[SERVER-4083] MongoDB crashed with corrupted double-linked list error Created: 17/Oct/11 Updated: 29/Feb/12 Resolved: 30/Jan/12 |
|
| Status: | Closed |
| Project: | Core Server |
| Component/s: | Stability |
| Affects Version/s: | 2.0.0 |
| Fix Version/s: | None |
| Type: | Bug | Priority: | Major - P3 |
| Reporter: | Evgeny Gridasov | Assignee: | Mathias Stearn |
| Resolution: | Done | Votes: | 1 |
| Labels: | vmware1 | ||
| Remaining Estimate: | Not Specified | ||
| Time Spent: | Not Specified | ||
| Original Estimate: | Not Specified | ||
| Environment: |
RHEL 5.2, running inside VmWare, 2 CPUs, 4Gb of RAM |
||
| Attachments: |
|
| Operating System: | Linux |
| Participants: |
| Description |
|
An error occurred during an intensive write operation and MongoDB shut down. That was a master server which was configured to replicate to other 3 servers using replication sets.
Please see attachment with full stack trace. |
| Comments |
| Comment by Ben Hu [ 22/Jan/12 ] |
|
Yah, my stack trace did not include miniwebserver, I will open a separate ticket https://jira.mongodb.org/browse/SERVER-4741 with my log info. |
| Comment by Mathias Stearn [ 21/Jan/12 ] |
|
The double-linked list error isn't a real error in most cases. That is, it is fallout from an earlier failure and the inability to recover from a bad state. @Evgeny: In the log that was posted with this ticket it looks like it was a failure in the built-in web server. Do you know what query caused that? Is there any chance a mongo client connected to the HTTP port rather than the mongo port? @Ben: If your stack trace didn't include miniwebserver, please open a separate ticket as it is unlikely to be related to this one. |
| Comment by Ben Hu [ 23/Nov/11 ] |
|
We had same crash with double-linked list error. We are using vmware and our system is Ubuntu 11.04. It happens in 2.0.1 as well. |
| Comment by Evgeny Gridasov [ 02/Nov/11 ] |
|
Have you checked the mongodb.log file that I've attached with this issue? |
| Comment by Mathias Stearn [ 01/Nov/11 ] |
|
Do you still have the logs? |
| Comment by Mathias Stearn [ 17/Oct/11 ] |
|
Could you attach the full log, not just the last few lines? |
| Comment by Mathias Stearn [ 17/Oct/11 ] |
|
We are looking into this now. Have you only seen this in vmware or also on bare metal? Also, what does your storage subsystem look like (local disk, FibreChannel, iSCSI, NFS, etc)? |