[ZBX-4374] Intermittent incorrect display of triggers Created: 2011 Nov 22  Updated: 2017 May 30  Resolved: 2015 Feb 16

Status: Closed
Project: ZABBIX BUGS AND ISSUES
Component/s: Frontend (F), Server (S)
Affects Version/s: 1.8.8
Fix Version/s: None

Type: Incident report Priority: Major
Reporter: Dmitrij Gavrilenko Assignee: Unassigned
Resolution: Cannot Reproduce Votes: 0
Labels: dependencies, trigger
Remaining Estimate: Not Specified
Time Spent: Not Specified
Original Estimate: Not Specified

Attachments: File 1. Monitoring-Triggers.JPG     File 2. Latest_data - PortStaus.JPG     File 3. Latest_data - History.JPG     File 4. Item_config.JPG     File 5. Trigger_config - ifOperStatus.JPG     File 6. Trigger_config - ifAdminStatus (dep).JPG     File 7. MySQL - ItemStatus.JPG     File 8. MySQL - TiggerStatus.JPG     File 9. MySQL - history_uint.JPG    

 Description   

google translate
-------------------------------------------------------------
After migration to version 1.8.8 began noticing intermittent incorrect display of flip-flops.
The logic of monitoring:
Hosts work in zabbix-proxy. Monitor the status of hosts by means of ladders. Pitfall traps is on the zabbix-proxy. All items / triggers are registered on the hosts locally, without templates (templates initially connected, and then detached from the persistence of the data).

Description of the file attachments and problems:
1. Monitoring-Triggers:
operating window «Monitoring» -> «Triggers», which can be seen, the trigger has the status of «PROBLEM», though the port is already «UP (1) '(display in the trigger occurs through« Value mapped »). Time the trigger in 11:11:00.

2. Latest_data - PortStaus:
window «Monitoring» -> «Latest data», the port is "administrative" and "operational." It is evident that both elements were 11:11:14 value "1", ie ports have moved into a state of «UP».

3. Latest_data - History
History shows the values ??for the "operational" state of the port. We see that in 11:11:00 port moved to the state «Down» (data item has a value of "0"), which corresponds to the time of the trigger, and then, at 11:11:14, a data item has the value "1".
At 11:11:00 and 11:11:14 element that collects "administrative" state of the port receiving the same data "0" and "1" respectively, which indicates proper operation handler traps and pass values ??to the proxy on the main server.

4. Item_config
given the current configuration of the element that collects "operational" state of the port «E5».

5. Trigger_config - ifOperStatus:
description of the configuration of the trigger for the "operational" state of the port «E5», for which there is an incorrect indication. Here comment from the trigger is removed.

6. Trigger_config - ifAdminStatus (dep):
description of the configuration of the trigger for "administrative" status of the port «E5», from which the dependence of the trigger for the "operational" state of the port.

7. MySQL - ItemStatus:
current state of the element in a MySQL database for "operational" state of the port «E5», after he obtained a value of "1": field «lastvalue» = 1.

8. MySQL - TiggerStatus:
current state of the trigger in a MySQL database for "operational" state of the port «E5», after the element was set to "1": field «value» = 1.

WHY?? After all, the trigger condition (see attachment "5. Trigger_config - ifOperStatus») is satisfied! And it shows in the screenshots:
"1. Monitoring-Triggers », in the line:" Interface E5 (**********): Up (1) »
"2. Latest_data - PortStaus », the last value of" 1 "
"3. Latest_data - History », the last value of" 1 "

9. MySQL - history_uint:
in any case given the history of the state for a gathering "operational" state of the port «E5».

PS: If you disable the dependency for a trigger that has already worked in my case, "Interface E5 (**********): Up (1)", the indication alarm message continues.
PSS: in file «triggers.inc.php» I made small changes that would patch up malfunctioning macro

{ITEM.VALUE} for WEB interface using «Valuemapped» (see ZBX-3423).
PSSS: while describing the problem, there was a decline and recovery of more than one port, but only with other equipment, and without the administrative block, as described here. Port fell to 3 minutes (in the situation described here for 14sek), and then got up and moved to the trigger status «OK». Configuration items and triggers for this equipment are identical to those described here above.

original
-------------------------------------------------------------
????? ???????? ?? ?????? 1.8.8 ???? ???????? ????????????? ???????????? ??????????? ?????????.
?????? ???????????:
????? ???????? ????? zabbix-????. ?????????? ????????? ?????? ?????????????? ? ??????? ??????. ??????? ?????? ????? ?? zabbix-??. ??? ????????/?????? ????????? ?? ?????? ??????, ??? ???????? (????? ??????? ????????????, ? ????? ????????????? ? ??????????? ??????).

???????? ????????? ?????? ? ????????:
1. Monitoring-Triggers:
??????? ???? «Monitoring» -> «Triggers», ? ??????? ???, ??????? ????? ?????? «PROBLEM», ???? ????????? ????? ??? «UP(1)» (????????? ? ???????? ?????????? ????? «Value mapped»). ????? ???????????? ???????? ? 11:11:00.

2. Latest_data - PortStaus:
???? «Monitoring» -> «Latest data», ????????? ????? «??????????????» ? «??????????». ?????, ??? ? 11:11:14 ??? ???????? ???????? ???????? «1», ?.?. ????? ??????? ? ????????? «UP».

3. Latest_data - History
???????? ??????? ???????? ??? «???????????» ????????? ?????. ?????, ??? ? 11:11:00 ???? ??????? ? ????????? «Down» (????? ?????? ??????? ???????? «0»), ??? ????????????? ??????? ???????????? ????????, ? ?????, ? 11:11:14, ??????? ?????? ??????? ???????? «1».
? 11:11:00 ? 11:11:14 ?????, ?????????? «??????????????» ????????? ????? ??????? ????? ?? ??????, «0» ? «1» ??????????????, ??? ??????????????? ? ?????????? ?????? ??????????? ?????? ? ???????? ???????? ? ?????? ?? ???????? ??????.

4. Item_config
????????????? ??????? ???????????? ??????, ??????????? «??????????» ????????? ????? «E5».

5. Trigger_config - ifOperStatus:
???????? ???????????? ???????? ??? «?????????????» ????????? ????? «E5», ?? ???????? ??????????? ???????????? ?????????. ????? ??????????? ? ???????? ??????.

6. Trigger_config - ifAdminStatus (dep):
???????? ???????????? ???????? ??? «???????????????» ????????? ????? «E5», ?? ???????? ??????????? ??????????? ??? ???????? ? «??????????» ?????????? ?????.

7. MySQL - ItemStatus:
??????? ????????? ???????? ? ?? MySQL ??? «?????????????» ????????? ????? «E5», ????? ????, ??? ?? ??????? ???????? «1»: ???? «lastvalue» = 1.

8. MySQL - TiggerStatus:
??????? ????????? ???????? ? ?? MySQL ??? «?????????????» ????????? ????? «E5», ????? ????, ??? ??????? ??????? ???????? «1»: ???? «value» = 1.

????????? ???? ??????? ???????? (??. ???????? «5. Trigger_config - ifOperStatus») ?????????!!! ? ??? ????? ?? ??????????:
«1. Monitoring-Triggers», ? ????: «??????? E5(**********): Up (1)»
«2. Latest_data - PortStaus», ????????? ???????? "1"
«3. Latest_data - History», ????????? ???????? "1"


9. MySQL - history_uint:
?? ?????? ?????? ????????????? ????????? ??????? ??? ???????? ??????????? «????????????» ????????? ????? «E5».

P.S.: ???? ????????? ??????????? ??? ??????, ??????? ??? ????????, ? ???? ?????? «??????? E5(**********): Up (1)», ?? ????????? ?????????? ????????? ????????????.
P.S.S.: ? ???? «triggers.inc.php» ???? ??????? ????????? ?????????, ??? ?? ????????? ???????????? ?????? ??????? {ITEM.VALUE}

??? ??? ?????????? ??? ????????????? «Valuemapped» (??. ZBX-3423).
P.S.S.S.: ???? ???????? ??????, ????????? ??????? ? ?????????????? ??? ?????? ?????, ?? ?????? ?? ?????? ????????????, ? ??? ???????????????? ??????????, ??? ??????? ?????. ???? ???? ?? 3 ??????(? ????????? ????? ???????? ?? 14?), ????? ???? ???????? ? ??????? ??????? ? ?????? «OK». ???????????? ????????? ? ????????? ?? ???? ???????????? ????????? ???, ??????? ???????? ????? ????.

 Comments   
Comment by Dmitrij Gavrilenko [ 2011 Dec 07 ]

google translate :

After migration to version 1.8.9 the problem remains.

I think that it is linked to the fact that if the dependence of triggers, when a dependent element comes a new value, and the parent trigger (from which the dependence) has continued to state PROBLEM, a state-dependent trigger is ignored and, as a consequence He does not go into a state of OK. Although it has already received the item value, at which his status has already had to change. As a result, when the parent trigger changes state on OK, then appears on the screen with state-dependent trigger PROBLEM, despite the fact that the dependent element is the trigger sozherzhit value which is dependent trigger should move to OK status.
Accordingly, the equipment no longer comes with the operational state of the ladder and we have a trigger flip-flop with state podvisshy PROBLEM, but as the value of which is a number, saying that the port is raised.

PS: The situation I get when working with ladders of the IP-equipment, which in one package sends immediately the administrative and operational state of the port. Maybe the problem is the lack of processing of nanoseconds, and perhaps you can make changes to the processing logic based triggers.

----------------------------------------------------------------------------------------
После миграции на версию 1.8.9 проблема осталась.

Мне кажется, что она связана с тем, что если установлена зависимость триггеров, то, когда в зависимый элемент приходит новое значение, а родительский триггер(от которого установлена зависимость) имеет по прежнему состояние PROBLEM, то состояние зависимого триггера не ОБРАБАТЫВАЕТСЯ и, как следствие, он не переходит в состояние ОК. Хотя его элемент уже получил значение, при котором его статус уже должен был поменяться. В результате, когда родительский триггер меняет свое состояние на OK, то на экране появляется зависимый триггер с состояние PROBLEM, при том, что элемент зависимого триггера уже сожержит значение, при котором зависимый триггер уже должен был перейти в статус ОК.
Соответственно, от оборудования больше не приходит трап с операционым состоянием триггера и мы имеем подвисший триггер с состоянием PROBLEM, но в качестве значения элемента которого находится число, говорящее о том, что порт поднят.

P.S.: Данная ситуация у меня появляется при работе с трапами от IP-оборудования, которое в одном пакете шлет сразу административное и операционное состояние порта. Может быть проблема связана с отсутствием обработки наносекунд, а может быть можно внести изменения в логику обработки зависимости триггеров.

Comment by richlv [ 2014 Jun 19 ]

some data collection problems were recently fixed. could you please try 2.2.4rc3 and see whether the problem is still present ?

Comment by Alexander Vladishev [ 2015 Feb 16 ]

I close the issue. Feel free to reopen if it is reproducible in latest versions of Zabbix.

Generated at Sun Apr 06 20:37:19 EEST 2025 using Jira 9.12.4#9120004-sha1:625303b708afdb767e17cb2838290c41888e9ff0.