[SERVER-9141] sharded CursorCache should timeout unsharded cursor refs Created: 27/Mar/13 Updated: 12/Oct/15 Resolved: 12/Oct/15 |
|
| Status: | Closed |
| Project: | Core Server |
| Component/s: | Sharding |
| Affects Version/s: | 2.0.8, 2.2.3, 2.4.1 |
| Fix Version/s: | None |
| Type: | Bug | Priority: | Major - P3 |
| Reporter: | Greg Studer | Assignee: | Unassigned |
| Resolution: | Duplicate | Votes: | 0 |
| Labels: | None | ||
| Remaining Estimate: | Not Specified | ||
| Time Spent: | Not Specified | ||
| Original Estimate: | Not Specified | ||
| Issue Links: |
|
||||||||||||||||
| Operating System: | ALL | ||||||||||||||||
| Participants: | |||||||||||||||||
| Description |
|
The sharded CursorCache should not only timeout sharded cursors, but should also timeout references to unsharded cursors that have been idle for the same amount of time. Interactions with timeout flags (i.e. no timeout cursors) also need to be handled. |
| Comments |
| Comment by J Rassi [ 12/Oct/15 ] |
|
Yes. ClusterCursorManager::killMortalCursorsInactiveSince() kills both sharded and unsharded cursors. Closing as a duplicate of |
| Comment by Andy Schwerin [ 12/Oct/15 ] |
|
jason.rassi, is this fixed on master by the new cursor management code in mongos? |