[SERVER-9329] Invalid access at address - Segmentation fault Created: 11/Apr/13  Updated: 05/Mar/15  Resolved: 15/Apr/13

Status: Closed
Project: Core Server
Component/s: Stability
Affects Version/s: 2.4.1
Fix Version/s: None

Type: Bug Priority: Major - P3
Reporter: Brent Miller Assignee: Andy Schwerin
Resolution: Duplicate Votes: 0
Labels: None
Remaining Estimate: Not Specified
Time Spent: Not Specified
Original Estimate: Not Specified
Environment:

Ubuntu 12.04.02, 48 Gb RAM, dedicated server.


Issue Links:
Duplicate
duplicates SERVER-9014 Mongod and mongos crash induced by ma... Closed
Operating System: Linux
Steps To Reproduce:

We seem to be getting this anytime the server is under high cpu/iowait.

Participants:

 Description   

We're seeing Seg Faults when under high load. The first time this happened was when we were loading data from one collection to another. This caused the primary to crash and by the time we caught it had gotten too far behind the oplog on the other servers in the replica set. So, we dropped everything in the data dir and allowed it to resync. Now we're seeing the same "Invalid access at address" during the resync.

The log file shows:
Wed Apr 10 14:17:13.529 [conn8] authenticate db: apm

{ authenticate: 1, user: "apmui", nonce: "ccb12593a3cab52a", key: "58f35b791cc2537221649704a50caee6" }

Wed Apr 10 14:17:13.529 [conn9] authenticate db: apm

{ authenticate: 1, user: "apmui", nonce: "ccb12593a3cab52a", key: "58f35b791cc2537221649704a50caee6" }

Wed Apr 10 14:17:13.529 [conn8] auth: couldn't find user apmui@apm, apm.system.users
Wed Apr 10 14:17:13.529 [conn9] auth: couldn't find user apmui@apm, apm.system.users
Wed Apr 10 14:17:18.559 Wed Apr 10 14:17:18.559 Invalid access at address: 0x7f5de016fff0 from thread: conn9
Invalid access at address: 0x7f5de0371ff0 from thread: conn8

Wed Apr 10 14:17:18.586 Got signal: 11 (Segmentation fault).

Wed Apr 10 14:17:18.586 Got signal: 11 (Segmentation fault).



 Comments   
Comment by Andy Schwerin [ 15/Apr/13 ]

Duplicate of SERVER-9014.

Comment by Brent Miller [ 15/Apr/13 ]

We upgraded to 2.4.2-rc0 over the weekend and have been unable to reproduce the seg fault. Looks like SERVER-9014 was indeed the culprit.

Thanks for the help.

Comment by Andy Schwerin [ 12/Apr/13 ]

Thanks for the report. From MMS, I notice that over the last day your server has been accepting on average 3 new connections per second. If this is typical, you may be experiencing SERVER-9014, which is resolved in 2.4.2-rc0. Can you re-test with the release candidate, to see if this resolves your issue? The release candidate is available on the downloads page at mongodb.org.

Generated at Thu Feb 08 03:20:05 UTC 2024 using Jira 9.7.1#970001-sha1:2222b88b221c4928ef0de3161136cc90c8356a66.