[SERVER-36600] mongos failed on error - Failed to mlock: Unknown error Created: 12/Aug/18  Updated: 27/Oct/23  Resolved: 14/Aug/18

Status: Closed
Project: Core Server
Component/s: Sharding
Affects Version/s: 4.0.1
Fix Version/s: None

Type: Bug Priority: Major - P3
Reporter: jiancai.hao Assignee: Dmitry Agranat
Resolution: Works as Designed Votes: 0
Labels: None
Remaining Estimate: Not Specified
Time Spent: Not Specified
Original Estimate: Not Specified

Attachments: File mongos.log.tar.gz    
Operating System: ALL
Participants:

 Description   

I installed mongodb in sharding, but the mongos process was down suddenly not long after the start. There's error in the log file:

2018-08-11T22:11:18.010+0800 I NETWORK [listener] connection accepted from
2018-08-11T22:11:18.011+0800 I NETWORK [listener] connection accepted from 111.199.184.231:25765 #1609 (1539 connections now open)
2018-08-11T22:11:18.037+0800 F - [conn1275] Failed to mlock: Unknown error
2018-08-11T22:11:18.037+0800 F - [conn1275] Fatal Assertion 28832 at src/mongo/base/secure_allocator.cpp 246
2018-08-11T22:11:18.038+0800 F - [conn1275]

***aborting after fassert() failure

2018-08-11T22:11:18.051+0800 F - [conn1275] Got signal: 6 (Aborted).
0x561133b129b1 0x561133b11bc9 0x561133b120ad 0x7fe63b7ac5e0 0x7fe63b40f1f7 0x7fe63b4108e8 0x561132ef0e3f 0x561133aa6ca7 0x5611335b7559 0x5611335b83c8 0x5611332a9b50 0x5611332ab7d0 0x5611334e6789 0x56113303e4c0 0x56113303f927 0x561132f43b77 0x561132f6126a 0x561132f5bf27 0x561132f5f731 0x5611333ba7b2 0x561132f5a13f 0x561132f5d2d5 0x561132f5b667 0x561132f5bfad 0x561132f5f731 0x5611333bad15 0x561133a8dc44 0x7fe63b7a4e25 0x7fe63b4d234d

...



 Comments   
Comment by Dmitry Agranat [ 14/Aug/18 ]

Thank you for the clarification sword2000

Comment by jiancai.hao [ 13/Aug/18 ]

Thanks Dmitry,

Yes, this maybe one factor. I have set locked memory as the recommended value. At the same time, I found the direct cause – in one of my client, there is a NoSQLBooster for mongoDB process who is false death and is continously connecting to this mongos server, so that the mongos server exhausted all the locked memory in seconds.

I think this ticket can be closed now. Thanks.

Comment by Dmitry Agranat [ 12/Aug/18 ]

Hi sword2000,

The reported error suggests that the ulimit max locked memory setting is not set to our recommended value. Please adjust the ulimit settings for max locked memory and try starting the node again to see if that resolves the issue.

Thanks,
Dima

Generated at Thu Feb 08 04:43:35 UTC 2024 using Jira 9.7.1#970001-sha1:2222b88b221c4928ef0de3161136cc90c8356a66.