SNMP timeout in Zabbix

[Rishabh Raj Singh]

Hello,
We are using Zabbix version 4.2.
We have setup Zabbix Monitoring through SNMP for our DHCP/DNS servers in different security layers.
We are using Zabbix environment as follows -
1 Mgmt Server
Proxies in each security layer
SNMP agent on host machine is sending data to Zabbix Proxies, and It it sending data to Zabbix Mgmt server and database.
We are using SNMP Timeout for 30s in Zabbix
There are some DNS/DHCP servers in the external security layer. We are having some issues with monitoring of these servers.
There is data loss on zabbix. We assume that packets are dieing before coming to Zabbix proxy.
and we are not getting reply back from snmp in 30 s timeout
Attaching Graph screenshots.
We have tried to change the timeout value on zabbix conf file to 60s. It was not possible.
We have tried to retreive data by increasing number of retries through script. still data is dieing as there is timeout for script as well.
Let me know if there is any way to increase the timeout and retry via snmp
Please advise us on this topic. Please provide us if there is solution to this problem.

[Rishabh Raj Singh]

We have tried both with using bulk requests and without bulk requests, we are getting similar graphs in both cases, a slight difference of small dots and small lines, but data is breaking

[Rishabh Raj Singh]

We need to increase timeout and retries on zabbix, please help in this prospective

[NilsA]

Hi,
we've run into the same issue before with some network devices. In our case, they simply didn't answer every time, resulting in the SNMP interface flapping and lots of missing data in our graphs.
The fix for us was to set unreachable and unavailability delays in the proxy config to the lowest value possible (1). This fixed about 95% of our data loss.

You can test this by manually getting information with snmpget on your proxy or server. Try it a few times and see if the device always answers.
Of course this increases the work load significantly if you're having trouble with a lot of devices.
At the very least it doesn't seem like a Zabbix-problem to me.

[cyber]

Are you using bulk request or not? I seem to remember having some similar timeout cases and removing bulk requests helped... It will increase load on proxy, tho... It needs to do lot more queries that way...