ZABBIX Forums - Zabbix Help

There is a way to change the database failure message?

galavichid — Mon, 05 Oct 2026 19:57:40 GMT

I have few Zabbix servers, all using same smtp relay for sending e-mail notifications. Last week, I had failure in one of them, but I need to check first which one isn't instead just see it from the e-mail I got.

There is a way to change the message so I can make it more specific?

The message:

MySQL database "zabbix" on "localhost" is not available: Can't connect to local MySQL server through socket '/var/run/mysqld/mysqld.sock' (2)

Thanks

How can I monitor an on-premises ESXi host using Zabbix Cloud?

jmmalbas — Fri, 02 Oct 2026 16:31:29 GMT

Hi,

We are an MSP provider and have Zabbix deployed in the cloud, running the latest version, 7.4.13.
We would like to monitor our clients’ ESXi hosts using our centralized Zabbix server. Is this possible?

It looks like Zabbix Agent 2 is not available for installation on the ESXi operating system. What would be the recommended approach for monitoring ESXi hosts with Zabbix, especially when the Zabbix server is hosted in the cloud and the ESXi hosts are located at different client sites?

Any recommendations or guidance would be greatly appreciated.

Thank you!

Windows Event Log Unique Triggered Problems

jwrober — Thu, 01 Oct 2026 17:48:21 GMT

Hi all, I'm new to Zabbix. I'm running a newly installed v7.4.15 setup on a CentOS Stream 10 Linux VM. I am wanting to use the windows agent to capture windows event logs. In the "Windows by Zabbix Agent Active" template I have created 3 new items with the following code:

Code:

Name: Application Event Log Events (Critical/Error/Warning)
Key: eventlog[Application,,"Critical|Error|Warning",,,,skip]

Name: Security Event Log Events (Critical/Error/Warning)
Key: eventlog[Security,,"Critical|Error|Warning",,,,skip]

Name: System Event Log Events (Critical/Error/Warning)
Key: eventlog[System,,"Critical|Error|Warning",,,,skip]

I then created three triggers to log problems on the console:

Code:

Name: Windows: Event Log: Application (Critical / Error / Warning)
Expression: length(last(/Windows by Zabbix agent active/eventlog[Application,,"Critical|Error|Warning",,,,skip]))>0
Ok Event Gen: None
Problem Event Gen Mode: Single
Allow Manual Close: Yes

Name: Windows: Event Log: Security (Critical / Error / Warning)
Expression: length(last(/Windows by Zabbix agent active/eventlog[Security,,"Critical|Error|Warning",,,,skip]))>0
Ok Event Gen: None
Problem Event Gen Mode: Single
Allow Manual Close: Yes

Name: Windows: Event Log: System (Critical / Error / Warning)
Expression: length(last(/Windows by Zabbix agent active/eventlog[System,,"Critical|Error|Warning",,,,skip]))>0
Ok Event Gen: None
Problem Event Gen Mode: Single
Allow Manual Close: Yes

The challenge I have is I am getting problems for new log events associated with closed/resolved problems. Here is a sample. The original event was for a DLL not have perfmon data. I acknowledged it with some update entries

and closed. Then today it "came back" with this new event that isn't a repeat of the first one.

Is there a way to change the trigger to add some kind of unique values to the problem so as I resolve things Zabbix finds, it doesn't bring back old problems as new?

Thanks in Advance

problems with SNMP Bulk Request - bits send/received in Dependent Item

daniel.ramirez — Thu, 01 Oct 2026 16:46:34 GMT

i have a rare problem with data collection from SNMP Bulk Request.
this is Master Item:

and this is dependent item:

the dependent item sometimes do not poll on the configured time (5 minutes in the master item) and this produces strange peaks in values:

value table:

in the last polling, it tooks 9 minutes to poll a new value, after that only took 1 minute and generate that strange value (400Gb aprox.)

what could it be causing this behavior?

PD: the proxy is stable with normal vps and no queued values.

Can I apply a template to multiple hosts I have already discovered?

John_W — Thu, 01 Oct 2026 15:18:08 GMT

I don't see a way to assign a Template in the Discovery Rule. I can manually add the template after discovery, but that can take up some time, depending how many hosts I discover.

zabbix connections to host in time_wait state

tcweb — Wed, 30 Sep 2026 15:20:34 GMT

I have noticed that some of our monitored hosts will show hundreds of connections to port 10050 in time_wait state. (via netstat).
In this case, these are windows servers. We use zabbix 7.4, but the agents may be 7.2 in some case.

I do not think that it's necessarily a problem, but is it "normal"? These hosts are busy (with many zabbix items being collected), but I don't think we are over-monitoring. resources look fine on the server.

I welcome any thoughts or feedback.

Thanks,

-Tom

How to have Hosts that are created by discovery rules to have the host name listed?

John_W — Tue, 29 Sep 2026 19:01:24 GMT

Since I am new to Zabbix, I have a lot of devices to discover and hosts to create. I have to assume there is a way to get host info populated during the discovery of devices. I haven't found a screen to make that happen. I was thinking there would be an OID provided to determine the host name of the device. I have experimented with creating hosts, and the ones I have created have the graphs, dashboards, and web columns greyed out. This is shown under Monitoring-> Hosts in the side menu. What step do I need to take to get those features working?

Zabbix trigger with time ? (only trigger in working hours)

lbm — Mon, 28 Sep 2026 11:21:10 GMT

I have an trigger which looks like this.
But whenever the time is near 09:00 and 17:00 (5mins before), the trigger starts flapping insanely for 5mins. How can this be?

I simply just want the trigger to alert me in busyness hours and not 24/7.

count(/server/replicationhealth,10m,"like","CRITICAL")>0
and time() >= 090000
and time() <= 170000

Trigger on usrpara

gerwinnl — Mon, 28 Sep 2026 09:40:13 GMT

We use active agent usrpara to monitor on windows. When everything works it reports 0 / 1.
But when the script timeout or for example the script is wrong and the info message says: "value of type "string....." We get no data in the last value check.

Is it possible to create a trigger on the item when last value is empty or when the info message has a error?

Zabbix Proxy 8.0beta2: missing field "items.query"

Toorms — Fri, 25 Sep 2026 06:00:28 GMT

Hello Community,

in the latest image from Zabbix Proxy 8.0beta2

Code:

zabbix/zabbix-proxy-sqlite3@sha256:5be6f49243477e2c33d2e532d70a491326e8e6b35079244ec9ee90caf0119b04
zabbix/zabbix-proxy-sqlite3:alpine-trunk

I've got the following problem, without any changes in configuration or something else.

Code:

foo-srv-zapr01 | 12:20260925:055422.045 [configuration syncer #1] received configuration data from server at "monitor.idacen.de", datalen 498616
foo-srv-zapr01 | 12:20260925:055422.047 [configuration syncer #1] cannot process received configuration data from server at "monitor.foobar.local": missing field "items.query"

Did anyone else got these kind of problems after the latest trunk updates?

For investigations, here are my compose.yml file

Code:

services:
Zabbix-Proxy:
container_name: mad-srv-zapr01
restart: always
image: zabbix/zabbix-proxy-sqlite3:alpine-trunk
# image: zabbix/zabbix-proxy-sqlite3@sha256:fd9a923846fe632bbb01ac8f5c1ffc8daca af02d6c5ae05062895fbe8644d323
environment:
ZBX_TLSCONNECT: psk
ZBX_TLSACCEPT: psk
ZBX_TLSPSKIDENTITY: psk-id
ZBX_HOSTNAME: foo-srv-zapr01
ZBX_SERVER_HOST: monitor.foobar.local
ZBX_TLSPSKFILE: /var/lib/zabbix/enc/zabbix_agentd.psk
ZBX_SERVER_PORT: 10051
ZBX_PASSIVESERVERS: 192.168.178.0/24,172.24.0.0/24
ZBX_ACTIVESERVERS: 192.168.178.0/24,172.24.0.0/24
volumes:
- ./data/zabbix_agentd.psk:/var/lib/zabbix/enc/zabbix_agentd.psk
- prxdatabase:/var/lib/zabbix/db_data
ports:
- 0.0.0.0:10051:10051
ulimits:
nofile:
soft: 65536
hard: 65536
networks:
static-network:
ipv4_address: 172.20.128.2

networks:
static-network:
ipam:
config:
- subnet: 172.20.0.0/16

volumes:
prxdatabase:

Cheers

Simple display widget with regex and override host abilities

rlongfield — Fri, 25 Sep 2026 02:04:07 GMT

I am looking for a widget that simply displays a value of an item. The honeycomb is almost exactly what I need except it's missing the Override Host option.
The Item Value is also close except I can't use Regex to massage the item value, for example the Printer model information for a Lexmark printer has a string of 26 characters after the model information. I'm sure that string means something to someone but all we really need to know is what model number the printer is. With a Honeycomb widget I can use a Regex to strip out the extra 26 characters.

So, I guess I'm looking for a widget that functions like the honeycomb but has the Override Host option. Does such a widget exist?

How wo implement a Global Away Mode for Email Notifications?

S1ardust — Thu, 24 Sep 2026 18:49:35 GMT

Hello,

I have implemented Zabbix 7.4 in my Home-Network and set it up.
To get alerts by email, I have selected the desired triggers and added the tag "SendMail" = "true" to them:

I defined the condition in the Action accordingly:

This is working fine, but I get too many emails during everyday use.
I would like to receive only some alerts when I am "away".

Therefore I thought about introducing an "AwayMode" tag, which I would add only to selected triggers, that already have the "SendMail" Tag:

I would NOT touch the other Triggers and set "AwayMode" = "false".

Ideally, I would like to have a single global switch that changes the notification behavior without having to modify individual triggers.

Conceptually/Programmatically, I would want to achieve something like this:

Code:

if (Global "AwayMode" -eq  "true")
{
    if ($SendMail -eq "true" and $AwayMode -eq "true") {
        Send E-Mail Alert
    }
}
else
{
    if ($SendMail -eq "true") {
        Send E-Mail Alert
    }
}

What would be the recommended way to implement this in Zabbix?

Thank you!

Replicate Solarwinds "top 10 Interfaces" list

tcweb — Thu, 24 Sep 2026 17:51:31 GMT

Many people in the US are looking to move away from Solarwinds to other tools like Zabbix. Solarwinds suite of tools has been a staple for network administrators for years. They have "top usage" charts and graphs that seem to be difficult to reproduce in Zabbix. (for me anyway!)

For example, they have a chart that shows "top switch ports by percent utilization":

This is across ALL switches and ports! The problem I see is that "Top Hosts" or "Top Items" in zabbix ends up creating something like this:

Since the items names to not match, it will not create a single list that I can sort in reverse order.

Is there a way to hack one of these widgets? One thing I thought of was to use a powershell script that calls the API to pull data from zabbix, and then pull it back into zabbix as CSV or JSON.

Am I missing something? Is there an easier solution?

Thank you.

-Tom

Grafana Zabbix Datasource cannot connect to Zabbix 7.4

Edmilson António — Thu, 24 Sep 2026 11:43:21 GMT

Environment

Grafana: 13.2.2
Zabbix Server: 7.4.x
Zabbix Plugin for Grafana: tested with 6.7.0 and 6.8.0
OS: Ubuntu Server 26.04 LTS

Problem Description
I am unable to connect the Grafana Zabbix datasource to a Zabbix 7.4 server.
The datasource test in Grafana returns:

Could not connect to given URL

In some cases, Grafana also displays:

basic auth is not supported for Zabbix v7.2 and later:

Verified that the Zabbix API is reachable:
http:///zabbix/api_jsonrpc.php
Mostrar mais linhas
Tested API authentication manually using curl:
curl -X POST \
-H "Content-Type: application/json-rpc" \
-d '{
"jsonrpc":"2.0",
"method":"user.login",
"params":{
"username":"",
"password":""
},
"id":1
}' \
http:///zabbix/api_jsonrpc.php
Mostrar mais linhas Authentication succeeds and returns a valid token.
Created a dedicated API token in Zabbix and verified it works with curl using:
Authorization: Bearer
Captured traffic with tcpdump and confirmed Grafana is sending requests to:
POST /zabbix/api_jsonrpc.php
The Zabbix server replies with:
HTTP/1.1 200 OK
Despite the successful API responses, Grafana still reports:
Could not connect to given URL

Question
Has anyone successfully connected Grafana 13.2.2 with the Zabbix datasource plugin (6.7.x or 6.8.x) to Zabbix 7.4 using API tokens?
Could this be a compatibility issue between:

Grafana 13.2.2
alexanderzobnin-zabbix-app 6.7/6.8
Zabbix 7.4 API authentication changes

A source to learn the very basics of OIDs and Keys in Zabbix?

John_W — Tue, 22 Sep 2026 17:35:56 GMT

Coming from Solarwinds, how Zabbix handles OIDs is different and there were no Keys to work with in SW. I've read the manual and watched videos, but it appears these topics aren't covered in the detail I need.

OIDs. Technically, the OID is just the numerical listing of a given OID, but Zabbix allows action terms like walk or get and puts the OID in brackets. It looks like you can also have a list of OIDs. I think I saw one example with macros. Perhaps that was in the Key field.

Keys. It appears there are predefined keys that have names with periods separating the portions of the name but there are some that do not. Can you create your own key using either syntax? Do the keys need to match the descriptions of the OIDs?

If there is a video that goes into the basics of OIDs and Keys that goes into the real basics of these two topics, I would really like to get a link.