Actions
Bug #1676
closedShortMessageDatabase>>#waitFor: is "racing"
Start date:
03/29/2016
Due date:
% Done:
0%
Spec Reference:
Description
The code assumes that MongoTalk is picking the "right" response for the request but it mostly ignores the requestId in the response. This will lead to a race in:
tailableCursor := MongoTailableCursor new query: query; root: root; yourself. tailableCursor create. self smsPending ifTrue: [ ^ShortMessageWorkPending]. tailableCursor requestNext.
It might be that in self smsPending we get the response for the tailable cursor. In the best case we get a parse error which leads to a re-connect of the database and immediate re-try.
Updated by zecke about 7 years ago
- Status changed from New to Closed
There is now a "tail" thread that is waiting for changes from the database.
Actions