeds after a failure the escalation level is reduced.

There is no way to change this behavior, but restarting Sonarr will also clear it.

3

Actually, no, Backoff is stored in the db.

@Ken_Kyger Since it deescalates upon success, it means it fails +50% of the time, that’s some crappy ISP connection.

Either way, in version .3953 we already don’t escalate if the failure is due to a dns or connection failure. So i’m guessing it’s some other failure, timeout for example.

4

This topic was automatically closed 60 days after the last reply. New replies are no longer allowed.