Zabbix Active checks fail

[semiraue]

I'm moving my existing zabbix agents from passive to active monitoring and installing zabbix 4.0 server from the scratch. I followed the comments on config file as it is and manged to get testing machine and few other machines to work. I used agent active items and discovery rules and network interface auto discovery also works perfectly.

But when I try to config some other VM with the same config file it wont work perfectly. sometime in the latest data pane I can see information but it's not updating correctly. and Zabbix agent active ping always warning that host unreachable for 5 mins. please find my config file below.

##centos_________________________
PidFile=/var/run/zabbix/zabbix_agentd.pid
LogFile=/var/log/zabbix/zabbix_agentd.log
LogFileSize=0
DebugLevel=3
Server=xx.xxx.xx.xx
StartAgents=0
ServerActive=xx.xxx.xx.xx
Hostname=centos_jad11_xen
RefreshActiveChecks=60
Include=/etc/zabbix/zabbix_agentd.d/*.conf


since my zabbix server on different network (cloud) I cannot use simple passive checks. active checks are the only way to configure. I managed to setup my testing debian box even behind the NAT with this same config. but vm's with public IP's getting failed. And there is nothing unusual in agent logs

13205:20181023:110708.748 Zabbix Agent stopped. Zabbix 4.0.0 (revision 85308).
14407:20181023:111049.467 Starting Zabbix Agent [centos_jad11_xen]. Zabbix 4.0.0 (revision 85308).
14407:20181023:111049.468 **** Enabled features ****
14407:20181023:111049.468 IPv6 support: YES
14407:20181023:111049.468 TLS support: YES
14407:20181023:111049.468 **************************
14407:20181023:111049.468 using configuration file: /etc/zabbix/zabbix_agentd.conf
14407:20181023:111049.468 agent #0 started [main process]
14408:20181023:111049.468 agent #1 started [collector]
14410:20181023:111049.469 agent #2 started [active checks #1]

what are the other facts should I check for proper active monitoring ?

[zux]

are you absolutely sure, that the ServerActive has the right ip address?
does the host named "centos_jad11_xen" actually has active agent items in zabbix server?
are there any errors in server log, about this host?

[semiraue]

Yes Server-active is correct and host-name centos_jad11_xen have all zabbix agent active items on the items list. server logs has no entries matched to the hostname. In Queue window I have records that delayed by 10 mins from this host. but other hosts working fine

[kloczek]

I'm moving my existing zabbix agents from passive to active monitoring and installing zabbix 4.0 server from the scratch. I followed the comments on config file as it is and manged to get testing machine and few other machines to work. I used agent active items and discovery rules and network interface auto discovery also works perfectly.

Did you change on all those hosts all items type from "zabbix agent to "zabbix (active) agent"?

[semiraue]

yes I change all the items to zabbix agent active. not single item exist with zabbix agent.

[kloczek]

In what you've posted initially everything is OK.

[semiraue]

Yes that's what. with the same config some agents working perfectly fine. And I noticed in the latest data pane all the metrics from affected servers updated with like 30 mins delay.

[semiraue]

Hi does the timezone of server that I installed agent make any difference ? because I can see the last update time is the same time as guest server time. Previously I used passive check with the same server and no issues

[kloczek]

Yes that's what. with the same config some agents working perfectly fine. And I noticed in the latest data pane all the metrics from affected servers updated with like 30 mins delay.

This is because you have 30min interval to push new configuration.

[semiraue]

This is because you have 30min interval to push new configuration.

how can I reduce that time ? need to do it from agent or server ? please explain

[zux]

Manually - restart zabbix agent.
Or change this option:
RefreshActiveChecks =

[semiraue]

As I identified issue is different time zones. when logging data to server agent use remote machine time zone instead of actual server time zone. The latest data timestamp always displaying the remote machine time and agent ping item triggers because of that. i'm not quite sure why it logging data based on agent timestamp rather than actual timestamp that data arrives to the server. is there any way to fix this ?

[semiraue]

Manually - restart zabbix agent.
Or change this option:
RefreshActiveChecks =

As I mentioned in my question RefreshActiveChecks always set to 1 minute

RefreshActiveChecks=60

[semiraue]

As I identified issue is different time zones. when logging data to server agent use remote machine time zone instead of actual server time zone. The latest data timestamp always displaying the remote machine time and agent ping item triggers because of that. i'm not quite sure why it logging data based on agent timestamp rather than actual timestamp that data arrives to the server. is there any way to fix this ?

Hello can someone explain why this is happening ?? is there any way to resolve this rather than agent and server on the same time zone ?