[SERVER-740] Crash on geo search Created: 12/Mar/10  Updated: 12/Jul/16  Resolved: 12/Mar/10

Status: Closed
Project: Core Server
Component/s: Stability
Affects Version/s: 1.3.3
Fix Version/s: 1.3.4

Type: Bug Priority: Major - P3
Reporter: Valerii Hiora Assignee: Eliot Horowitz (Inactive)
Resolution: Done Votes: 0
Labels: None
Remaining Estimate: Not Specified
Time Spent: Not Specified
Original Estimate: Not Specified
Environment:

OS X 10.6.2


Attachments: Zip Archive dump.zip    
Participants:

 Description   

On my specific data it always crashes on geo searching using:
db.points.find({ loc : { $near : [50, 50] }}) - crash
db.points.find({ loc : { $near : [50, 50] }}).limit(26) - works
db.points.find({ loc : { $near : [70, 70] }}) - works

Typical data is: { "_id" : ObjectId("4b9a36c3a5ff9d0cb900c20c"), "loc" :

{ "lat" : 54.72271, "lon" : 56.02658 }

}
The problem is that I can't see which exactly data causes the problem.

Let me know if you need actual data and which method do you prefer to get it.



 Comments   
Comment by Hubert Lepicki [ 12/Mar/10 ]

I was just going to report that, was also happening on Linux 64 bit. Will check if it's gone now.

Comment by Valerii Hiora [ 12/Mar/10 ]

Thanks for quick fix.

Comment by Eliot Horowitz (Inactive) [ 12/Mar/10 ]

now actually fixed

Comment by Eliot Horowitz (Inactive) [ 12/Mar/10 ]

closed before by accident

Comment by auto [ 12/Mar/10 ]

Author:

{'login': 'erh', 'name': 'Eliot Horowitz', 'email': 'eliot@10gen.com'}

Message: fix boundary geo case SERVER-740
http://github.com/mongodb/mongo/commit/5cb4789317b8e93dd4127b22c9e52ead4d9aba69

Comment by Valerii Hiora [ 12/Mar/10 ]

I've extracted only locations

Comment by Eliot Horowitz (Inactive) [ 12/Mar/10 ]

Ok - not seeing it right away.
If you can send a mongodump of that collection, that would help a lot.
you can attach it here or open a SUPPORT case it its senstive

Comment by auto [ 12/Mar/10 ]

Author:

{'login': 'erh', 'name': 'Eliot Horowitz', 'email': 'eliot@10gen.com'}

Message: logRotate command to rotate logs SERVER-740
http://github.com/mongodb/mongo/commit/74e22e9d7377e1a275066a6f561f5955b0a80d18

Comment by Valerii Hiora [ 12/Mar/10 ]

Sure, here it is:

Fri Mar 12 16:53:20 Got signal: 11 (Segmentation fault).
Fri Mar 12 16:53:20 Backtrace:
0x1001f0a5c 0x7fff88a1ceaa 0x80000000 0x1001b117d 0x1001a7698 0x10017a00a 0x1000bf7bc 0x10017bbb9 0x10017f3d4 0x100184c8a 0x1000b401e 0x100140b17 0x100145b65 0x1001fc11b 0x100208e64 0x7fff889f5f8e 0x7fff889f5e41
0 mongod 0x00000001001f0a5c _ZN5mongo10abruptQuitEi + 332
1 libSystem.B.dylib 0x00007fff88a1ceaa _sigtramp + 26
2 ??? 0x0000000080000000 0x0 + 2147483648
3 mongod 0x00000001001b117d _ZN5mongo9GeoSearch4execEv + 2413
4 mongod 0x00000001001a7698 _ZNK5mongo9Geo2dType9newCursorERKNS_7BSONObjES3_i + 744
5 mongod 0x000000010017a00a _ZNK5mongo9QueryPlan9newCursorERKNS_7DiskLocEi + 698
6 mongod 0x00000001000bf7bc _ZN5mongo11UserQueryOp4initEv + 1676
7 mongod 0x000000010017bbb9 _ZN5mongo12QueryPlanSet6Runner6initOpERNS_7QueryOpE + 25
8 mongod 0x000000010017f3d4 _ZN5mongo12QueryPlanSet6Runner3runEv + 596
9 mongod 0x0000000100184c8a _ZN5mongo12QueryPlanSet5runOpERNS_7QueryOpE + 74
10 mongod 0x00000001000b401e _ZN5mongo8runQueryERNS_7MessageERNS_12QueryMessageERNS_5CurOpE + 2238
11 mongod 0x0000000100140b17 _ZN5mongo13receivedQueryERNS_6ClientERNS_10DbResponseERNS_7MessageE + 647
12 mongod 0x0000000100145b65 _ZN5mongo16assembleResponseERNS_7MessageERNS_10DbResponseERK11sockaddr_in + 3973
13 mongod 0x00000001001fc11b _ZN5mongo10connThreadEv + 619
14 mongod 0x0000000100208e64 thread_proxy + 132
15 libSystem.B.dylib 0x00007fff889f5f8e _pthread_start + 331
16 libSystem.B.dylib 0x00007fff889f5e41 thread_start + 13
Fri Mar 12 16:53:20 dbexit:
Fri Mar 12 16:53:20 shutdown: going to close listening sockets...
Fri Mar 12 16:53:20 going to close listening socket: 4
Fri Mar 12 16:53:20 Listener on port 27017 aborted
Fri Mar 12 16:53:20 going to close listening socket: 5
Fri Mar 12 16:53:20 Listener on port 28017 aborted.
Fri Mar 12 16:53:20 shutdown: going to flush oplog...
Fri Mar 12 16:53:20 shutdown: going to close sockets...
Fri Mar 12 16:53:20 shutdown: waiting for fs preallocator...
Fri Mar 12 16:53:20 shutdown: closing all files...
Fri Mar 12 16:53:20 closeAllFiles() finished
Fri Mar 12 16:53:20 shutdown: removing fs lock...
Fri Mar 12 16:53:20 dbexit: really exiting now
ERROR: Client::~Client _context should be NULL

Comment by Eliot Horowitz (Inactive) [ 12/Mar/10 ]

The first think is can you send the server log?
Hard to know what "crash" means.

Generated at Thu Feb 08 02:55:01 UTC 2024 using Jira 9.7.1#970001-sha1:2222b88b221c4928ef0de3161136cc90c8356a66.