SNMP extension V2 and bulk requests

[olivier_coliber]

Hi,

I have some snmp v2c devices to monitor.

I tryed to poll them with snmp generic extension.

In ruxitagent_extensionsmodule_xxx.log i found messages like this :

{"extensions":[{"id":"snmp","extension":"com.dynatrace.extension.snmp-generic-device","tasks":[{"task":"-7234474320797872862","configurationId":"750b27da-1633-3d8f-bd40-0c0d9afa1fe5","compositionStatus":"DEVICE_CONNECTION_ERROR","message":"DEVICE_CONNECTION_ERROR: (host=10.63.207.117): connection problem due to wrong community name or network problems.: Query caused a network error with the need to reconnect (host=10.63.207.117): read udp 10.117.72.54:36721->10.63.207.117:161: i/o timeout: GetBulk returned an error: request timeout (after 3 retries) [status code=27] [status code=38] [status code=38]; "}

But when i poll this device with snmpwalk from ActiveGate, it work's, I retrieve interfaces status... etc.

There is no configuration mistakes (same community, same IP, same ActiveGate).

Is this could be come from plugin GetBulk request? My device coud be not compliant with these requests? In this case ,someone knows how to force extension to use Get Next request?

Thanks

Olivier

[ChadTurner]

@olivier_coliber are you still having this issue?

[GardenGroom]

Hello there,

diggin' out that topic as i got similar problems:

activated the generic snmp devices plugin and added a configuration with a bunch of devices to monitor.

Running into errors as the log tells me the same things like Oliver described, bulk requests on a specific OID runs into timeout on a devices, which causes the devices from the whole configuration to get inactivated (the icons are greyed out and no data is available, sometimes it works, sometimes not).

My workaround was creating configuration per devices, so no other devices would be affected because one isn't reachable, but i'm sure it's not the way it was intended to use.

[Mike_L]

If it is working sometimes and sometimes not I’d open a support ticket for it if I were you. 

[GardenGroom]

Already done - seems to be a problem when you got more devices in a single configuration. Having only one device per configuration works better. Got the information it should be fixed in Q3/23.

[Mike_L]

By then we will automatically split up all endpoints within one monitoring configurations to different AGs in the group.

Until then, you can also try to increase the limit on the AG to high or (if you are on 265+) dedicated limits.

[Mike_L]

SNMP extensions always use GetBulk. Maybe try to execute such a request manually and see if it works on your device?