ZABBIX Forums - Zabbix Help

Dedicated Zabbix Sql Server restarts

bep6vaku — Fri, 10 Apr 2026 21:26:11 GMT

I have a dedicated Zabbix SQl running MariaDB ver.15.1 on an Ubuntu 24.04 and when the memory utilization approached 78 or better precent, the SQL service restart. Not sure what is causing this or what settings need to ne adjusted. This is a physical server with 380GB of ram. I looked at the InnoDB settings but not sure if any need to be adjusted: Current version of Zabbix is 7.4.8

innodb_buffer_pool_size = 285G
innodb_log_file_size = 16G
innodb_log_buffer_size = 256M
innodb-file-per-table = 1
innodb_buffer_pool_instances = 16
innodb_old_blocks_time = 1000
innodb_stats_on_metadata = off
innodb-flush-method = O_DIRECT
innodb-flush-log-at-trx-commit = 2
innodb_read_io_threads = 64
innodb_write_io_threads = 64
innodb_io_capacity = 2000
innodb_io_capacity_max = 15000
connect_timeout = 10
table_open_cache_instances = 32
table_open_cache = 16000

Zabbix HA cluster strange behavior

olegus — Fri, 10 Apr 2026 19:32:23 GMT

Hi, we use Zabbix HA cluster functionality (2 servers) and every time we need to do maintenance on active server, we see ~5 min gap in charts. It looks like HA cluster does not work as expected even though a stand-by server correctly switched to Active according to SystemInfo page . We have fallover delay set to 1 min (which is probably too much) - but a gap is longer - up to 7min in some cases.
So when we bring the first server back again, we see all processes running high, struggling to process queues. All our proxies set both server's IPs correctly according to docs.
I wonder if it is expected behavior ? Would it help if we reduce failover delay down to say, 10 sec? What is the best practice here?

How we can discover host but just want discover selected interface only?

kronicklez — Fri, 10 Apr 2026 09:18:39 GMT

HI all,

By default when we discover host using snmp it will discover all the interface in that host. But i want it just discover selected interface only. How to do that?

Appreciate any help.

Best OS choice for a school district?

John_W — Thu, 09 Apr 2026 17:43:05 GMT

Our school district is considering moving from SolarWinds to Zabbix. I can see Zabbix can be installed on many Linux versions. I have read CentOS is "end of life". Beyond that, is there a "best" Linux variant that is both free and secure to install Zabbix on?

How to preserve SNMP Trap Operational data after new input

jdfranklin2 — Thu, 09 Apr 2026 17:38:10 GMT

Hello,

I have an SNMPv3 Trap set up on my host with triggers for higher severity traps. The problem is, my host is a Broadworks PBX switch with >4000 subscribers and it sends a very large number of SNMP traps, so when my trigger expression is met, the operational data on the problem gets quickly overwritten by new traps that arrive. The problem name doesn't get overwritten, but the operational data does.

The trap text is fairly custom and usually quite verbose (additionally, it does SNMP on port 8001 rather than the standard 161/162), so I have some dependent items to be able to do value mapping for some of the fields and for easier trigger configuration:

We'll focus on the trap.alarmName, snmptrap.fallback, and trap.severity. Here's the fallback setup:

trap.alarmName setup:

and the trap.severity setup

And then here's the trigger expression:

And here's an example. The alarm type bwSipServiceUnavailableReceived came through at 12:42pm local time and has a severity of 3, but it's being overwritten by a trap sent at 1:23pm local time with the alarm name bwSipMaxRetriesExceeded and severity of 1 (which doesn't trigger an alert)

Is it possible to make it so that operational data doesn't get overwritten?

And since I'm on the subject, partially, is there any way to throw away certain trap data based on regex before it gets past the "fallback" step? That might solve my problem, since we get ~2000 traps every hour, most of them low or informational severity. I would still like the operational data to not be overwritten, if possible.

Zabbix version 7.0.25 released?

hrrapozo — Thu, 09 Apr 2026 02:12:23 GMT

So today (4/8/26 - 3:30 pm HST) I was doing the weekly backup of my Zabbix MySQL database. Normally I shutdown the zabbix server, do the backup, perform any Ubuntu OS updates and then reboot the system.

There is also a Alert Trigger Action (email) dealing with Zabbix server. The email informed me that Zabbix server changed to 7.0.25 (it used to be 7.0.24). I do know that 7.0.25rc1 had been released on March 31, 2026 but there was no indication of the stable version of 7.0.25 was ready.

I am not complaining that 7.0.25 came out early but I also noticing that the System Information widget (shown below) says the server version is 7.0.25 but the latest release says 7.0.24.

Is a maintenance period the right way to stop Zabbix triggering alerts?

G0nz0uk — Wed, 08 Apr 2026 14:39:35 GMT

Hello,

We have maintenance on out systems between 2-3am every days and 12-1am for some other devices, what is the easiest way to stop any triggers being fired during this time please?

I've found Data Collection > Maintenance

I'm guessing I can create too here which stay active for some far off date and use tags?

Thanks

I can't clone a host because it contains a supposedly incompatible trigger

toxi — Tue, 07 Apr 2026 22:46:45 GMT

Hi! I can't clone a host because it contains a supposedly incompatible trigger. However, in Zabbix Server 7, it's impossible to select/configure/disable problematic triggers in the host configuration window (as was the case in Zabbix Server 5). So, I have to go somewhere (into the templates), disable the incompatible trigger, and only then try to clone the host.
Is this a system bug, or was it designed so that triggers in the host settings can't be edited until the host is actually cloned and a corresponding record for the new host is created in the database?

Server: Zabbix 7.0.24

Could you tell me how I can disable "hiding" the PSK key and PSK name?

toxi — Tue, 07 Apr 2026 22:35:24 GMT

Hi! Could you tell me how I can disable "hiding" the PSK key and PSK name in the host settings on Zabbix Server? It's inconvenient when the entries are hidden and the settings haven't been saved previously.

Server: Zabbix 7.0.24

Mass Update Host Names

jimmy.mccubbins — Tue, 07 Apr 2026 12:37:30 GMT

We have discovered over 1000 hosts across the network but they all came in as IP only for the host name.
What option do I have to update them with a Host Name to replace the IP Name without going 1 x 1?
Script examples would be great.

Error in creating an alert rule for EIGRP and DMVPN

bboddu — Tue, 07 Apr 2026 09:48:41 GMT

Hello All,

We have recently deployed a new Zabbix to monitor our network devices. The deployment was successful and we are trying to configure an alert rule to trigger an alert for EIGRP down and DMVPN down alerts, but we are unable to get succussed in configuring the alert rules. Could someone help in resolving or help in configuring the alert rules.
Currently we are monitoring the Cisco routers and Switches and Firewalls

Zabbix version: 7.0.23
Also, could someone help in sharing the default password for root
username root
password: ?

Regards,
Bharath Kumar

Best practice of working with psk.key

Aer0 — Mon, 06 Apr 2026 07:15:19 GMT

Hello everyone!

I'm working with both systems, Windows and Linux.

Both systems has Agent2 with PSK Encryption setting.

Zabbix has recomendation: "Set access rights to PSK file - it must be readable only by Zabbix user" and we did it for Linux system, but what about Windows?

Agent 2 service started with System access rights, do we need to left only "System Read" setting for that file?

What is your Best practice for that on Windows?

Thank you.

Difference in Discovery Action Behavior Between Zabbix 5.0 and 7.0

rody — Mon, 06 Apr 2026 02:36:03 GMT

Hello,
We are using Zabbix discovery actions to detect newly launched servers created by AWS Auto Scaling.
After migrating from Zabbix 5.0 to 7.0, we noticed that a discovery action that used to work correctly no longer triggers. It appears that there is a difference in the behavior of discovery actions between Zabbix 5.0 and 7.0.

Environment / Usage

Detecting newly launched servers via AWS Auto Scaling
Using Discovery Actions for detection
Discovery condition is based on SNMP agent checks

Observed Behavior

It seems that the handling of “Discovery time” when performing SNMP agent discovery differs between Zabbix 5.0 and Zabbix 7.0.
Regarding the behavior when re‑discovering already discovered devices
(devices shown as grayed out):

Zabbix 5.0:
When a discovered device is scanned again, the discovery time is reset.
Zabbix 7.0:
When a discovered device is scanned again, the discovery time is not reset.

Because of this difference, discovery actions that worked correctly in Zabbix 5.0 are not triggered in Zabbix 7.0.

Questions

We would like to confirm the following points:

Is this behavior in Zabbix 7.0 an intentional change in specification?
Is there any configuration option or workaround to reset the discovery time when re‑discovering an already discovered device?
Are there any known issues, limitations, or similar reports related to this behavior?

Any guidance or insight would be greatly appreciated.
Thank you.

If you want, I can also:

Shorten this for a quick forum post
Adjust the tone (more formal / more casual)
Add version numbers, screenshots references, or logs wording

Just tell me

Zabbix server: More than 75% used in the trends cache

G0nz0uk — Fri, 03 Apr 2026 16:35:26 GMT

Hello,

I'm pretty new to Zabbix and it's been running well until I added 30 Cisco switches today. I already had 20 in there. However now I've getting this:

Code:

Zabbix server: More than 75% used in the trends cache

On my Zabbix server I have never set this, should I and what would you use?

Code:

### Option: TrendCacheSize
# Size of trend write cache, in bytes.
# Shared memory size for storing trends data.
#
# Mandatory: no
# Range: 128K-16G
# Default:
# TrendCacheSize=4M

Here is the rest of that section:

#

Code:

## Option: HistoryCacheSize
#Size of history cache, in bytes.
#Shared memory size for storing history data.
#
# Mandatory: no
# Range: 128K-16G
# Default:
# HistoryCacheSize=16M

### Option: HistoryIndexCacheSize
#Size of history index cache, in bytes.
#Shared memory size for indexing history cache.
#
# Mandatory: no
# Range: 128K-16G
# Default:
# HistoryIndexCacheSize=4M

### Option: TrendCacheSize
#Size of trend write cache, in bytes.
#Shared memory size for storing trends data.
#
# Mandatory: no
# Range: 128K-16G
# Default:
# TrendCacheSize=4M

### Option: TrendFunctionCacheSize
#Size of trend function cache, in bytes.
#Shared memory size for caching calculated trend function data.
#
# Mandatory: no
# Range: 128K-2G
# Default:
# TrendFunctionCacheSize=4M

### Option: ValueCacheSize
#Size of history value cache, in bytes.
#Shared memory size for caching item history data requests.
#Setting to 0 disables value cache.
#
# Mandatory: no
# Range: 0,128K-64G
# Default:
ValueCacheSize=512M

I was wondering what tweaks or graphs I should be looking at to help?

This is what I have in my setup:
1 x Zabbix server (v7.4.8)
1 x Postgres TSDB server
3 x Proxy servers

On my proxy servers I have this below set, is this too low?

Code:

ProxyMemoryBufferSize=16M

All linux Ubuntu. CPU, Disk and Memory look ok.

Thanks

How to set tag on discovered trigger based on match discovery value vs macros

GIL — Fri, 03 Apr 2026 13:01:04 GMT

Hi 2 all. To properly summ SLA i need tags on services triggers based on Windows Template.
I have build some macros in template with names of services that are really critical for server to work:
{$SERVICES.CRITICAL} = ^(?:Winmgmt|W32Time|RpcSs|NTDS|Netlogon|MSTDC|Lanm anServer|Kdc|IsmServ|DNS|DHCPServer|DFSR|Dfs|ADWS) $
Now i need somehow assign tag or tag vs value based on discovered service name match macros expression.
How to do that?