ZABBIX Forums - Zabbix Help

SSH - SFTP agentless check (Zabbix 7.4)

MikeP — Thu, 28 May 2026 11:11:31 GMT

Hello,

does anyone have an experience using SFTP subsystem with Zabbix ssh.run item that was announced to be available since 7.2?
SSH check on the host works just fine and results are returned as expected, using sftp from command line from Zabbix machine works as well.

However I was not able to find any example/explanation on use of SFTP subsystem, though it is mentioned in documentation https://www.zabbix.com/documentation...hecks#overview (ssh.run[SFTPBackup,,,,,sftp])

My natural assumption was that it should be possible to use genuine SFTP commands here like version / ls / get / bye / etc.
No matter the commands the item jut times out.

Does anyone have an experience/example with this item?

//br Mike

Zabbix Database Information

Manuel Deschambault — Wed, 27 May 2026 15:49:14 GMT

Zabbix server 7.4, Grafana server 12.3.0

I have a PostgreSQL 18 database under the hood, both for Zabbix and Grafana

I have a PostgreSQL datasource setup in Grafana using a read-only user I have setup in the Zabbix PostgreSQL database.

The problem I have is the 'host' table does not have an field containing the host name (at least in readable format), and I can't seem to find the other table that would have this information (that I could do a SQL join query on).

Is there a complete map of the Database available somewhere, or does someone have some exploration steps documented?

Email notification issue after recently updated sendgrid password.

beingomkardalvi — Wed, 27 May 2026 05:24:53 GMT

We have Zabbix deploved in a Docker container and recently updated the SendGrid password. Since then, Zabbix email notifications have stopped working, even after updating the new password in the media type configuration. I tested email sending manually from inside the container, and emails are being received successfully. However, when testing through the Zabbix media type, it fails with the following error:
"Timeout was reached: Connection timed out after 40002 milliseconds."
Any idea how can I resolve it.?

Using below command I am able to receive email when I am trying inside container.

curl -X POST "https://api.sendgrid.com/v3/mail/send" -H "Authorization: Bearer abc" -H "Content-Type: application/json" -d '{"personalizations":[{"to":[{"email":"abc@xyz.com"}],"subject":"Test Email from Zabbix"}],"from":{"email":"xyz@abc.com"},"content":[{"type":"text/plain","value":"Test message from Zabbix server"}]}'

Zabbix 8 stopped connecting to database following apt upgrade

spyro — Mon, 25 May 2026 23:42:55 GMT

Hi,

My zabbix-server has stopped connecting to the postgres database since updating to the current version.

The database is available only via unix sockets, and is running on a nonstandard "port" (5433, not 5432).

It appears to be ignoring the DBPort variable in the configuration file, wheras previously it would use this value to determine the correct unix socket path (which for postgres seems to incorporate the port number (even if you dont listen on tcp).

This was working perfectly prior to the update, and the database is still accessible from the command line.

Is there a known fix for this?

The log contains messages such as:

[Z3001] connection to database 'zabbix' failed: [0] connection to server on socket "/var/run/postgresql/.s.PGSQL.5432" failed: FATAL: password authentication failed for user "zabbix"

(note the incorrect socket path, it should be /var/run/postgresql/.s.PGSQL.5433)

Monitoring WebRTC live streaming servers with Zabbix

abhaydevresource — Fri, 22 May 2026 13:27:58 GMT

hi there, i am looking for recommendations for WebRTC live streaming monitoring using zabbix ? i have been searching just curious what templates or approaches others are using and i have seen some red5 live streaming platform which support WebRTC does anyone heard about it or tried ? or is there any suggestions on this request.

zabbix-server service failed to start

dango — Tue, 19 May 2026 15:55:57 GMT

Hi there, I am getting this error on my newly deploy zabbix.

Code:

zabbix-server.service: Referenced but unset environment variable evaluates to an empty string: MAINPID

OS is Ubuntu 26.04 LTS with zabbix version 7.4.10. This was fresh deployment with mysql and apache. It was working fine the 1st day when I build some templates in the portal. A couple of days later, I noticed browser state the service is not running.

Hosts randomly switching to "Unknown" while data collection still works

sparkly — Tue, 19 May 2026 09:01:24 GMT

Hello everyone,

I am currently facing a strange issue with Zabbix where the agent interface of hosts randomly switches to "Unknown".

What is confusing is that:

The hosts are still sending fresh data.
Triggers continue to fire correctly.
Metrics are updated normally.

So communication between proxy and agents still seems to work.

The issue can temporarily be resolved by:

Restarting the Zabbix proxy
Or disabling and re-enabling the affected hosts

Environment

The infrastructure is identical for all customers:

2 physical servers per customer
Around 7 hosts per customer in Zabbix
All hosts monitored through a proxy

Versions

Zabbix Server/Proxy version: 7.4.10
Zabbix Agent versions: 7.4.7 - 7.4.10
Zabbix agents versions: 7.2.12 - 7.4.10

At the beginning, the issue appeared randomly and usually affected all 7 hosts of a single customer at once.

However, during the last two days, the problem escalated:

All 46 monitored hosts became "Unknown" overnight.

Troubleshooting done

I checked both proxy logs and agent logs.
I could not find any related errors or warnings.
On Sunday, I configured the proxy service to restart automatically at 5:00
Yesterday and today at 08:00 all hosts were still / already unavailable/unknown.
According to the proxy status, the service had already been running for 3 hours, so the restart itself worked correctly.
After manually restarting the proxy again at 08:00, all hosts immediately became available again.

Has anyone experienced similar behavior or has ideas what could cause this?

Thanks in advance.

annoyances on problems.

EStreet706 — Mon, 18 May 2026 05:28:44 GMT

I have noted that when I am in areas like, say, problems. It does not show the total number found. It does have pages of 50, but nowhere is a total count listed. To do anything productive, it doesn't seem that intuitive.

For example, on one site I am working with, I installed Zabbix site-wide. On the server, about 2 years ago, there was a ransomware attack where I left several hundred readme text files in place. The scanner detects them, and Zabbix lists them as Disaster, which it should. However, when I tag all of them, I am unable to close them as resolved. The only way I can see to do that is to first reclassify them as something other than Disaster, then retag them again, and then close it. Should be a painless way to do that.

Also, using 8.0 pre-release.

logrt or log?

tcweb — Fri, 15 May 2026 20:02:41 GMT

We have an application that writes to a log file "log.0". When log.0 reaches a configured number of lines, it is renamed to "log.1", and a new "log.0" is created. If there was already a log.1, it woudl be renamed to log.2, etc, etc.
This way, the app can keep N number of log files. (we usuallly keep 10 or 20). Note: I did not come up with this, it's a legacy application.

This being said, I would THINK that I could use the log[] function, since the log I wish to monitor is always "log.0". But I've been seeing posts elsewhere that say the logrt[] should be used because log[] could get confused when the log copy/switch happens.
The Zabbix documentation doesn't cover this sort of example.

As a follow-on question, is it advisable to set a polling interval of 3s for a log or logrt function? Our log is very busy, we may write hundreds of lines per second.

Another follow-on question: I know that I can have multiple log or logrt items on the same file. But are there any concerns with doing this? Performance?

Thanks in advance for any guidance.

-Tom

Ubiquiti AirOS SNMP: Interface ath1 – High Error Rate (>2 for 5m)

rafaelpierotti — Thu, 14 May 2026 20:27:45 GMT

Hi everyone,

I’m having an issue with a trigger in my monitoring dashboard related to a Ubiquiti AirOS device.

The trigger “Interface ath1: High error rate (>2 for 5m)” is constantly being activated, and I’m trying to understand the real cause instead of simply increasing the macro threshold to hide the alert.

At the moment, the interface appears to be working normally, but the error counter keeps increasing and generating alerts frequently. I’m not sure if this could be related to:

Wireless interference
Signal quality issues
Hardware problems
Duplex/speed mismatch
High traffic load
SNMP polling behavior
Something specific in AirOS

Has anyone experienced the same issue before?

I’d like to properly troubleshoot and identify the root cause rather than just suppressing the alert. Any suggestions on what I should check first, or which metrics/logs are most useful for diagnosing this kind of problem?

Thanks in advance.

Getting a "recently closed problems" Table

Wolfsbane2k — Thu, 14 May 2026 09:02:05 GMT

Hi all!

I'm having some issues with trying to pull out a list of "recently closed problems" to present on a Zabbix dashboard, such that if something we've "known about" suddenly closes we can investigate why.

Ideally this would be organised by "length problem was open for", but I can't find out a way to get the "raw" data first.

This is primarily because we have got a bit "blind" to long list of going issues (some over a year that "suddenly" closed and we missed it) and I want to ensure that we catch the changes in the state in a tabluated form, allowing a rigorous approach.

Is this possible in Zabbix (7.4.10) natively?

Ta!

Template importing correctly through web interface but not through API

mmansfie — Wed, 13 May 2026 21:17:33 GMT

Hi!
I'm managing a Zabbix server via the API, and having trouble using the configuration import endpoint to load in templates. After some troubleshooting, I've simplified it down to this: I have a simple template (see attached) which imports correctly when using the web interface, but imports without its discovery rule when imported via the API. Below is the template, my API testing code, and screenshots of the results.

I'm on Zabbix server 7.0.25 (happy to try newer versions, but if this is a known bug I was unable to find it). I couldn't find anything in the Zabbix server logs about the attempted configuration, let alone any errors, even when the debug level was at maximum.

Here's the template YAML file:

Code:

zabbix_export:
  version: '7.0'
  template_groups:
    - uuid: a571c0d144b14fd4a87a9d9b2aa9fcd6
      name: Templates/Applications
  templates:
    - uuid: aa15021bf5e04380bfff52fbeda29a84
      template: 'NRPE by Zabbix agent 2 user parameters'
      name: 'NRPE by Zabbix agent 2 user parameters'
      groups:
        - name: Templates/Applications
      discovery_rules:
        - uuid: a7d2d2fd1d9045dbb9f020db7da32c20
          name: 'NRPE discovery'
          key: nrpe.discovery
          delay: 1h
          item_prototypes:
            - uuid: f53fed58623649efbe1e6eea83585ef0
              name: '{#NRPE_NAME}'
              key: 'nrpe.query[{#NRPE_SCRIPT}]'
              value_type: TEXT
              trends: '0'

The python code I'm using to test the API:

Code:

import zabbix_utils
import socket
import yaml

api = zabbix_utils.ZabbixAPI(url="https://" + socket.getfqdn())
api.login(user="Admin", password=[REDACTED])

# load the desired Zabbix configuration
with open("test.yaml","r") as config_file:
template = yaml.load(config_file, Loader=yaml.Loader)

source = yaml.dump(template)
message = { "format": "yaml", "rules": { "templates": { "createMissing": True, "updateExisting": True }}, "source": source}
print(message)
print(api.configuration.import_(message))

I've confirmed with the output of this that the YAML parsing is correct and the API endpoint returns good.

Here's a screenshot of the template when imported with the web interface:

And when imported with the API:

I've also confirmed that exporting it after importing it via the API similarly shows the discovery rule missing.

Would appreciate any help on this. Thanks!

Too many long running queries in Zabbix PG "SELECT h.* FROM history..."

g8a — Tue, 12 May 2026 07:46:54 GMT

Hello!
I'm experiencing a lot of long running queries on our zabbix postgresql, and they come from Grafana dashboards (it's the main assumption - actually the DB overload could be caused by Grafana Direct DB Connection bypassing the API and querying history tables directly. Reducing pm.max_children helps with RAM but does not stop the DB flood)

seconds_running | table_name | application_name | query_preview
-----------------+--------------+------------------+--------------------------------------------------------------
57249 | history_uint | | SELECT h.* FROM history_uint h WHERE h.itemid IN (XXXX)
57249 | history | | SELECT h.* FROM history h WHERE h.itemid IN (XXXX)

Tables affected are "history_uint" and "history" as you can see, and this is affecting our environment because Grafana dashboards never load information.

psql (PostgreSQL) 13.23
zabbix_server (Zabbix) 7.0.21

We have tried restarting, scaling the server ram, upgrading the pm.max_children from 50 to 180...

All of these still not affecting and queries will still take too much to run.

Any ideas how to troubleshoot this? this happens since last Azure maintenance (last weekend, they restarted the zabbix db)

Thanks in advance.

Zabbix Appliance ISO stuck at loop2 on physical machine (Lenovo ThinkCentre M710q)

hei_zabbix_user — Tue, 12 May 2026 03:09:09 GMT

Hi,

I'm trying to install Zabbix Appliance on a physical machine but the installation always hangs at the same point.

Machine specs:
- Model: Lenovo ThinkCentre M710q Tiny (10MRCT01WW)
- Storage: 2.5" SATA SSD (256GB)
- RAM: 16GB
- BIOS: Legacy mode, Secure Boot disabled, SATA configured as AHCI

What I've tried:
- Zabbix 7.4 and 7.0 LTS ISO (netinstall)
- MBR partition scheme, both ISO and DD write modes in Rufus
- Added kernel parameters: rd.multipath=0 rd.md=0 rd.dm=0 multipath=off
- Boot mode set to Legacy Only

The installation always hangs after this line with no further output:
loop2: detected capacity change from 0 to 34359738368

The download completes successfully (100%, 918MB), network connection is fine (1000 Mbps Full Duplex), and "Started cancel waiting for multipath siblings of sda" shows OK. But after loop2 mounts, the screen freezes completely with no keyboard input accepted.

No error messages are shown. The system just stops responding.

Has anyone encountered this issue on a physical machine? Any suggestions would be appreciated.

Thanks

Performance monitoring of Windows services

Neodemon — Mon, 11 May 2026 20:07:46 GMT

Hi, we have a virtual machines running a few windows services, some of them might be running using the same executable like:
Service1 using "c:\services\service.exe P1"
Service2 using "c:\services\service.exe P2"
where P1 and P2 are command line parameters.

I tweaked Windows Registry to have access to include PIDs in the performance counters (HKEY_LOCAL_MACHINE\SYSTEM\CurrentControlSet\Servi ces\PerfProc\Performance, ProcessNameFormat DWORD 2),
set up a discovery rule (see LLD_js.txt) that identifies specific service and saves some info including running PID and then
created an item prototype with a key like perf_counter_en["\Process({#SERVICE.INSTANCE})\Private Bytes"].

Total schema works successfully until servcie restart, as after restart it gets another PID, means another item gets created for the host and started collecting metrics into new item
Just wondering if someone faced the same issue and resolved it to have statisitics continue gathering for the same service after service restart

Regards, Dmytro