[SERVER-16759] Query uses unexpected index Created: 07/Jan/15 Updated: 06/Apr/23 Resolved: 10/Jan/15 |
|
| Status: | Closed |
| Project: | Core Server |
| Component/s: | Index Maintenance |
| Affects Version/s: | 2.6.6 |
| Fix Version/s: | None |
| Type: | Bug | Priority: | Critical - P2 |
| Reporter: | Alexander S. | Assignee: | Ramon Fernandez Marina |
| Resolution: | Done | Votes: | 0 |
| Labels: | None | ||
| Remaining Estimate: | Not Specified | ||
| Time Spent: | Not Specified | ||
| Original Estimate: | Not Specified | ||
| Operating System: | ALL |
| Participants: |
| Description |
|
We didn't have such problems with 2.4 so this is more like a regression in 2.6.6. We have the next indexes:
And the next query:
Previously the query didn't took more than a few ms, now 15 seconds in average. |
| Comments |
| Comment by Ramon Fernandez Marina [ 10/Jan/15 ] | |||||||||||||||
|
Hi heaven, thanks for letting us know. I'm going to resolve this ticket as "gone away", but if the issue appears again after you've refilled the collection please reopen the ticket and send us the information requested above so we can investigate if there's a bug here. Regards, | |||||||||||||||
| Comment by Alexander S. [ 10/Jan/15 ] | |||||||||||||||
|
It seems the problem gone after re-creating the index. Need a few more days to refill the collection and retry. | |||||||||||||||
| Comment by Ramon Fernandez Marina [ 08/Jan/15 ] | |||||||||||||||
|
Hi heaven, we'll need some more information to diagnose this behavior. Could you please follow the steps below and upload the resulting information?
You can put each piece of information in a file, zip all files along with the mongod.log log files from your cluster and post it to this ticket. If the resulting file is too large (hundreds of megabytes) please let me know and I'll send you different upload instructions. Thanks, | |||||||||||||||
| Comment by Alexander S. [ 08/Jan/15 ] | |||||||||||||||
|
Any suggestions please? This really hurts us. We also tried:
And the next query:
This worked with 2.4, I remember we had over 50 000 000 records in the queue and didn't experience any performance issues. But none of the above work in 2.6.6. | |||||||||||||||
| Comment by Alexander S. [ 07/Jan/15 ] | |||||||||||||||
|
Please change the type from feature to bug. UPD: sometimes it does work, sometimes — doesn't. We didn't change anything in our code but just started getting performance alerts after upgrade. db.upgradeCheckAllDBs() shows everything is fine with the database. |