Zabbix Server Syncing Issues and Downtime Concerns

[baweryasin]

Hello everyone,

I've been encountering some recurring issues with my Zabbix server lately and would appreciate any insights or advice on how to address them. Here's a rundown of the problems I'm facing:

1. **Irregular Syncing of Trend and History Data**: Normally, the syncing of trend and history data occurs daily around 17:00 after regular maintenance tasks. However, following a significant insert command, the server occasionally halts these processes but resumes after a few minutes. Strangely, this month, the server has initiated similar syncing processes around 00:00 on three separate occasions. After these episodes, the server stops as expected but remains offline for 10-12 hours before restarting.

2. **Performance Impact of Insert Commands**: Large insert commands seem to disrupt regular server operations, causing temporary halts in data syncing and sometimes prolonged downtime.

I've checked the configuration files and ensured that the parameters related to housekeeping and data syncing are set appropriately. However, I'm still puzzled by these irregularities and their impact on server uptime and performance.

If anyone has encountered similar issues or has suggestions on optimizing Zabbix server performance under heavy data load scenarios, your input would be greatly appreciated.

Thank you in advance for your help!



Here the log entries:
2767426:20240627:225724.050 Got signal [signal:15(SIGTERM),sender_pid:2856301,sender_uid:0 ,reason:0]. Exiting ...
2767427:20240627:225724.241 HA manager has been paused
2767452:20240627:225724.242 syncing history data in progress...
2767452:20240627:225724.572 syncing history data... 100.000000%
2767452:20240627:225724.572 syncing history data done
2767427:20240627:225725.647 HA manager has been stopped
2767426:20240627:225725.653 syncing trend data...
2767426:20240627:225728.718 slow query: 3.035438 sec, "insert into trends

inserts

2767426:20240627:225735.190 syncing trend data done
2767426:20240627:225735.194 Zabbix Server stopped. Zabbix 6.0.28 (revision 1f9d541d29a).
73635:20240628:112610.830 Starting Zabbix Server. Zabbix 6.0.28 (revision 1f9d541d29a).
73635:20240628:112610.832 ****** Enabled features ******
73635:20240628:112610.832 SNMP monitoring: YES
73635:20240628:112610.832 IPMI monitoring: YES
73635:20240628:112610.832 Web monitoring: YES
73635:20240628:112610.832 VMware monitoring: YES

[Markku]

I'm not sure I understand your question.

2767426:20240627:225724.050 Got signal [signal:15(SIGTERM),sender_pid:2856301,sender_uid:0 ,reason:0]. Exiting ...

means that a process with ID of 2856301 sent the Zabbix server a SIGTERM signal, i.e. someone asked the Zabbix server service to stop. Did you stop it to demonstrate a slow query when syncing the data, or are you asking why the server was stopped at 22:57:24 in the first place?

Markku

[baweryasin]

In first place im asking why the zabbix server stopped at 22:57 and did not start again after a few minutes (what it should), and is restarting after a few hours.

[Markku]

Well, if someone/something specifically stopped the Zabbix server service, it won't start automatically (at least in systemd environments). Try to find out who stopped it.

Markku