ZABBIX Forums - Forums

Postgres path issue when trying to rebuild srpms on rocky 8

Jason — Fri, 10 Jul 2026 14:27:09 GMT

I've noticed an issue when trying to rebuild the srpms on rocky 8.

After a while I'm getting "configure: error: PostgreSQL library not found" despite the libraries for 16 from the official postgres repo installed.

The issue is the path of /usr/pgsql-16/bin doesn't seem to be being picked up. Could this be added to the list of locations it searches? I know it should be in the path. I've removed and re-added it a few times and it's not appearing in the path by default so probably more robust to have it look in those locations checking 14, 15, 16 or 17 if they exist.

LLD SNMP. Вопрос по SNMPINDEX из разных таблиц

Diesel315 — Fri, 10 Jul 2026 06:06:15 GMT

Добрый день!
Вопрос по SNMPINDEX из разных таблиц. Требуется консультация в понимании логики работы

Zabbix Server 7.4.12
Debian 12

Суть:
Есть базовый шаблон мониторинга SAN коммутаторов *Brocade FC by SNMP*. К сожалению он не совсем верный логически, так как "да" вроде бы автором и закладывалась логика обнаружения только значимых портов через фильтрацию:
- {$NET.IF.IFADMINSTATUS.NOT_MATCHES} :: ^2$ :: Ignore down(2) administrative status

И оно вроде как даже работает, но только для относительно новых версий моделей. Потому что только в новых моделях (новая версия FOS Brocade), если нет SFP Transceiver, то порт автоматически переходит в - down(2) administrative status
На старых версиях так не работает, по умолчанию все порты в up administrative status, если они лицензированы.
В итоге картина примерно такая:

Code:

{"{#SNMPINDEX}":"1073741824","{#IFOPERSTATUS}":"2" ,"{#IFADMINSTATUS}":"1","{#IFNAME}":"FC port 0/0","{#IFDESCR}":"FC port 0/0","{#IFTYPE}":"56"},
{"{#SNMPINDEX}":"1073741825","{#IFOPERSTATUS}":"2" ,"{#IFADMINSTATUS}":"1","{#IFNAME}":"FC port 0/1","{#IFDESCR}":"FC port 0/1","{#IFTYPE}":"56"},
...
{"{#SNMPINDEX}":"1073741835","{#IFOPERSTATUS}":"2" ,"{#IFADMINSTATUS}":"1","{#IFNAME}":"FC port 0/11","{#IFDESCR}":"FC port 0/11","{#IFTYPE}":"56"},
{"{#SNMPINDEX}":"1073741836","{#IFOPERSTATUS}":"2" ,"{#IFADMINSTATUS}":"2","{#IFNAME}":"FC port 0/12","{#IFDESCR}":"FC port 0/12","{#IFTYPE}":"56"},
{"{#SNMPINDEX}":"1073741837","{#IFOPERSTATUS}":"2" ,"{#IFADMINSTATUS}":"2","{#IFNAME}":"FC port 0/13","{#IFDESCR}":"FC port 0/13","{#IFTYPE}":"56"},
...
{"{#SNMPINDEX}":"1073741846","{#IFOPERSTATUS}":"2" ,"{#IFADMINSTATUS}":"2","{#IFNAME}":"FC port 0/22","{#IFDESCR}":"FC port 0/22","{#IFTYPE}":"56"},
{"{#SNMPINDEX}":"1073741847","{#IFOPERSTATUS}":"2" ,"{#IFADMINSTATUS}":"2","{#IFNAME}":"FC port 0/23","{#IFDESCR}":"FC port 0/23","{#IFTYPE}":"56"}]

Все первые 12 порт (лицензированные) в статусе UP ("{#IFADMINSTATUS}":"1"), соответственно LLD их находит и добавляет в мониторинг.

Изучал разные MIB базы и нашел, что в базе SW-MIB есть нужная таблица swFCPortTable c параметром swFCPortPhyState (INTEGER { noCard ( 1 ) , noTransceiver ( 2 ) , laserFault ( 3 ) , noLight ( 4 ) , noSync ( 5 ) , inSync ( 6 ) , portFault ( 7 ) , diagFault ( 8 ) , lockRef ( 9 ) , validating ( 10 ) , invalidModule ( 11 ) , noSigDet ( 14 ) , unknown ( 255 ) } ) по которому как раз можно было бы отфильтровать. Только вот беда...

В этой таблице SNMPINDEX имеет значения - 1/2/3 и т.д.
А в нужной нам (.ifXEntry), из которой мы потом строим элементы данных SNMPINDEX имеет значения - 1073741824/1073741825/1073741826 и т.д.
И получается итоговый JSON имеет две разные таблицы сопоставления:

Code:

{"{#SNMPINDEX}":"1073741824","{#IFOPERSTATUS}":"2","{#IFNAME}":"FC port 0/0","{#IFDESCR}":"FC port 0/0","{#IFTYPE}":"56"},
{"{#SNMPINDEX}":"1073741825","{#IFOPERSTATUS}":"2","{#IFNAME}":"FC port 0/1","{#IFDESCR}":"FC port 0/1","{#IFTYPE}":"56"},
{"{#SNMPINDEX}":"1073741826","{#IFOPERSTATUS}":"2","{#IFNAME}":"FC port 0/2","{#IFDESCR}":"FC port 0/2","{#IFTYPE}":"56"},
{"{#SNMPINDEX}":"1073741827","{#IFOPERSTATUS}":"2","{#IFNAME}":"FC port 0/3","{#IFDESCR}":"FC port 0/3","{#IFTYPE}":"56"}


{"{#SNMPINDEX}":"1","{#IFPORTPHYSTATE}":"4"},
{"{#SNMPINDEX}":"2","{#IFPORTPHYSTATE}":"2"},
{"{#SNMPINDEX}":"3","{#IFPORTPHYSTATE}":"2"},
{"{#SNMPINDEX}":"4","{#IFPORTPHYSTATE}":"2"}

Вопрос:
Я правильно понимаю, что их скрестить/подружить/сопоставить друг с другом нельзя?
То есть используя фильтрацию {#IFPORTPHYSTATE} из одной таблицы (swFCPortTable), потом правильно формировать элементы данных, которые основываются на параметрах из другой таблицы (ifXEntry) при различных значениях - {#SNMPINDEX}

Discovery problem in 7.4.12

BiMW — Fri, 10 Jul 2026 05:56:52 GMT

After upgrading from 7.4.11 to 7.4.12, the discovery rule started detecting hosts that had been previously discovered and deleted. Is anyone else experiencing this issue?

After downgrade all is normal.

Suggestion Maintenance periods

sapsegundo — Thu, 09 Jul 2026 17:55:33 GMT

Ability to create maintenance by TAG, without requiring the host or host group.

Cannot update templates

techadmin — Thu, 09 Jul 2026 13:26:08 GMT

Versions:
zabbix-docker: 7.4.11
docker: 29.6.1
OS: Debian 13.5

We're unable to update several templates. F.e. the "Linux by Zabbix agent" (https://git.zabbix.com/projects/ZBX/...Ftags%2F7.4.12) with the diff

fails with the following error

I can't pinpoint specific log messages in 'docker compose logs zabbix-server'.
And I'm getting the same error for db/postgresql_agent2.

How can I make this work?

Best wishes,
Fabiano

Zabbix 7.4.12 and 8.0.0beta2 versions have been released!

Michael_Kammer — Thu, 09 Jul 2026 13:25:30 GMT

The latest Zabbix 7.4.12 and 8.0.0beta2 versions are out now!

Take a look at the new features and improvements:

7.4.12: https://www.zabbix.com/rn/rn7.4.12

8.0.0beta2: https://www.zabbix.com/rn/rn8.0.0beta2

All new Zabbix releases can be downloaded here: https://www.zabbix.com/download

Zabbix Queue backlog after radius failure

ericg — Wed, 08 Jul 2026 22:42:56 GMT

Zabbix Infrastructure
OS: Ubuntu 24.04
Zabbix: 7.4.10
Zabbix servers: 2 in HA
Zabbix Proxies: 5 in active mode, one is stand alone the other four are in proxy groups grouped in 2's
PostgreSQL: 18.3, w/timescaleDB 2.23.1
PGbouncer: 1.24.1
Compression: on after 7d

Issue experienced: Around 4am yesterday two of our Zabbix proxies stopped using their value cache and they started backlogging items. This all seems to coincide with our Radius server failing to authenticate requests during this time, the radius server has failed to auth at other times but not necessarily at this scale.
After Radius resumed working after a few moments, the two Zabbix proxies never caught up on the backlog of items you can seen in the queue of the screenshot I've included (both proxy 3 and 4 were about the same).
To resolve the backlog I Restarted the primary Zabbix server at 9:18 AM which failed over to the second Zabbix server, at this point certain metrics started recording again, but the queue stayed backlogged. At 1:12 PM I restarted both proxy 3 and 4 and that resolved their backlog of queued items.

Any suggestions for a bug this could be? or perhaps what I should see in the logs to help narrow down what the problem could be?

Sudden dops & spikes on interface graphs

t4h3r0 — Wed, 08 Jul 2026 11:09:49 GMT

Hi everyone,
New to the forum and relatively new to Zabbix.

I recently finished deploying a distributed Zabbix HA cluster with proxies.
Monitoring Cisco NCS-5500 & NCS-540X series devices using "Cisco IOS by SNMP", ZTE ZXCTN 6120H series devices with a cloned version of "Cisco IOS by SNMP" named "ZTE ROSNG by SNMP" and Huawei ATN 910 series devices with "Huawei VRP by SNMP" templates. Minimal minor changes have been made to tamplates.

On the Cisco and ZTE devices, interface traffic graphs are exhibiting a recurring sudden drops & spikes cycles or rubber-banding pattern sharp drops followed by immediate spikes every 30 to 40 minutes. The actual physical traffic is perfectly smooth.
This is only happening on the Cisco and less dramatically on ZTE hosts. The Huawei devices are rendering perfectly smooth, accurate graphs under the exact same network conditions. Increasing the polling interval from 1m to 2m reduced the amplitude of the spikes but did not solve the root issue.

I would genuinely appreciate any insights, configuration tweaks or pointers you might have. Thanks in advance.

Delay problem generation if some tag is configured

riBoon — Wed, 08 Jul 2026 10:25:12 GMT

Hi,
we are using zabbix version 7.

We have a template which checks for the availability of zabbix agent and other stuff.
This template is configured to important hosts and not so important hosts.

Is it possible to have the problem generation delayed by some time for the not so important hosts? We would like to configure a tag to the not so important hosts at host level.

Our goal is to get the alarms for the important hosts directly and the not so important hosts delayed (or no alarm, if they are ok after a while).

I thought about configuring this in actions but actions won't reduce the problem generation itself.

Avoid duplicates when managing hosts both via hostname only and network discovery?

mmansfie — Tue, 07 Jul 2026 16:50:52 GMT

Hi!

I'm operating a Zabbix site where we want to both manage the hosts via Puppet and run network discovery (in order to catch un-managed hosts)... but across the same range of hosts. We'd like to be able to specify hosts in Puppet by DNS name rather than IP address, but network discovery then creates duplicates as discovery actions only use the IP address to identify unique hosts. We could ameliorate this somewhat by making Puppet decline to overwrite the interface, but then we would still run into duplicates if Puppet created hosts before they would be network discovered.

We're thinking of some other methods:
- moving the discovered hosts into a separate view group and not even monitoring them, just using them as a list of hosts on the network to be compared to the list of host managed by Puppet
- using Puppet to determine if the host is managed by puppet, then fill in something into the agent metadata to prevent network discovery actions from firing
I'm not stoked by either of these methods. Do people have other methods to use to fix this issue?

I'm running this on Zabbix 7.0 LTS but if newer features would help this (I haven't found any) I'd be happy to try upgrading.

Thanks!
Mac

Zabbix 6.0.47 and 7.0.28 stable versions are out now!

Michael_Kammer — Tue, 07 Jul 2026 13:50:15 GMT

The latest Zabbix 6.0.47 and 7.0.28 stable versions have been released!

Take a look at the new features and improvements:

6.0.47: https://www.zabbix.com/rn/rn6.0.47

7.0.28: https://www.zabbix.com/rn/rn7.0.28

All new Zabbix releases can be downloaded here: https://www.zabbix.com/download

IIS by Zabbix agent active - Appool is not in Running state, trigger after 3 minutes

PeterBa — Tue, 07 Jul 2026 13:19:22 GMT

Hello everyone,

I'm in a pickle with IIS by Zabbix agent active - IIS: Application pool {#APPPOOL} is not in Running state.

Due to network / security complications I'm not able to let Zabbix know that a website in IIS under maintenance. Our desired workaround is to not trigger the IIS: Application pool {#APPPOOL} is not in Running state prototype trigger until the application pool is not in running state (3) for longer then 3 minutes.

It doesn't matter how we adjust the trigger. It either doesn't see the application pool go down at all, or the trigger fires right away.

Our latest attempt which also failed;
Problem expression

Code:

min(/IIS by Zabbix agent active/perf_counter_en["\APP_POOL_WAS({#APPPOOL})\Current Application Pool State"],3m)<>3
and {$IIS.APPPOOL.MONITORED:"{#APPPOOL}"}=1

Recovery expression

Code:

last(/IIS by Zabbix agent active/perf_counter_en["\APP_POOL_WAS({#APPPOOL})\Current Application Pool State"])=3

Is there a way to fire the trigger only if the application pool is not in running state for more then 3 minutes?

I'm using Zabbix 7.4.11

How notification design Zabbix RBAC when users/groups are managed via Entra SAML JIT?

Rudlafik — Tue, 07 Jul 2026 10:24:06 GMT

Hello,

I have a question about optimizing notifications when users and user groups (UG) are managed via SAML (Entra).

Before implementing SAML, I used a straightforward approach: I created users, assigned them to user groups, and then attached notifications to those groups. I stopped assigning notifications directly to users a long time ago.

With SAML, the situation has become more complicated:

Everything is managed automatically by SAML (Entra).
I have 10 SAML groups mapped to corresponding Zabbix user groups, each with a specific role derived from the three base roles.

The problem arises with notifications:

I have around 30 notifications, each tied to a different user group. However, users created via SAML JIT provisioning cannot be manually added to Zabbix user groups. Does this mean I need to create another 30 groups in Entra and effectively delegate the entire RBAC model to the O365/Azure team?

From this, it seems the RBAC model must follow these rules:

When multiple SAML mappings apply, Zabbix evaluates the base role type first. The higher base role wins. If roles share the same base type, the more restrictive (weaker) role is selected.
All baseline users without notifications will have the ZBX ReadOnly role.
All NOTIFY groups should map to the same neutral role, ideally the lowest usable one (ZBX ReadOnly).
SAML controls both the assignment to base Zabbix groups/roles and to NOTIFY groups/roles, ensuring conflict-free permission management.

I currently see two possible approaches:

Option A: Everyone sees everything

Metrics are not considered sensitive within the company, so all users can see everything. The open question is how to handle external users — I have not received clear guidance from management.
Notification groups with users assigned in Entra would then effectively act as pure Exchange distribution lists.

Option B: NOTIFY groups also define visibility of monitored objects and metrics

Visibility of hosts is determined by Zabbix user groups mapped from SAML groups (e.g., SAML IIS → ZBX IIS).
However, this significantly increases the number of groups from a permissions perspective. For example, for IIS I would need SAML groups like: SAML IIS Admin, SAML IIS Operator L2, SAML IIS Operator L1, SAML IIS ReadOnly — and similarly for MS SQL, RabbitMQ, LAN/WAN, etc.

How are you handling this in your environments?

Notifikace, uživatelské skupiny, role v prostředí s nasezeným SAML(Entra)

Rudlafik — Tue, 07 Jul 2026 09:04:36 GMT

Ahoj,
Mám dotaz jak optimalizovat notifikace když jsou uživatelé a user group (UG) řízeny SAML(Entra).
Před zavedením jsem pracoval způsobem, že jsem měl uživatele přiřadil je do UG a té přiřadil notifikace. Přiřazovaní notifikací na uživatele jsem opustil už dávno.
Se SAML se situace zkomplikovala.
- Vše si řídí automat SAML(Entra).
- Mám 10 SAML G ke každé ZBX UG a k té zase specifickou roli vycházející ze základních 3 rolí.

Problém nastává s notifikacemi:
Mám 30 notifikací a každá na jinou UG. Uživatele generovaného SAML JIT nelze ručně přidávat do ZBX UG. Pak tedy v Entra založit dalších 30 skupin a skutečně celý RBAC hodit na kluky z o365/Azure?

Z celého mi vychází, že RBAC model musí být postaven na těchto pravidlech:
- Při souběhu více SAML mapování Zabbix nejdřív vyhodnocuje základní typ role. Vyšší základní typ vyhrává. Pokud jsou role stejného základního typu, vybere se restriktivnější/slabší role.
- Všichni základní uživatelé bez notifikací budou mít roli ZBX ReadOnly
- Všechny NOTIFY skupiny mapovat na stejnou neutrální roli, ideálně tu nejnižší použitelnou, ZBX ReadOnly.
- SAML řídí rozdělení, jak do základních ZBX skupin/rolí tak do NOTIFY skupin/rolí. Tímto je zajištěno bezkonfliktní řízení přístupových Oprávnění

A tady mám dvě varianty:
Varianta A všichni vidí všechno:
- měřené veličiny nejsou v rámci firmy tajné a všichni vidí vše. Problém je co externisti. Na to jsem odpověď od vedení nedostal.
- pak notify skupiny s přiřazenými uživateli v Entra fungují jako čistě distribuční skupiny Exchange.

Varianta B NOTIFY skupiny jsou i skupinami pohledu na monitorované objekty a metriky:
- Informace o tom jaké hosty uvidí je součástí ZBX UG mapované na SAML G př: SAML IIS -> ZBX IIS
Je tu však poté nárůst skupin z pohledu oprávnění v samotné aplikaci. Pak pro IIS musím mít SAML skuipnu: SAML IIS Admin, SAML IIS Operátor L2, SAML IIS Operátor L1, SAML IIS ReadOnly obdobně MS SQL, Rabbit, LAN/WAN atp.

Jak tuhle problematiku řešíte vy ve vašich prostředích?

KeeneticOS Template by SNMP

NicKerrr — Mon, 06 Jul 2026 14:08:36 GMT

Hi everyone, I recently created a template for monitoring Keenetic devices. I’m looking for volunteers with different OS versions and hardware models to help test it.

https://github.com/NicKerrr/zabbix-t...eticOS_by_SNMP