[SERVER-3517] Received signal 6 and mongos crashed Created: 02/Aug/11 Updated: 12/Jul/16 Resolved: 09/Aug/11 |
|
| Status: | Closed |
| Project: | Core Server |
| Component/s: | Sharding |
| Affects Version/s: | 1.8.2, 1.8.3 |
| Fix Version/s: | 1.9.2 |
| Type: | Bug | Priority: | Critical - P2 |
| Reporter: | Edward Wei | Assignee: | Greg Studer |
| Resolution: | Done | Votes: | 0 |
| Labels: | None | ||
| Remaining Estimate: | Not Specified | ||
| Time Spent: | Not Specified | ||
| Original Estimate: | Not Specified | ||
| Environment: |
/usr/bin/mongos db version v1.8.3-rc0, pdfile version 4.5 starting |
||
| Attachments: |
|
||||
| Issue Links: |
|
||||
| Operating System: | Linux | ||||
| Participants: | |||||
| Description |
|
I setup 3 config servers , 3 mongod servers for one replicaset and 2 mongos. After running a sequence of operation(create collection -> enablesharding -> insert -> findandmodify / query / delete collection) several times, mongos received signal 6 and crashed. I ran into this error both on 1.8.2 and 1.8.3 rc0. log snippet attached as following: Tue Aug 2 08:18:39 [conn83] CMD: shardcollection: { shardcollection: "dummy.coll_5", unique: false, key: { shardkey1: 1, shardkey2: 1 } } Tue Aug 2 08:18:39 [conn83] about to create first chunk for: dummy.coll_5 max: { shardkey1: MaxKey, shardkey2: MaxKey }Tue Aug 2 08:18:41 [mongosMain] connection accepted from 172.16.0.18:48305 #135 Tue Aug 2 08:19:05 [mongosMain] connection accepted from 172.16.0.18:48307 #140 Tue Aug 2 08:42:08 [conn83] enable sharding on: dummy.coll_3 with shard key: { _id: 1 }Tue Aug 2 08:42:08 [conn83] about to create first chunk for: dummy.coll_3 max: { _id: MaxKey }Tue Aug 2 08:42:15 [mongosMain] connection accepted from 172.16.0.18:42693 #153 |
| Comments |
| Comment by Greg Studer [ 09/Aug/11 ] |
|
SHould be fixed in latest 1.9.2 |
| Comment by auto [ 09/Aug/11 ] |
|
Author: {u'login': u'gregstuder', u'name': u'gregs', u'email': u'greg@10gen.com'}Message: fix semantics of the getChunkManager() method, add non-erroring version |
| Comment by auto [ 09/Aug/11 ] |
|
Author: {u'login': u'gregstuder', u'name': u'gregs', u'email': u'greg@10gen.com'}Message: reload full db when dropping collection - |
| Comment by Greg Studer [ 05/Aug/11 ] |
|
Thanks for the logs, very helpful. We're working on a fix, track the ticket for progress. |
| Comment by Edward Wei [ 04/Aug/11 ] |
|
Hi, |
| Comment by Greg Studer [ 03/Aug/11 ] |
|
can you reproduce this issue with mongos at loglevel 2? -vv or admin.runCommand( { setParameter : 1, logLevel : 2 })? |
| Comment by Greg Studer [ 03/Aug/11 ] |
|
Tracking down the issue, as a temporary workaround it may be possible to just wait a few seconds before deleting the collection (or check that writebacksQueued = false on the mongod instances). |