[SERVER-3947] mongod segmentation fault Created: 26/Sep/11 Updated: 30/Mar/12 Resolved: 16/Oct/11 |
|
| Status: | Closed |
| Project: | Core Server |
| Component/s: | Querying, Stability, Storage |
| Affects Version/s: | 2.0.0 |
| Fix Version/s: | None |
| Type: | Bug | Priority: | Blocker - P1 |
| Reporter: | Dennis Hoene | Assignee: | Unassigned |
| Resolution: | Duplicate | Votes: | 1 |
| Labels: | FRVIa | ||
| Remaining Estimate: | Not Specified | ||
| Time Spent: | Not Specified | ||
| Original Estimate: | Not Specified | ||
| Environment: |
mongo php-driver (pecl) 1.2.5 |
||
| Operating System: | Linux |
| Participants: |
| Description |
|
I'm getting a strange misbehaviour during heavy doc inputs and transformations with multiple connections. I'm running a single server instance 2.0.0, no replication/sharding. The problem occured since server update from 1.8 to 2.0 and php-driver update from 1.1.4 to 1.2.5. mongod terminates with the following log output: Mon Sep 26 14:52:22 Invalid access at address: 0 Mon Sep 26 14:52:22 Got signal: 11 (Segmentation fault). Mon Sep 26 14:52:22 Backtrace: Logstream::get called in uninitialized state |
| Comments |
| Comment by Eliot Horowitz (Inactive) [ 16/Oct/11 ] |
|
See |
| Comment by kafka0102 [ 15/Oct/11 ] |
|
I have the same errors and make me roll back to 1.8. I just run 2.0 some time and it crashed. It seems not do special commands.Bellow is the relative logs.Hope it can help. Tue Oct 11 16:42:02 Got signal: 6 (Aborted). Tue Oct 11 16:42:02 Backtrace: Tue Oct 11 16:42:02 Invalid access at address: 0x4 Tue Oct 11 16:42:02 Got signal: 11 (Segmentation fault). Tue Oct 11 16:42:02 Backtrace: mongod: /opt/extra/include/boost/thread/pthread/mutex.hpp:50: void boost::mutex::lock(): Assertion `!pthread_mutex_lock(&m)' failed. Tue Oct 11 16:42:23 Backtrace: |
| Comment by Eliot Horowitz (Inactive) [ 14/Oct/11 ] |
|
@dennis - not sure that's the issue - but we're looking into this |
| Comment by Dennis Hoene [ 14/Oct/11 ] |
|
Could dropping one or several indexes in a collection and simultaneously sent queries to the corresponding collection lead to such a breakup? In another case it happened exactly after dropIndexes() and build of a new index (see log). Unfortunately queries aren't being logged. Fri Oct 14 12:38:33 [conn337] CMD: dropIndexes db1.auftraege background Fri Oct 14 12:38:33 Got signal: 11 (Segmentation fault). Fri Oct 14 12:38:33 Backtrace: Logstream::get called in uninitialized state |
| Comment by Dennis Hoene [ 12/Oct/11 ] |
|
After further investigation I can track back the segmentation fault: In my case it also occurs during removal/addition of multiple indexes to multiple collections in the background (during queries / live operation). |
| Comment by Joao Paulo Farias [ 12/Oct/11 ] |
|
Similar issue here, got it while adding an index to a large collection (1.8 million documents) in the background. Log bellow: Wed Oct 12 11:13:44 [conn9] build index done 1812030 records 105.616 secs Wed Oct 12 11:13:45 Got signal: 11 (Segmentation fault). Wed Oct 12 11:13:45 Backtrace: Logstream::get called in uninitialized state |
| Comment by Harald Lapp [ 11/Oct/11 ] |
|
i have a similar issue. The backtrace is: Tue Oct 11 09:11:52 [initandlisten] connection accepted from 10.10.0.12:58806 #68483 Tue Oct 11 09:11:53 Got signal: 11 (Segmentation fault). Tue Oct 11 09:11:53 [initandlisten] connection accepted from 10.10.0.12:58828 #68484 Logstream::get called in uninitialized state Tue Oct 11 09:11:54 Backtrace: |
| Comment by Dennis Hoene [ 06/Oct/11 ] |
|
Got a slightly different backtrace this time: Thu Oct 6 14:08:20 Invalid access at address: 0 Thu Oct 6 14:08:20 Got signal: 11 (Segmentation fault). Thu Oct 6 14:08:21 Backtrace: Logstream::get called in uninitialized state |