High utilization of LLD Worker internal process issue periodically

[andyway85]

Hello,

I am having issues with high utilization of LLD Workers internal process.
These are the syntoms:

1) The lld worker internal process get up to 100%




2) On the zabbix server logs I see this kind of logs. Correct me if i'm wrong, this has to do with low history syncer value??




3) The Zabbix Server and Proxy Server present high Disk read/write request response

I would like to understand first if LLD worker internal process has to do with the ammount of low level discovery rules that I have for several devices, and there is an option on the zabbix_server.conf that has 2 StartLLDProcessors by default, if I top it a bit, like 5 or 10 instead of 2, would this benefit this situation I am experiencing?
### Option: StartLLDProcessors
# Number of pre-forked instances of low level discovery processors.
#
# Mandatory: no
# Range: 1-100
# Default:
# StartLLDProcessors=2

Thanks for your help

Best regards,

Andres

[NgRox]

Hi,
Basically, Zabbix has 1 process called LLD Manager which is responsible for managing discovery rule operations, the LLD Manager dynamically sends LLD transactions to workers (LLD workers/processes)

Therefore, if you have a large amount of high-level discovery, the LLD Manager will manage the sending of grants to LLD workers, if you have few LLD processes (workers) they will be overloaded... so the explanation is that.

Answering your question about the StartLLDProcessors parameter, the answer is YES, if you increase its help. Now it's time to understand if the amount of LLD occurring in your environment is normal and lacks resources or if there is some wrong configuration in your LLD that is discovering things that were not meant to be discovered and are causing great use of this process.