Issue with monitoring service on MongoDB

Percona Monitoring and Management (PMM) PMM 1.x
1

Hello,

I am new to MongoDB and am setting up the PMM client on a 3-node cluster for the first time. On the first node I am working on, I got the linux metrics monitoring to work with no issues. But I am having trouble getting the mongodb:metrics service into a Running state.

When I run pmm-admin list, it shows the service at not running. I have tried numerous commands to get the service to start. Each time I run one of the below commands, I see a return message stating “OK, started mongodb:metrics service for <servername>”

These are the commands I have run so far:

pmm-admin add mongodb:metrics

pmm-admin add mongodb:metrics --uri mongodb://dbuser:@0.0.0.0:27017/admin --cluster rs0

pmm-admin add mongodb:metrics --uri mongodb://dbuser:@localhost:27017/admin --cluster rs0

I verified I can log into the Mongo instance using the dbusert account, which has admin privileges.

I have also run the start command numerous times, but am seeing the same results.

Any help with this would be appreciated- Thanks

2

can you try start service by the following line?

pmm-admin start mongodb:metrics
3

Thanks Mykola- I have tried that several times, but did not work. I just tried again and get this message: OK, started mongodb:metrics service

but the service still showing as not running.

I am seeing this in the mongodb metrics log:

E0413 16:23:36.979584 381 server_status.go:126] Failed to get server status.
panic: collector has no descriptors

Thanks,

Ron

4

Hi Ron,

I am trying to reproduce issue, what mongodb version do you use?

maybe I’ll prepare some custom patch with better error debug output on the next week…

5

Hi Mykola,

We are using version 3.2.11. Please let me know if you need any additional information

Ron

6

Hi Ron,

I started MongoDB from this rpm [url]https://repo.mongodb.org/yum/redhat/7/mongodb-org/3.2/x86_64/RPMS/mongodb-org-3.2.11-1.el7.x86_64.rpm[/url]
after that I created user

echo 'db.createUser({user: "someuser", pwd: "somepass", roles: [ { role: "userAdmin", db: "admin" } ] })' \
| mongo localhost:27017/admin

and checked credentials by this command

echo 'db.runCommand( { serverStatus: 1, recordStats: 0 } )' \
| mongo --username someuser --password somepass --host localhost --port 27017 --authenticationDatabase admin

and succesfully configured pmm by this command

pmm-admin add mongodb --cluster somecluster --uri someuser:somepass&#64;localhost:27017 some-instance

can you try these commands on your system?

7

Hi Mykola,

After running the above commands, I think the issue is with the privileges granted to the userAdmin role on the servers here. It looks like there are no privileges assigned to this role- I am not sure if this is by default or was done by the initial installer.

I have not been able to find a listing of the privileges needed for the pmm db user - do you have this info handy?

Thanks,

Ron

8

can you try command from mongodb_exporter documentation?

db.getSiblingDB("admin").createUser({
user: "mongodb_exporter",
pwd: "s3cr3tpassw0rd",
roles: [
{ role: "clusterMonitor", db: "admin" },
{ role: "read", db: "local" }
]
})

source [URL]GitHub - percona/mongodb_exporter: A Prometheus exporter for MongoDB including sharding, replication and storage engines

9

Hi Mykola- I actually went ahead and granted the privileges in your above command to my pmm user and it works now. Thank you very much for the assistance!

Ron

10

Ron, thank you

Tim created ticket about improving documentation - PMM-785
so it will improved in future :slight_smile:

11

Great- thanks again for the help!

Ron