Details
-
Bug
-
Status: Closed
-
Major
-
Resolution: Duplicate
-
None
-
None
-
None
-
None
Description
While investigating TS-3440 (and writing the subsequent test cases) I have yet to find a case where I can trigger this feature (hostdb caching a real as down). In the case of TS-3440 we know that the real is unavailable, but we end up re-connecting anyways. Regardless of the broken re-connect logic we shouldn't re-connect since hostdb should have marked the real as down and cached it for 300s.
Note: if this is broken, we'll have a rough time fixing it as it would effectively be a backwards incompatible change. If it is broken we'll probably want to change the default config to match the current behavior.
Attachments
Issue Links
- duplicates
-
TS-2836 TS attempt to connect to dead server at least once
- Closed