@baris only get this
Aug 28 16:39:27 hostname mongod: 2018-08-28T16:39:27.604+0800 I CONTROL [main] Automatically disabling TLS 1.0, to force-enable TLS 1.0 specify --sslDisabledProtocols 'none'
MongoDB crashes sometimes
We are running nodeBB on a testserver at the moment. Sometimes (maybe once a week or less) the mongod seems to crash. The service is nonresponding and so nodeBB also goes down. I don't see anything that helps us inside the logs. nodeBB only tells me that it cannot connect to the database and mongod just shows when I restarted the service.
I would like to investigate this before we go productive. Is there any way to exactly see what causes this crashes? The server itself is a clean Ubuntu 16.04 LTS virtual server with nothing else running on it.
Take a look at
/var/log/syslogusually mongod gets killed because the system is OOM, if that's the case you should see it in the syslog. Add more swap or increase the RAM of your server.
$ free -h total used free shared buff/cache available Mem: 2.0G 236M 1.7G 1.2G 68M 1.8G Swap: 4.0G 183M 3.8G
This doesn't look like a memory issue could happen. Noone is using the server atm. I go on there alone. Don't see anything special inside syslog. I will activate some SNMP sensors and hopefully see what happens shortly before it crashes.
Thanks for your try to help Baris!
@Tekin: You should also check /var/log/mongodb/mongod.log. What is the output of "sudo service mongod status"?
BTW: What version of MongoDB are you using?
Dravere last edited by
How are you using MongoDB? Installed directly on the system or running for example in a docker container? I still would highly recommend to check if the MongoDB wasn't killed by the Linux OOM killer. MongoDB has a tendency to eat up all the available memory in your system. During testing for my own forum I ran into several issues with this.
Sorry for the late answer.
1.) MongoDB is installed directly.
2.) Installed is version 3.6.5.
3.) The log of mongod only says that it started up. Everything before this is missing. Don't ask me why.
4.) Didn't see anything special inside syslog.
What we did now is that we watch over the system memory via SNMP.
I know it has been a while, but what did your investigation find?