CentOS, Apache

Our specific use-case for this would be to use the {EVENT.TAGS} macro in a global script, - this would allow us to start a Windows service that had stopped (using a trigger action is not suitable, as there will be times when we don't want a service to automatically be restarted).

This feature would bring so many benefits to us. This would allow us to start a Windows service that had stopped. In our case, instead of 1st Line support teams get in touch with 2nd Line support teams, they would have a means to start the services, because the don´t have access to servers. Using a trigger action is not suitable.

is there any prevision for release of the feature?

Tks

Hi,

same here it would be a great benefit for us also.

I added a similar request:

https://support.zabbix.com/browse/ZBXNEXT-6029

Background we need an option to "manually" trigger our CreateIncident.pl script which creates an Incident from the selected event. Therfore to get data from the event to handover to Service Manager like user who opened the incident (thankfully is available since 5.0.2 and it's working) and we need the event data. We need at least the {EVENT.ID} to query additional data from the RestAPI to enrich the Incident within global script to get additional data (occurence, message, system name etc.) from the event.

Please can you check if it is possible to implement?

Many thanks!

We currently evaluating if Zabbix could be a replacement for our current Monitoring solution (BMC Patrol). The most pain point we got is that it is not possible to handover any type of MACROS or ITEM values ({{$HOST.HOST}:<KEY>}) to global script as argument which can be triggered manually by our operators in a event/problem context. Like described above, the {EVENT.ID} will be at least the argument we need to query additional data from the specific event where the operator executes the remote script from to query additional information by RestAPI from this specific event for our Incident creation intefrace.

Would be awesome if you can provide this in a future release.

Many thanks.

Would be great to get it integrated. Thanks in advance.

Hello guys,

is a very important feature, do you know when it will be available? maybe in version 5.4 or 6.0? =)

I hope you will get something with ZBXNEXT-6368 in Zabbix 5.4.

thank you, solved in version 5.4

Related: ZBXNEXT-3229

ZBXNEXT-1906 is distantly related as well. Although, it's about trigger expressions.
However, when finding a good algorithm for action conditions, this is then possibly applicable to trigger expressions as well

Merged fix in https://support.zabbix.com/browse/ZBXNEXT-3588

Merged fix in https://support.zabbix.com/browse/ZBXNEXT-3588

Merged fix in https://support.zabbix.com/browse/ZBXNEXT-3588

Merged fix in https://support.zabbix.com/browse/ZBXNEXT-3588

Merged fix in ZBXNEXT-3588

Fixed in development branch svn://svn.zabbix.com/branches/dev/ZBXNEXT-3588-trunk

(1) [S] In functions check_condition_event_tag() check_condition_event_tag_value() and a lot more the header says they might return successful or not value while in reality they always return SUCCEED.

vso RESOLVED in r64752

s.paskevics CLOSED

Attached optimized and non optimized pictures for following action, 1 item and 1000 triggers + multiple event generation.

A Maintenance status not in maintenance
B Application like test
C Template = test_template_level2
D Trigger = test_template_level2: trap_trig1000
E Host = test server
F Host group = test

#!/bin/bash
COUNTER=0
while [  $COUNTER -lt 200 ]; do
./bin/zabbix_sender -z 127.0.0.1 -p 10051 -s "test server" -k trap_template1 -o 0
let COUNTER=COUNTER+1
done

(2) [S] In lots of functions to check conditions (e. g. check_condition_event_tag()) there is similar code snippet:

               ret = FAIL;

               if (CONDITION_OPERATOR_NOT_EQUAL == condition->operator ||
                               CONDITION_OPERATOR_NOT_LIKE == condition->operator)
               {
                       ret_continue = SUCCEED;
               }
               else
                       ret_continue = FAIL;
 
               ret = ret_continue;

Looks like the first assignment is redundant.

vso RESOLVED in r64779

<dimir> CLOSED

(3) Some headers have - before IN/OUT and is_escalation_event() is missing header

vso RESOLVED in r64933

s.paskevics CLOSED

(4) Redundant distinct check in internal events host condition

Distinct is used with primary key.

"select distinct itemid"

vso RESOLVED in r64935

glebs.ivanovskis Well spotted!
CLOSED

(5) There are places where cache might be used instead of optimized SQL queries for example templates

see In DCdump_htmpls() implemented in ZBXNEXT-3659

 24393:20170331:094031.407 In DCdump_htmpls()
 24393:20170331:094031.407 hostid:1
 24393:20170331:094031.408   templateid:7
 24393:20170331:094031.408 hostid:10
 24393:20170331:094031.408   templateid:7
 24393:20170331:094031.408 End of DCdump_htmpls()

This should be better done as separate task probably, but must be properly investigated.

So it might be worth removing template processing optimization part of this task and deciding to implement it through cache as a separate task.

vso WON'T FIX, not possible to use only cache here.

(6) Short-circuit evaluation.
When this task is finished, must not forget to create new task and investigate possibility of proper short-circuit evaluation.
vso created ZBXNEXT-3960, WON'T FIX.

(7) event_match_condition() seems to be quite small and simple function with a lot of debug logging ("In ..." and "End of ..."). Probably worth moving logging to check_action_conditions(), will be just one line per event per condition.

vso RESOLVED in r68473

glebs.ivanovskis Nice!
CLOSED

(8) check_action_condition() is a symbol with external visibility. Would be nice to add zbx_ prefix.

vso RESOLVED in r68474

glebs.ivanovskis Good!
CLOSED

(9) I had a belief that DB_* types should match database schema as close as possible, thus the idea to store

zbx_vector_uint64_t	objectids;

in DB_CONDITION seems strange to me. I think it decreases readability of the code using DB_CONDITION. Maybe I'm wrong, let's get some clarification from wiper who started the trend with

zbx_vector_ptr_t	tags;

in DB_EVENT.

wiper ... who simply looked at DB_TRIGGER insider DB_EVENT and added more properties.

glebs.ivanovskis I've found the source of this movement. sasha said that it's better to rename such type.

vso RESOLVED in (26)

s.paskevics CLOSED

(10) Function prepare_actions_eval() does not belong to dbconfig.c, neither it accesses configuration cache nor uses configuration-specific definitions. This file is huge on its own, let's not pollute it with unrelated code.

vso RESOLVED in r68484,68486

glebs.ivanovskis Great! Minor style fixes in r68641. Let's remove zbx_action_eval_free() from dbconfig.c too. REOPENED

vso RESOLVED in r68938

s.paskevics CLOSED

(11) In the wake of (9) I made some improvements to db.h and its surroundings. Please review my changes in r68265, r68270, r68273, r68274, r68276.

vso CLOSED

(12) I couldn't say by the name of check_events_conditions() what this function does and it's purpose "check if multiple events matches multiple conditions" wasn't helpful either. Same story with conditions_vectors_create() and conditions_vectors_destroy(). I tried to inline them and was quite pleased with result. Reader would need to look at their code anyway to understand what they are doing. Take a look at r68638. If you feel that process_actions() is too long, there are more logical ways to divide it into multiple functions.

vso yes, I agree CLOSED

(13) As I understand zbx_dc_get_actions_eval() simply copies data from shared memory into local memory of history syncer. We may consider optimization similar to zbx_dc_correlation_rules_get() (maybe as a subtask or independent task):

	/* The correlation rules are refreshed only if the sync timestamp   */
	/* does not match current configuration cache sync timestamp. This  */
	/* allows to locally cache the correlation rules.                   */
	if (config->sync_ts == rules->sync_ts)
	{
		UNLOCK_CACHE;
		return;
	}

glebs.ivanovskis BTW, probably no need to lock cache just for timestamp check, as vjaceslavs does in ZBXNEXT-3006.

vso CLOSED nice observation, created ZBXNEXT-3913

(14) Incorrect function name in comment for check_template_id_item_sql_alloc function.

vso RESOLVED in r68955

s.paskevics CLOSED

(15) I don't like is_escalation_event function, because after new changes (ZBXNEXT-18) the event can be escalated without checks specified in is_escalation_event function.

s.paskevics WON'T FIX, function is correct for current version.

(16) I propose to replacing if else statement with switch in check_discovery_condition function, as is done in check_trigger_condition, check_auto_registration_condition and check_internal_condition functions.

RESOLVED in r69288.

vso CLOSED

(17) Did some refactoring, please review r69288, r69289.

vso CLOSED

(18) Objectids that we do select on shall be stored for possibility to use in between.

vso RESOLVED in r69378

s.paskevics CLOSED

(19) Please check changes in r69379. Callbacks were deleted from check_object_hierarchy function.

vso CLOSED with small style fix in r69411

(20) When internal events are generated for trigger and for item then there is no guarantee that objectid will not be same for both. We must add object to the criteria.
RESOLVED in r69394

s.paskevics CLOSED

(25) Please check get_object_ids_internal() function refactoring. r69633 r69641

vso CLOSED

(26) There is no bulk selection in the check_acknowledged_condition function, but in practice, this function always process only one event.

s.paskevics This function can be called only from escalator. I added some changes in code r69676, 69687. RESOLVED.

vso CLOSED

(27) SQL error:

[Z3005] query failed: [1054] Unknown column 's.type' in 'field list' [select s.type from dservices s where s.dserviceid=1]

vso this was reported and fixed under ZBX-12299, WON'T FIX, we can merge fix from trunk

(28) Please check my minor changes in r69736.

vso CLOSED

Successfully tested.

Fixed in:

• pre-4.0.22rc1 3c0775329b5
• pre-5.0.2rc1 b47d63a55d4
• pre-5.2.0alpha1 (master) f6dc85c79b2

By stating to "[...] make a unified view [..]", is it meant to provide this view in addition or to replace the current views Triggers and Events, resp. to implement one view that supports each of these aspects?

While I can confirm that it is indeed confusing to new or non-power-users, it is pretty valuable to have these two aspects on issues. One aspect for the current state and another aspect for identifying what happened at a certain period in time.

A replacement view for Triggers and Events that provides several aspects sounds to be the best approach to me but is possibly also the most challenging one from a UX perspective.

However, if a good implementation design is found, that's to say one view that allows different kind of aspects with flexible but intuitive filter options, then this promises to become a huge improvement to users.

Btw, yet another aspect could be ZBXNEXT-2695.

The new view is supposed to support all functionality we already have under Monitoring->Triggers and Monitoring->Events along with additional filtering options and much better performance.

Available in development branch svn://svn.zabbix.com/branches/dev/ZBXNEXT-3201

(1) Changed translation strings:

Strings added:

• RESOLVED
• Recent problems
• Recovery time
• Show unacknowledged only

iivs CLOSED

Available in pre-3.1.0 (trunk) r61049.

(2) Related to ZBXNEXT-2163.

Administration -> Media types: "exec_params_count" can be removed.
Monitoring -> Problems, Monitoring -> Web: "page" input parameter must be validated more strictly

sasha RESOLVED in trunk@61073,61074

iivs CLOSED

(3) Links to event details is missing from problem view

sasha RESOLVED in trunk@61077

iivs CLOSED

(4) Broken "Host" column in Monitoring->Triggers view

sasha RESOLVED in trunk@61081

iivs CLOSED

(5) Bulk acknowledge

sasha RESOLVED in dev branch svn://svn.zabbix.com/branches/dev/ZBXNEXT-3201-1 r61346.

gunarspujats CLOSED

(6) [D] API documentation updated

• https://www.zabbix.com/documentation/3.2/manual/api/reference/problem
• https://www.zabbix.com/documentation/3.2/manual/api/reference/problem/object
• https://www.zabbix.com/documentation/3.2/manual/api/reference/problem/get
• https://www.zabbix.com/documentation/3.2/manual/api/reference/event/object
• https://www.zabbix.com/documentation/3.2/manual/api/reference/event/get
• https://www.zabbix.com/documentation/3.2/manual/api/changes_3.0_-_3.2

RESOLVED

gunarspujats CLOSED

(8) [F] Acknowledging problems, view switches to last page. So does the cancel.

sasha RESOLVED in trunk@r61145

iivs CLOSED

(9) [F] Problems page shows 8689+ records at the bottom, while other pages show only 1001+. Also API count returns 9849 records. Filter was reset, so shouldn't everything be selected?

sasha Problem view does not show old resolved problems. It depends on "Display OK triggers for" option.

sasha RESOLVED in trunk@r61143

iivs CLOSED

(10) [F] Sorting by severity doesn't change. Looks exactly like sorted by time.

alexei RESOLVED in trunk@r61120
sasha incorrect sorting by problem, host and severity. RESOLVED in trunk@r61141

iivs Looks good now.
CLOSED

(11) [F] As discussed, the extension of Monitoring -> Problems view from ZBXNEXT-3277 task is now moved here and I quote: "The view will be extended to have another column INFO placed after STATUS. The columns will contain an icon if problem was resolved by correlation."

sasha RESOLVED in r62098

Strings added:

• Correlation rule
• Resolved by correlation rule "%1$s".
• Resolved by correlation rule.
• Resolved by user "%1$s".
• Resolved by user.

gunarspujats CLOSED

(12) [A] Specification states that possible sorting columns for problem.get are eventid and clock, however objectid is implemented as well. Also there is no clock index on problem table.

sasha RESOLVED in dev branch svn://svn.zabbix.com/branches/dev/ZBXNEXT-3201-1 r61320.

gunarspujats CLOSED

(13) [F] Coding style and other minor issues:

• ZBase.php
  • L:371 type must be set to CRouter and parameter is not described.
• monitoring.problem.view:
  • L21: Missing space before the first line of code
• CControllerProblemView.php:
  • L24: unused variable
  • L89-102: commented code
• CProblem.php:
  • L25: We don't write this anymore.
  • L42: PHPdoc for function states that there is an option count? And itemids? Seems like the Problem API options are quite different for that get method.
  • L293: We don't write this anymore.

sasha RESOLVED in dev branch svn://svn.zabbix.com/branches/dev/ZBXNEXT-3201-1 r61316

gunarspujats CLOSED

(14) [F] sorting by priority in problem view does not work

alexei RESOLVED in revision 61120.

iivs How is this different from (10)?

sasha CLOSED as duplicate of (10)

(15) [F] after first in-line refresh all sorting links in table header become incorrect

sasha RESOLVED in dev branch svn://svn.zabbix.com/branches/dev/ZBXNEXT-3201-1 r61336.

gunarspujats CLOSED

(16) [F] Performance issue
Monitoring->Problems use a lot of memory. To see 50 problems need at least 512 RAM for PHP

(mod_fastcgi.c.2673) FastCGI-stderr: PHP Fatal error:  Allowed memory size of 134217728 bytes exhausted (tried to allocate 72 bytes) in /var/www/zbxnext_3274/include/classes/api/CRelationMap.php on line 77
(mod_fastcgi.c.2673) FastCGI-stderr: PHP Fatal error:  Allowed memory size of 134217728 bytes exhausted (tried to allocate 78 bytes) in /var/www/zbxnext_3274/include/classes/api/CApiService.php on line 286
(mod_fastcgi.c.2673) FastCGI-stderr: PHP Fatal error:  Allowed memory size of 134217728 bytes exhausted (tried to allocate 72 bytes) in /var/www/zbxnext_3274/include/classes/api/CApiService.php on line 286
(mod_fastcgi.c.2673) FastCGI-stderr: PHP Fatal error:  Allowed memory size of 134217728 bytes exhausted (tried to allocate 72 bytes) in /var/www/zbxnext_3274/include/classes/api/CApiService.php on line 286
(mod_fastcgi.c.2673) FastCGI-stderr: PHP Fatal error:  Allowed memory size of 134217728 bytes exhausted (tried to allocate 72 bytes) in /var/www/zbxnext_3274/include/classes/api/CApiService.php on line 286
(mod_fastcgi.c.2673) FastCGI-stderr: PHP Fatal error:  Allowed memory size of 134217728 bytes exhausted (tried to allocate 64 bytes) in 
(mod_fastcgi.c.2673) FastCGI-stderr: PHP Fatal error:  Allowed memory size of 134217728 bytes exhausted (tried to allocate 72 bytes) in /var/www/zbxnext_3274/include/classes/api/CRelationMap.php on line 48
(mod_fastcgi.c.2673) FastCGI-stderr: PHP Fatal error:  Allowed memory size of 134217728 bytes exhausted (tried to allocate 80 bytes) in /var/www/zbxnext_3274/include/db.inc.php on line 599
(mod_fastcgi.c.2673) FastCGI-stderr: PHP Fatal error:  Allowed memory size of 134217728 bytes exhausted (tried to allocate 85 bytes) in /var/www/zbxnext_3274/include/db.inc.php on line 599
(mod_fastcgi.c.2673) FastCGI-stderr: PHP Fatal error:  Allowed memory size of 134217728 bytes exhausted (tried to allocate 78 bytes) in /var/www/zbxnext_3274/include/classes/api/CRelationMap.php on line 73

Moved from ZBXNEXT-3277 (25)

sasha RESOLVED in dev branch svn://svn.zabbix.com/branches/dev/ZBXNEXT-3201-1 r61291:r61312

sandis.neilands It is better now however with 105k problems the "Problems" page still runs out of memory. This shouldn't happen since we show only 20 pages, e.g. 1000 problems by default.

gunarspujats CLOSED

(17) [F] Moved from ZBXNEXT-3277 (27)

sandis.neilands: What does the 'Action' column in 'Problems' page is supposed to show? Just the actions taken for problem event, problem and ok event?

sasha RESOLVED in r62041

Strings added:

• Actions on
• Failures

Strings deleted:

• Not sent

gunarspujats CLOSED

sasha added displaying of total number of actions in r62141

RESOLVED

gunarspujats CLOSED

(18) Different filter controls in Monitoring->Triggers, Monitoring->Overview and Monitoring->Problems

sasha RESOLVED in dev branch svn://svn.zabbix.com/branches/dev/ZBXNEXT-3201-1 r61335.

gunarspujats CLOSED

(19) [F] if sorting by host empty list, errors

    Undefined variable: triggers_hosts [zabbix.php:21 → require_once() → ZBase->run() → ZBase->processRequest() → CView->getOutput() → include() → CScreenProblem->get() in include/classes/screens/CScreenProblem.php:166]
    Invalid argument supplied for foreach() [zabbix.php:21 → require_once() → ZBase->run() → ZBase->processRequest() → CView->getOutput() → include() → CScreenProblem->get() in include/classes/screens/CScreenProblem.php:166]

picture no_data_sort_host.jpg

sasha RESOLVED in dev branch svn://svn.zabbix.com/branches/dev/ZBXNEXT-3201-1 r61348.
natalja.zabbix CLOSED

(20) [F] Based on specification "Displayed tag and tag value length is limited to N pixels". No limit to N pixels, now displaying all 255 symbols of tag and tag value
picture tag_lenght.png

sasha Already fixed in ZBXNEXT-3274 (18). CLOSED

(21) [F] no space between third tag and '...' , with space looks better

sasha RESOLVED in dev branch svn://svn.zabbix.com/branches/dev/ZBXNEXT-3201-1 r61436

gunarspujats CLOSED

svn://svn.zabbix.com/branches/dev/ZBXNEXT-3201-1 was merged to trunk r61447.

2nd phase was implemented in pre-3.2.0alpha2 r61801.

(22) 2nd phase translation strings:

Strings added:

• a time is expected

Strings deleted:

• DISCOVERED
• DOWN
• LOST
• Latest events
• UP
• Recent problem removed in r61447

oleg.egorov CLOSED

(23) changelog entry currently says :

improved event.get() and problem.get() methods

it might be worth specifying what was improved (performance ? something else ?)

sasha CLOSED as duplicate of (6)

richlv hmm, will the person updating the API docs remember to change the changelog entry ? maybe that's worth adding in that subissue ?

sasha You are right! FIXED in r64478

CLOSED

(24) [F] when use filtering options "Show:History", "Tags:any tag name and tag value" geting error

Error in query [SELECT e.eventid,e.objectid,e.clock,e.ns FROM events e WHERE e.source='0' AND e.object='0' AND EXISTS (SELECT NULL FROM event_tag et WHERE e.eventid=et.eventid AND et.tag='TagName' AND UPPER(pt.value) LIKE'%TESTVALUE%' ESCAPE '!') AND e.clock>='1408860125' AND e.clock<='1471932125' AND e.value='1' ORDER BY e.eventid DESC LIMIT 1001 OFFSET 0] [Unknown column 'pt.value' in 'where clause']

sasha RESOLVED in r61872

oleg.egorov CLOSED

(25) [F] Time bar issue in Monitoring->Problems if selected all period

sasha Cannot reproduce. Seems the bug is related to timebar. CLOSED

(26) [F] not renamed labels according specification in Administration->General->Trigger displaying options:
Display OK triggers for -> Display OK events for
On status change triggers blink for -> New PROBLEM events blink for

sasha it is decided to leave these labels without changes

CLOSED

(27) [F] according to the specification "Show events not older than (in days)" and "Max count of events per trigger to show" should be removed in Administration->General->GUI

sasha These fields already used in tr_status.php. Specification must be updated.

WON'T FIX

(28) [I] Missed upgrade patch, event_expire and event_show_max should be removed from config table.

sasha These fields already used in tr_status.php. Specification must be updated.

WON'T FIX

(29) [F] sorting by any type, change current page number
Steps to reproduce:
1. select second page
2. sort by severity
Result: page changed to third

oleg.egorov After bulk acknowledge happens same issue
I have page number 2
Period from 2016-03-13 22:14:40 till 2016-03-13 21:30:40
After bulk acknowledge happens redirect to page number 3

sasha RESOLVED in r61873

natalja.zabbix UI successfully tested

oleg.egorov CLOSED

(30) [A] Slow query

SQL (65.140725)

SELECT   e.eventid,e.objectid,e.clock,e.ns FROM events e WHERE e.source='0' AND e.object='0' AND (e.objectid BETWEEN '15353' AND '15398' OR e.objectid BETWEEN '15411' AND '15423' OR e.objectid BETWEEN '15493' AND '15505' OR e.objectid BETWEEN '16165' AND '16177' OR e.objectid BETWEEN '16180' AND '16189' OR e.objectid IN ('15440','15441','15442','15443','15543','15547','15548','15550','15551','15553','15561','15563','16162')) AND EXISTS (SELECT NULL FROM functions f,items i,hosts_groups hg,triggers t WHERE e.objectid=f.triggerid AND f.itemid=i.itemid AND i.hostid=hg.hostid AND (hg.groupid BETWEEN '4' AND '9' OR hg.groupid BETWEEN '16' AND '25' OR hg.groupid IN ('1','2','11','12','14','30','32','43','45','46','47','48')) AND i.hostid='10343' AND e.objectid=t.triggerid AND t.priority='5') AND e.acknowledged=0 AND e.clock>='1436821200' AND e.clock<='1457902800' AND e.value='1' ORDER BY e.eventid DESC LIMIT 1001 OFFSET 0

zabbix.php:21 → require_once() → ZBase->run() → ZBase->processRequest() → CView->getOutput() → include() → CScreenProblem->get() → CScreenProblem->getData() → CScreenProblem->getDataEvents() → CFrontendApiWrapper->get() → CApiWrapper->__call() → CFrontendApiWrapper->callMethod() → CApiWrapper->callMethod() → CFrontendApiWrapper->callClientMethod() → CLocalApiClient->callMethod() → call_user_func_array() → CEvent->get() → DBselect() in include\classes\api\services\CEvent.php:379

Debug info:

Total time: 69.799992
Total SQL time: 66.112774000004
SQL count: 2339 (selects: 1523 | executes: 816)
Peak memory usage: 63.25M
Memory limit: 3072M

Installation:

3,055,333 - events
2,167 - triggers

sasha I can't reproduce this issue.

SQL (0.232991): SELECT   e.eventid,e.objectid,e.clock,e.ns FROM events e WHERE e.source='0' AND e.object='0' AND e.objectid IN ('13666','13782','13784','13786','13788','13790','13792','13794','13796','13798','13800','13802','13804','13806','13808','13810','13812','13814','13816','13818','13820','13822','13824','13826','13828','13830','13832','13834','13836','13838','13840','13842','13844','13846','13848','13850','13852','13854','13856','13858','13860','13862','13864','13866','13868','13870','13872','13874','13876','13878') AND EXISTS (SELECT NULL FROM functions f,items i,hosts_groups hg,triggers t WHERE e.objectid=f.triggerid AND f.itemid=i.itemid AND i.hostid=hg.hostid AND (hg.groupid BETWEEN '4' AND '16' OR hg.groupid IN ('1','2')) AND i.hostid IN ('10084','10105') AND e.objectid=t.triggerid AND t.priority IN ('2','3','4','5')) AND e.acknowledged=0 AND e.clock>='1408909797' AND e.clock<='1471981797' AND e.value='1' ORDER BY e.eventid DESC LIMIT 1001 OFFSET 0
zabbix.php:21 → require_once() → ZBase->run() → ZBase->processRequest() → CView->getOutput() → include() → CScreenProblem->get() → CScreenProblem->getData() → CScreenProblem->getDataEvents() → CFrontendApiWrapper->get() → CApiWrapper->__call() → CFrontendApiWrapper->callMethod() → CApiWrapper->callMethod() → CFrontendApiWrapper->callClientMethod() → CLocalApiClient->callMethod() → call_user_func_array() → CEvent->get() → DBselect() in include/classes/api/services/CEvent.php:379

Debug info:

******************** Script profiler ********************
Total time: 0.298484
Total SQL time: 0.262083
SQL count: 58 (selects: 35 | executes: 23)
Peak memory usage: 6.75M
Memory limit: 512M

Installation:

1,135,455 - events
402 - triggers

sasha WON'T FIX

(31) [F] In Monitoring->Problems displaying event acknowledge errors.

Open in the new tab Event acknowledge , then try to submit space till page refreshed without error message.

sasha WON'T FIX

(32) [F] Triggers (tr_status.php) regression
Missed stime and period

In "Trigger" page select old, problem trigger, then on popup select "Problems" link... and there is no result, because it's was old issue.

sasha Already discussed with alexei. It was decided to show only recent problems from triggers popup menu.

WON'T FIX

(33) [F] macros not opened in trigger input after filtering

sasha RESOLVED in r61876

natalja.zabbix UI successfully tested

oleg.egorov CLOSED

(34) [F] Possible memory leak, tested on IE11
Leeks speed ~6kb\sec

Will be tested on other browsers.

Night test results:
IE memory leak by 16 hours is 422MB of RAM or ~7kb\sec
Chrome memory leak by hours is 241MB of RAM or ~4kb\sec

iivs My IE11 is eating 28kb/s. Over night it ate around 1.6GB of RAM. Trying to resize browser in the morning and using navigation to check if browser still works, it just freezes up and still continues to consume more memory. At this point I'm not sure if this is the only page or any page with 30s (default) refresh and left unattended for long periods of time.
FF crashed at unknown time.
This needs more investigation.

oleg.egorov Won't fix. CLOSED

Fixed in pre-3.2.0beta1 r61938.

When housekeeper remove an event (or in our test we truncated events table), the Monitoring => problems page shows an event age equal as the start of time, or unixtime at least. The events page seems to be OK.

See the attachments below.

backoffice.team, it is already fixed under ZBX-11132. Will be released in 3.2.0beta2.

(35) Show details filter option is not implemented

sasha RESOLVED in r62059

gunarspujats CLOSED

(36) Problems in CLOSING state are not blinking

sasha RESOLVED in r62060

gunarspujats CLOSED

(37) Applied new icons

old	new

sasha RESOLVED in r62125

gunarspujats CLOSED

(11), (17), (35), (36) and (37) are fixed in pre-3.2.0beta2 r62146.

(38) [F] incorrect values in "action" and "tag" column in exported csv file from Monitoring->Problem page. Tag column empty, but in action column tag values

iivs RESOLVED in r62308

natalja.zabbix ui successfully tested

sasha CLOSED

(39) [F] fixed displaying of actions popup; added footer for truncated action list

sasha RESOLVED in r62326.

Strings deleted:

• Actions on

oleg.egorov Fixed coding style in r62332. CLOSED

(38) and (39) are fixed in pre-3.2.0beta3 r62334.

any more detail on what the feature request is about ?

(1) [S] Database patch ready for testing in development branch svn://svn.zabbix.com/branches/dev/ZBXNEXT-3193
CLOSED

Server side ready for testing.

(2) No translation string changes.

sasha CLOSED

Apparently there is nothing to do for event API, since there is only event.get and the field is already read-only.

Updated server to conform the specification changes

(3) [D] Tables 'problem' and 'event recovery' have 'eventid' from 'events' as foreign key. However table 'events' can be partitioned (https://www.zabbix.org/wiki/Docs/howto/mysql_partitioning). At least MySQL does not support foreign keys on partitioned tables (http://dev.mysql.com/doc/refman/5.7/en/partitioning-limitations.html).

This is a showstopper for upgrading to 3.2 for those that have the 'events' table partitioned.

wiper 'events' table partitioning is not recommended anymore and we will stop supporting it. It should be clearly stated in upgrade notes and we should also update relevant documentation.

sandis.neilands Should we also provide a guide on how to unify partitioned 'events' table?

gunarspujats Discussed with sasha, guide won't be provided. CLOSED

(4) design Why 10000 events per batch during DB upgrade? How was this number determined? Considering that we'll hold the IDs of all events in memory during the upgrade - what is the maximum number of events that we have seen?

wiper Just and arbitrary number, not too small and not too large.
The following data is stored in memory during patch processing:

1. all event sources (triggers/items) - this is less what configuration cache has to keep in shared memory
2. active (not recovered) event identifiers - in theory this can be large if there are no OK events for triggers with multiple problem generation

wiper One solution would be to use temporary database table to store open events. However that would slow things down. For now we will just log warning when an event source reaches 10m open events.

sandis.neilands Another possible solution - add log that states percentage of events processed after each 5%. This would allow to forecast how much time is still needed for the upgrade.

wiper Would be nice, but there are upgrade steps we can't show step progress - for example changes to history table structure. So user will not be able to forecast to total time needed for upgrade.
RESOLVED in r60339.

sandis.neilands CLOSED with a comment. After upgrade upon receiving an OK event the server will be stuck while processing all those PROBLEM events.

(5) design API: why not expose problems via API? The workflow of automating common tasks fails if we do not have 1-to-1 correspondance between API and UI.

wiper This will done later in ZBXNEXT-3201
CLOSED

(6) [S] trigger_events_hash_func() - hash overwritten, hash of source computed twice.

wiper the hash is used as seed for the next hash step calculation. Fixed duplicate source addition to hash.
RESOLVED in r60337

sandis.neilands CLOSED.

(7) [S] assign_recovery_events() or update_event_recovery() (comment vs. reality)?

wiper RESOLVED in r60338.

sandis.neilands CLOSED.

(8) design If item and/or trigger removed while problem still unresolved then it is never removed from 'problem' table. Is this intended?

wiper It will be removed together with events by housekeeper
CLOSED

(9) design If trigger changed to generate single problem event, PROBLEM event arrives, then OK event arrives. All the open PROBLEM events for the trigger are closed. Is this intended?

wiper Yes, currently OK event closes all open PROBLEM events.
CLOSED

(10) design When is 'event_recovery' table supposed to be cleaned? How will it be used (which ZBXNEXT)?

wiper 'event_recovery' table records will be automatically removed with events (e.g. housekeeper).
CLOSED

richlv,
I asked myself the same. Judging the comment on (1), I assume it's about mapping things on database level to allow more efficient database queries... just a guess though.
A brief look at the specification would be helpful. But that's obviously supposed to be a surprising gift to us - we can get ready to be excited!

I assume it's about mapping things on database level to allow more efficient database queries...

Your assumption is absolutely correct. It's a quite technical feature request, which does not bring any end-user benefits if implemented alone.

(11) [S] Memory leak during upgrade in update_event_recovery().

==27269== 53,252,160 bytes in 1,633 blocks are possibly lost in loss record 1,065 of 1,065
==27269==    at 0x4C2AB80: malloc (in /usr/lib/valgrind/vgpreload_memcheck-amd64-linux.so)
==27269==    by 0x4E898B3: my_malloc (in /usr/lib/x86_64-linux-gnu/libmysqlclient.so.18.0.0)
==27269==    by 0x4E853CD: alloc_root (in /usr/lib/x86_64-linux-gnu/libmysqlclient.so.18.0.0)
==27269==    by 0x4E6A33A: cli_read_rows (in /usr/lib/x86_64-linux-gnu/libmysqlclient.so.18.0.0)
==27269==    by 0x4E6CB97: mysql_store_result (in /usr/lib/x86_64-linux-gnu/libmysqlclient.so.18.0.0)
==27269==    by 0x53F2E7: zbx_db_vselect (db.c:1392)
==27269==    by 0x53F5D4: __zbx_zbx_db_select (db.c:241)
==27269==    by 0x53F477: zbx_db_select_n (db.c:1624)
==27269==    by 0x5095FA: DBselectN (db.c:405)
==27269==    by 0x4FA4D8: update_event_recovery (dbupgrade_3010.c:276)
==27269==    by 0x4FA133: DBpatch_3010021 (dbupgrade_3010.c:350)
==27269==    by 0x4F1ECB: DBcheck_version (dbupgrade.c:784)
==27269== 
==27269== LEAK SUMMARY:
==27269==    definitely lost: 680 bytes in 85 blocks
==27269==    indirectly lost: 6,568,800 bytes in 426 blocks
==27269==      possibly lost: 76,010,400 bytes in 2,258 blocks
==27269==    still reachable: 814,257 bytes in 13,152 blocks
==27269==         suppressed: 0 bytes in 0 blocks
==27269== Reachable blocks (those to which a pointer was found) are not shown.
==27269== To see them, rerun with: --leak-check=full --show-leak-kinds=all

wiper RESOLVED in r60351.

sandis.neilands CLOSED.

Successfully tested the functionality.

We should check the performance once those features that depend on this feature are implemented.

Released in:

• pre-3.1.0 r60365

• leaving comments w/o acknowledging is a duplicate of ZBXNEXT-1629
• "unacknowledging" an ack is a duplicate of ZBXNEXT-1882

note that escalations, dashboard and other places are customisable based on ack status (operation condition, frontend filters), they do not always take it into account - that's a user preference. having a hard dependency would remove that flexibility

Marc, could you please describe what would be your desired outcome of this feature request?

Enabling "Multiple PROBLEM events generation" means that an event is generated whenever a trigger evaluates to PROBLEM, as can be seen in process_trigger() function in src/libs/zbxdbhigh/db.c, and that leads to event processing sequence, as you observed with the SQL queries. Function process_trigger() also walks dependencies of that trigger and, most importantly, executes a database query to update "lastchange" field in the database. So a significant increase in server load is expected.

The fact that one history syncer is busy for a long time is expected, too, because at most one history syncer can process one item's values, which is the case if you send millions of values for a single log item. If that is the main point of the report, then the issue might be better handled in already reported issues like ZBX-6353, ZBX-6925 and ZBX-7725.

The outcome, if that's possible at all, should be faster processing of cached values when having "Multiple PROBLEM events generation" set.
Or maybe cleverer somehow separated processing in general? Don't know whether that would be reasonable.
Well, and avoidance of (lots of) database queries for quite static data.

I didn't had concerns about the blocked history syncer. It was just something i noticed and wanted to mention to draw the complete picture of my observation.

However, what concerned me was the fact that the time after the high load (~17:15+ and ~21:20) the text cache appeared effectively not to be cleared anymore. Subjectively it seemed to grow by the rate of regular incoming text based data.
I expect the high loaded items to lag behind. But it seemed to affect other load unrelated text values as well, what could then lead to false-positive alerts.

I've no certain idea about the real root cause yet and thus no idea how it could be improved.
The only thing I saw was, having "Multiple PROBLEM events generation" enabled might lead to a problem that can only be solved by identifying the causing trigger and disabling this option temporarily.

Btw, it turned out that there were only two hosts with limited logging involved for testing. So the total count of send log values were just ~300k log values.
I falsely just took the information from the process information of the blocked history syncer, which processed >1 million values per run. Obviously it prcessed not just log values.

Edit:
As for the history syncer processes I have in mind that at 17:15 they all synced data in a "normal" period of 0.05-2 seconds. I think I should redo this test again to be more sure regarding some things....

Edit:
One question, am I right that no more than 400 events/seconds are processed per item and per interval?

One question, am I right that no more than 400 events/seconds are processed per item and per interval?

There is no such (hardcoded) limit. What lead you to make this conclusion?

I picked three or four blocks of events that were made at the very same second. These blocks always consisted of exactly 400 events.

linux operating system

in zabbix, trigger functions along with their parameters are supposed to be used for this

Can you please show an example on this?

please use zabbix irc, forums and other channels for community support. see https://www.zabbix.org/wiki/Getting_help for more detail

platform independent

Sorry, but, are you actually serious?

absolutely. At the company I work there is high correlation between the lunar phases and useless incoming mail which at the same time can not be qualified as spam. Such mapping seem interesting to our managers of activities. Of course, such monitoring could be made using various tools, but having all IT related monitoring in one place would be nice.

Jānis, while I definitely agree that different ways of looking at time have their proper use, such a feature would be improper for an IT product meant for general public.

OK, dropping the specific periods like moon pahses - lets look at it other way

• could it be possible for the user set the starting point in time for charting
  and to define its own periodicity/granularity according to his own needs?

As for the curiosity - while it looks ridiculous, such data can be used for
resource planning. I already know 3 organisations takinf into account such
"stellar" phonomena and those are not ones practicizing astrology or dark arts.

One of the characteristics of lunar days is that they are not equal in length, so adding the ability to define periodicity would only be of approximate value. Lunar day is also only one aspect of time and you might wish to consider other aspects, too. So what you are looking for in order to be precise is a general framework that would let user define a calendar system by providing some callback interface. A user would then provide such self-written callbacks to the calendar framework in Zabbix. So in order to graph by lunar days you would be required to write code anyway, and if you are required to write code anyway to address this custom need, you should probably just patch Zabbix yourself. Developing such a framework would complicate Zabbix code dramatically for the benefit of a small group of users. Perhaps, you could sign up for sponsored development, but it is unlikely to be developed on a voluntary basis. That was the reason for closing the issue originally.

Perhaps a way to add labels to graphs through API might solve your problem: see ZBXNEXT-117.

other use cases would be monitoring lighting and watering systems in correlation with lunar phases... but that is so highly specialised, i'd agree that for now it is unlikely to arrive in zabbix core. it would look nice on the feature list, though - but i guess per user timezone support is more useful, for example

if interest in this feature reappears, let's reopen this feature request.

still, was the interest to show lunar phases on the graph or or to have triggers react different ? triggers might be easier with an external datasource. graphs would be harder...

No, the idea was not to show phases on the graph, but graph according to lunar calendar - another item in granularity between week and standard month. Unfortunately, lunar month it is not a precise number of "solar" days (29,53 to be precise enough), as well as lunar week.

And what about potential Saudi Arabia clients? http://en.wikipedia.org/wiki/Lunar_calendar

You did not mention ITEM.VALUE macro, which is different from mentioned ITEM.LASTVALUE macro.

If you didn't know the ITEM.VALUE macro then it should resolve your requirements as I see.

Does ITEM.VALUE always point to the value that occurred at the time of the Event?

yes, exactly, with nanoseconds precision.

Thanks Oleksiy, I'll test and confirm.

Well i believe that this issue still important.

Even though ITEM.VALUE retaining the detail as long as the data lives in history and ITEM.LASTVALUE showing the current value of the item, many times you don't actually keep the history for long. Thats the purpose of trends: you erase a bunch of data from history to keep it small (or at least manageable). It is just too expensive to keep what an enterprise environment needs from events inside the history.

But the event, is somewhat more important than the history itself, at least in many cases. Considering Zabbix as a monitoring tool, the events it trigger are key for its existence.

With that in mind, an Trigger Name within the Event Details should be persisted as long as the event stays in the database. It should not be related to history, as it can (and many times it will) return as UNKNOWN.

We could implement this in two ways, i imagine:
1) set a flag inside the history so housekeeper does not erase that entry in particular.
2) instead of saving the trigger relation in events, we could save the full resolved name of that trigger inside the event.

What do you guys think about this?

Closing as duplicate of ZBXNEXT-2087.

as per https://www.zabbix.org/wiki/Docs/specs/ZBXNEXT-1575, displaying them during the initial development was "Discussed but will not be included"
as far as i recall, the rationale was that we could just show them, but to show them properly we would have to redesign that page quite a bit, so it was decided not to show them at all...

Thanks for the answer, but that was then, things maybe different now? I don't see why the Events tab would need to be changed that much to display internal events, but I'm more interested in being able to display all unsupported items, which from my point of view should probably be displayed in the Triggers tab. It is really great that we can get an email when an item becomes unsupported, but it would be even nice if we could view all unsupported items.

configuration -> hosts/templates, any item link, expand filter, set state to 'not supported' and clear host field

That view is awesome! The only thing it's missing is to be able to filter by maintenance status.

Zabbix could really use a web front-end overhaul. It has so many great features which are hard to find, or never even discovered.

I'd appreciate to optionally see internal events in a chronological order with proper filter functionality. This helps to see what recently happened without the need to create actions that fire lots of operations just to get this information.

Beside the fact that it sounds a little bit cumbersome to use e.g. MUA filter functions to query recent internal events, the alerter is due to lack of ZBXNEXT-2442 blocked quite quickly.

Inclusion in "Monitoring -> Events -> Source: Internal" would be awesome

This forum thread presents a good use case why it would be helpful to have this UI feature added.

Not having an easy way of viewing internal events means that problems with monitoring hosts and triggers on hosts themselves can't easily be identified and cleaned up, leading to other related issues.

Same Problem here, i need to check if trigger goes to unknown. But if I restart Tomcat unfortunately all jmx triggers go to unknown and i get a lot of messages in my maintenance period.

Also affects Zabbix version 2.2.2 and 2.2.0

This patch implements the requested behavior. The maintenance option shows up in the UI (and works) just like for trigger based actions. We are running this in production.

The patch is against svn r43648 on /branches/2.2 .

Not solved in 2.4

also asked in ZBXNEXT-3104

There could be other, much more simple cases why I don't want to receive alerts for internal events if a host in maintenance.
One such cases mentioned in ZBXNEXT-3140

Updated patch for zabbix 3.4.

By now the latter might be achieved partly by populating host inventory fields via items. Theses fields may then be used via macros.
ZBXNEXT-336 might gain usefulness for this use case a lot.

Since events are objects that don't exist before triggers change, I don't expect this to become an customizable object.
But maybe one approach could be ZBXNEXT-1645. This would allow to add macros on trigger level.

The last missing part would then be to add proper filter functionality based on macros(macro values) - what should be easy to implement in most places
A useful side-effect in using macros might be its support for inheritance:
global macro <- template macro <- host macro <- trigger macro

ZBXNEXT-2976 asks for "support of tags for configuration entities".

(1) [S] Database patch created in development branch svn://svn.zabbix.com/branches/dev/ZBXNEXT-2087

sasha Successfully tested! CLOSED

[S] Server side ready for review & testing.

(2) [A] trigger.create(), triggerprototype.create(), trigger.update(), triggerprototype.update(): tags array must be validated:

• tag cannot be empty
• tag cannot contain character '/'

gunarspujats RESOLVED in r59993

sasha the reason must be in such error messages instead of real value

1. _s('Incorrect value for field "%1$s": %2$s.', 'tag', $tag['tag'])
   _s('Incorrect value for field "%1$s": %2$s.', 'tag', _('a character string is expected'))
   _s('Incorrect value for field "%1$s": %2$s.', 'tag', _('unacceptable characters are used'))
2. the original value must be checked here

   if (trim($tag['tag']) === '') {

REOPENED

sasha RESOLVED in r60004

gunarspujats CLOSED

(3) [A] CTriggerGeneral:504,678: check for an empty array is useless

if (array_key_exists('tags', $trigger) && $trigger['tags']) {
        $this->validateAddTags($trigger['tags']);
}

gunarspujats RESOLVED in r59993

sasha CLOSED

(4) [A] frontends/php/include/classes/api/services/CTriggerGeneral.php:884:906: $db_trigger['tags'] and $trigger['tags'] must be sorted before processing

gunarspujats RESOLVED in r60002

sasha CLOSED

(5) [A] "Undefined index" errors

Undefined index: tag [hostgroups.php:63 → CFrontendApiWrapper->update() → CApiWrapper->__call() → CFrontendApiWrapper->callMethod() → CApiWrapper->callMethod() → CFrontendApiWrapper->callClientMethod() → CLocalApiClient->callMethod() → call_user_func_array() → CTrigger->update() → CTriggerGeneral->validateUpdate() → CTriggerGeneral->validateAddTags() → CArrayHelper::findDuplicate() in include/classes/helpers/CArrayHelper.php:204]
Undefined index: value [hostgroups.php:63 → CFrontendApiWrapper->update() → CApiWrapper->__call() → CFrontendApiWrapper->callMethod() → CApiWrapper->callMethod() → CFrontendApiWrapper->callClientMethod() → CLocalApiClient->callMethod() → call_user_func_array() → CTrigger->update() → CTriggerGeneral->validateUpdate() → CTriggerGeneral->validateAddTags() → CArrayHelper::findDuplicate() in include/classes/helpers/CArrayHelper.php:207]
Undefined index: tag [hostgroups.php:63 → CFrontendApiWrapper->update() → CApiWrapper->__call() → CFrontendApiWrapper->callMethod() → CApiWrapper->callMethod() → CFrontendApiWrapper->callClientMethod() → CLocalApiClient->callMethod() → call_user_func_array() → CTrigger->update() → CTriggerGeneral->updateReal() in include/classes/api/services/CTriggerGeneral.php:891]

gunarspujats RESOLVED in r60002

sasha Have a look at my changes in r60009

gunarspujats CLOSED

sasha checkTriggerTags() returns incorrect value if tags is not present. REOPENED

sasha RESOLVED in 60013.

gunarspujats CLOSED

(6) [A] trigger.get() and triggerprototype.get() produces PHP errors with selectTags = API_OUTPUT_COUNT

array_merge(): Argument #1 is not an array [hostgroups.php:64 → CFrontendApiWrapper->get() → CApiWrapper->__call() → CFrontendApiWrapper->callMethod() → CApiWrapper->callMethod() → CFrontendApiWrapper->callClientMethod() → CLocalApiClient->callMethod() → call_user_func_array() → CTrigger->get() → CTrigger->addRelatedObjects() → CTriggerGeneral->addRelatedObjects() → CApiService->outputExtend() → array_merge() in include/classes/api/CApiService.php:238]
array_flip() expects parameter 1 to be array, null given [hostgroups.php:64 → CFrontendApiWrapper->get() → CApiWrapper->__call() → CFrontendApiWrapper->callMethod() → CApiWrapper->callMethod() → CFrontendApiWrapper->callClientMethod() → CLocalApiClient->callMethod() → call_user_func_array() → CTrigger->get() → CTrigger->addRelatedObjects() → CTriggerGeneral->addRelatedObjects() → CApiService->outputExtend() → array_flip() in include/classes/api/CApiService.php:238]
array_keys() expects parameter 1 to be array, null given [hostgroups.php:64 → CFrontendApiWrapper->get() → CApiWrapper->__call() → CFrontendApiWrapper->callMethod() → CApiWrapper->callMethod() → CFrontendApiWrapper->callClientMethod() → CLocalApiClient->callMethod() → call_user_func_array() → CTrigger->get() → CTrigger->addRelatedObjects() → CTriggerGeneral->addRelatedObjects() → CApiService->outputExtend() → array_keys() in include/classes/api/CApiService.php:238]
Undefined index: triggerid [hostgroups.php:64 → CFrontendApiWrapper->get() → CApiWrapper->__call() → CFrontendApiWrapper->callMethod() → CApiWrapper->callMethod() → CFrontendApiWrapper->callClientMethod() → CLocalApiClient->callMethod() → call_user_func_array() → CTrigger->get() → CTrigger->addRelatedObjects() → CTriggerGeneral->addRelatedObjects() → CApiService->createRelationMap() in include/classes/api/CApiService.php:327]
Undefined index: triggerid [hostgroups.php:64 → CFrontendApiWrapper->get() → CApiWrapper->__call() → CFrontendApiWrapper->callMethod() → CApiWrapper->callMethod() → CFrontendApiWrapper->callClientMethod() → CLocalApiClient->callMethod() → call_user_func_array() → CTrigger->get() → CTrigger->addRelatedObjects() → CTriggerGeneral->addRelatedObjects() → CApiService->createRelationMap() in include/classes/api/CApiService.php:327]

gunarspujats RESOLVED in r60007

sasha CLOSED

(7) [A] Have a look at my changes in r59985.

gunarspujats CLOSED

(8) [A] better naming of validateAddTags(): checkTriggerTags()?

gunarspujats RESOLVED in r60002

sasha CLOSED

(9) [A] "selectTags" => API_OUTPUT_EXTEND returns triggertagid and triggerid

sasha RESOLVED in r60010 (CTriggerGeneral), r60012 (CEvent)

gunarspujats Deletion of old tags is broken. REOPENED

sasha RESOLVED in r60015

gunarspujats CLOSED

API was successfully tested!

(10) XML import/export

sasha RESOLVED in r60018

gunarspujats CLOSED

(11) [A] trigger tags inheritance doesn't work

sasha RESOLVED in r60020

gunarspujats CLOSED

(12) [F] Tags input form in triggers and trigger prototypes.

gunarspujats RESOLVED in r60028 (triggers), r60033 (trigger prototypes).

sasha

• tags must be sorted by ['tag', 'value'] in the forms
• have a look at my changes in r60082

REOPENED

gunarspujats RESOLVED in r60091

sasha Tags sorting should be applied in the form.

REOPENED

gunarspujats RESOLVED in r60111

sasha CLOSED

(13) [I] added new field value2 into conditions table

wiper CLOSED

(14) [S] action conditions processing doesn't respect tag field for "Tag value" condition

wiper RESOLVED in r60084

sasha CLOSED

(15) [S] lld_trigger.c trigger tags are not validated

wiper RESOLVED in r60096.

sasha Trigger tag cannot contain '/'. It must be validated. REOPENED

wiper RESOLVED in r60112.

sasha CLOSED with minor fix in r60123

(16) [S] src/zabbix_server/events.c:193 Only tag cannot contain character '/'.

if (0 != strchr(event_tag_local.tag->tag, '/') || 0 != strchr(event_tag_local.tag->value, '/'))

wiper RESOLVED in r60110

sasha CLOSED

(17) [S] src/zabbix_server/events.c:195 seems this code is useless

zbx_free_tag(event_tag_local.tag);
events[i].tags.values[j] = events[i].tags.values[--events[i].tags.values_num];
j--;

wiper RESOLVED in r60110

sasha CLOSED

(18) [S] Have a look at my changes in r60102, r60103, r60104, r60105, r60106, r60107, r60108 and r60109

wiper CLOSED

(19) [AF] Tags sorting should be removed from event.get and trigger.get.

gunarspujats RESOLVED in r60111

sasha CLOSED

(20) [AF] New action conditions "Tag" and "Tag value".

gunarspujats RESOLVED in r60111

sasha CLOSED with minor changes in r60125

Successfully tested!

Available in pre-3.1.0 (trunk) r60130.

This seems to have caused ZBX-10876.

(21) Updated documentation:

• What's new
• Event tags
• Trigger configuration
• Supported macros

Trigger configuration screenshots in LLD will be updated in the viewable LLD items/triggers/graphs task.

RESOLVED

glebs.ivanovskis As I understand {ITEM.LASTVALUE} is supported in Event tags too.

sasha {ITEM.LASTVALUE} is supported in Event tags and values. There are no differences between these macros. Documentation must be updated.

martins-v Updated:

• Supported macros for 3.2
• Event tags for 3.2
• Supported macros for 3.4
• Event tags for 3.4

RESOLVED

sasha Thanks! CLOSED

(22) [I] Database upgrade can get stuck if there are more than 10k discovery/autoregistration events in row.

wiper RESOLVED in r60700

sasha CLOSED

Fixed in pre-3.1.0 (trunk) r60821

(23) [F] Trigger tag

{{ITEM.VALUE}.regsub("CLASS:([a-zA-Z0-9/]+)","\1")}

fails validation while being valid. Probably because of / symbol in macro function parameters. Macros can be used in both - trigger tags and tag values.

sasha RESOLVED in r60802

gunarspujats Coding style

• missing PHPDoc in include/classes/validators/CTagValidator.php:42

sasha RESOLVED in r60819

gunarspujats CLOSED

Fixed in pre-3.1.0 (trunk) r60821

(25) [D] API documentation:

• https://www.zabbix.com/documentation/3.2/manual/api/changes_3.0_-_3.2
• https://www.zabbix.com/documentation/3.2/manual/api/reference/action
• https://www.zabbix.com/documentation/3.2/manual/api/reference/event
• https://www.zabbix.com/documentation/3.2/manual/api/reference/trigger
• https://www.zabbix.com/documentation/3.2/manual/api/reference/triggerprototype

sandis.neilands Description for tags except for "correlation_tag" is missing.

iivs Updated:

• https://www.zabbix.com/documentation/3.2/manual/api/reference/action/object
• https://www.zabbix.com/documentation/3.2/manual/api/reference/action/get
• https://www.zabbix.com/documentation/3.2/manual/api/reference/event/get
• https://www.zabbix.com/documentation/3.2/manual/api/reference/trigger/create
• https://www.zabbix.com/documentation/3.2/manual/api/reference/trigger/get
• https://www.zabbix.com/documentation/3.2/manual/api/reference/trigger/update
• https://www.zabbix.com/documentation/3.2/manual/api/reference/triggerprototype/create
• https://www.zabbix.com/documentation/3.2/manual/api/reference/triggerprototype/get
• https://www.zabbix.com/documentation/3.2/manual/api/reference/triggerprototype/update
• https://www.zabbix.com/documentation/3.2/manual/api/reference/configuration/export
• https://www.zabbix.com/documentation/3.2/manual/api/reference/configuration/import
• https://www.zabbix.com/documentation/3.2/manual/api/changes_3.0_-_3.2
• https://www.zabbix.com/documentation/3.4/manual/api/reference/action/object
• https://www.zabbix.com/documentation/3.4/manual/api/reference/action/get
• https://www.zabbix.com/documentation/3.4/manual/api/reference/event/get
• https://www.zabbix.com/documentation/3.4/manual/api/reference/trigger/create
• https://www.zabbix.com/documentation/3.4/manual/api/reference/trigger/get
• https://www.zabbix.com/documentation/3.4/manual/api/reference/trigger/update
• https://www.zabbix.com/documentation/3.4/manual/api/reference/triggerprototype/create
• https://www.zabbix.com/documentation/3.4/manual/api/reference/triggerprototype/get
• https://www.zabbix.com/documentation/3.4/manual/api/reference/triggerprototype/update
• https://www.zabbix.com/documentation/3.4/manual/api/reference/configuration/export
• https://www.zabbix.com/documentation/3.4/manual/api/reference/configuration/import

RESOLVED

sasha The wrong example for trigger prototypes methods. LLD macros are not included in the description and expression.

REOPENED

iivs Couldn't think of anything logical, so I deleted the that example and added tags for the first example. Updated:

• https://www.zabbix.com/documentation/3.2/manual/api/reference/triggerprototype/create#examples
• https://www.zabbix.com/documentation/3.2/manual/api/reference/triggerprototype/update#examples
• https://www.zabbix.com/documentation/3.2/manual/api/reference/triggerprototype/get#examples
• https://www.zabbix.com/documentation/3.4/manual/api/reference/triggerprototype/create#examples
• https://www.zabbix.com/documentation/3.4/manual/api/reference/triggerprototype/update#examples
• https://www.zabbix.com/documentation/3.4/manual/api/reference/triggerprototype/get#examples
  You're more than welcome to correct those examples and propose better ones.

RESOLVED

sasha Looks good to me. Thank you! CLOSED

(26) [S] Thanks to gunarspujats for discovering a crash!
Judging by backtrace and objdump Zabbix server crashes in lld_triggers_validate() on attempt to zbx_vector_ptr_clear_ext(&db_triggers, (zbx_clean_func_t)lld_trigger_free) where elements of db_triggers do not have a properly initialized tags field.

wiper I remember fixing this in one event correlation related branch. As the corresponding branches (3274, 3277) are already merged in trunk - this might be fixed. Have to check.

glebs.ivanovskis Yes, I confirm that.

+++ src/libs/zbxdbhigh/lld_trigger.c	(revision 61049)
@@ -1905,6 +1941,7 @@
 			zbx_vector_ptr_create(&db_trigger->functions);
 			zbx_vector_ptr_create(&db_trigger->dependencies);
 			zbx_vector_ptr_create(&db_trigger->dependents);
+			zbx_vector_ptr_create(&db_trigger->tags);
 
 			zbx_vector_ptr_append(&db_triggers, db_trigger);
 		}

CLOSED

To clarify,
Please add a regular expression filter. like vmware.eventlog[<url>, <regexp>]

It will be implemented in Zabbix 5.0, the work is already in progress.

Updated documentation:

• Preprocessing examples: 4.4, 5.0 (note that URLs and page naming is slightly different between 4.4 and 5.0, because preprocessing has a new dedicated page in 5.0)
• VMware monitoring items: 4.4, 5.0 (link added to the example page from vmware event log item)

Linux CentOS

could also be shown directly on the graph - ZBXNEXT-414

I'd say it indeed duplicates ZBXNEXT-414, idea is the same.
So I'm closing this one.

In first comment here http://blog.zabbix.com/multiple-servers-for-active-agent-sure/858/ I mentioned opposite - from 2.0 agents do not ask list of active checks by default.

Fixed in development branch svn://svn.zabbix.com/branches/dev/ZBXNEXT-1859

(1) an empty transaction

   518:20161011:131121.928 In process_areg_data()
   518:20161011:131121.928 query [txnlev:1] [begin;]
   518:20161011:131121.928 In DBregister_host_flush()
   518:20161011:131121.928 In process_events() events_num:0
   518:20161011:131121.928 End of process_events()
   518:20161011:131121.928 End of DBregister_host_flush()
   518:20161011:131121.928 query [txnlev:1] [commit;]
   518:20161011:131121.929 End of process_areg_data():SUCCEED

vso RESOLVED in r63089

sasha CLOSED

(2) This solution is not optimal:

1. hosts existence can be checked by one SQL statement for all discovery records;
2. same with auto-registration actions;
3. same with auto-registration data

vso RESOLVED in r63107, r63108, r63121, r63122

sasha CLOSED

(3) DB_DSICOVERED_HOST

1. typo in a name of structure DB_DSICOVERED_HOST; please, use new syntax for naming of such structures, for example: zbx_autoreg_host_t
2. this structure has string fields with static length. It badly works with multibyte strings
3. this structure must be moved into src/libs/zbxdbhigh/db.c module

vso RESOLVED in r63107, r63108, r63121, r63122

sasha CLOSED

(4) add_dhost()

1. this function can be rewritten to use only one structure DB_DSICOVERED_HOST
2. why here is used strncmp() function? Will be better strcmp()

vso RESOLVED in r63107, r63108, r63121, r63122

sasha CLOSED

(5) Incorrect formatting for SQL statements:
Incorrect:

    "select autoreg_hostid"
    " from autoreg_host"     
    " where proxy_hostid%s"
    " and host='%s'"
    ZBX_SQL_NODE,

Correct:

    "select autoreg_hostid"
    " from autoreg_host"     
    " where proxy_hostid%s"
        " and host='%s'"
        ZBX_SQL_NODE,

vso RESOLVED in r63107, r63108, r63121, r63122

sasha CLOSED

(6) DBregister_host_active(): check for actions must be called only when auto-registration data is not empty

vso RESOLVED in r63131

sasha CLOSED

(7) Have a look at my changes in r63128
vso CLOSED

(8) The code was completely rewritten in r63182.
vso Tested and looked into the changes. CLOSED

(9) Have a look at my changes in r63222
In each bulk first host would not show up in front end.

sasha Thanks a lot! CLOSED with minor fix in r63245.

(10) Have a look at my changes in r63234
Fixed memory leak in no proxy scenario.

sasha Thanks! CLOSED

Fixed conflicts in development branch svn://svn.zabbix.com/branches/dev/ZBXNEXT-1859-3.0

sasha Successfully tested with minor fix in r63255.

Fixed conflicts in development branch svn://svn.zabbix.com/branches/dev/ZBXNEXT-1859-3.2

sasha Successfully tested with minor improvements in r63287.

(11) [D] broken 2.2/ChangeLog

-New features:
+New features:i

vso RESOLVED in r63283

sasha CLOSED

Fixed in:

• pre-2.2.16 r63258
• pre-3.0.6 r63259
• pre-3.2.2 r63288
• pre-3.3.0 (trunk) r63290

(12) Documentation

• Upgrade notes 2.2.16, 3.0.6 and 3.2.2
• What's new 2.2.16, 3.0.6 and 3.2.2
• Active agent auto-registration 2.2, 3.0, 3.2 and 3.4

martins-v RESOLVED in upgrade notes/what's new.

Auto-registration pages already mention that an auto-registration action must be present. It does not mention explicitly what happens with event generation. Do we need to update anything?

sasha Thanks! It will be enough. CLOSED

RedHat el6.3

I guess this example is the same as in ZBX-6756
Personally I don't see as a reasonable approach to use multiple problem generating for such trigger expression.

ZBXNEXT-391 probably could help to resolve this issue.
Closed as duplicate.

Hint: create a template with these items, remove existing ones from a host, link the template to the host and in a result these two items will be polled with 1 second interval between them (there is a quite complex logic and it depends on item IDs).

remotely related - i tried to automate this with curl some time ago and there were undefined indexes : ZBX-7120

duplicating ZBX-9965 suggests a patch.

Now, Have any update for this change request?

In current version 4.2 there no separate page for Events.
Currently there is a Problems page and it export exactly the page you have opened.

kodai, can we close this? This is not the case anymore since 4.0.15 (see ZBXNEXT-4825).

ZBX-5719 (2) is related to this issue.

(1) database upgrade fails on DB2

22349:20130204:140424.147 query [txnlev:1] [create unique index events_1 on events (source,object,objectid,eventid)]
22349:20130204:140424.220 IBM DB2 ERROR: [-668] 57016 [[IBM][CLI Driver][DB2/LINUXX8664] SQL0668N Operation not allowed for reason code "7" on table "DB2INST1.EVENTS". SQLSTATE=57016]

After manual reorg on the table the database upgrade finishes successfully.

sasha RESOLVED in r33397

dimir CLOSED

(2) trigger nanosecond field might contain random value

src/libs/zbxdbcache/nextchecks.c:140

timespecs = zbx_malloc(timespecs, nextcheck_num * sizeof(zbx_timespec_t));
...
for (i = 0; i < nextcheck_num; i++)
{
	itemids[i] = nextchecks[i].itemid;
	timespecs[i].sec = nextchecks[i].now;
	errors[i] = nextchecks[i].error_msg;
}

The nanosecond part of timespecs[] items is not initialized.

sasha RESOLVED in r33545.

dimir CLOSED

The frontend is ready for testing.

The development branch is available in svn://svn.zabbix.com/branches/dev/ZBXNEXT-1575.

Updated documentation:
https://www.zabbix.com/documentation/2.2/manual/appendix/macros/supported_by_location

(3) [F] Small suggestion. The spec says about "Report not supported items" action condition:

Item state not supported

but the frontend says:

Item in "not supported" state

Also on the actions list page the "Conditions" have nested quoted strings:

Event type = "Item in "not supported" state"

I suggest changing these messages in frontend like specs say:

<SOMETHING> in "<STATE>" state -> <SOMETHING> state "<STATE>"

and getting rid of nested quoted strings on actions list page.

<Sasha> The strings was changed after discussion with Martins. Specification was changed too.

jelisejev We've also decided, that the value of an action condition will be displayed in italic, instead of quotes. RESOLVED in r34751.

<Sasha> Successfully TESTED. Code review are needed.

oleg.egorov CLOSED

(5) [F] Unlinking a default host "Zabbix server" after clean install from both templates gives error when trying to save host:

Cannot delete templated trigger "Zabbix alerter processes more than 75% busy:

>75".

Image (zbxnext-1575-frontend-template-unlink-error.png) attached.

oleg.egorov Impossible unlink template not only for default host. I can't unlink windows template from new host.

jelisejev RESOLVED in r34566.

oleg.egorov CLOSED

(6) [F]
Monitoring->Events Source: Discovery
New installation without events

Incorrect object "0" (trigger) for source "1" (discovery), only the following objects are supported: 1 - discovered host, 2 - discovered service.

jelisejev RESOLVED in r34604.

oleg.egorov CLOSED

(7) [F]
Please, fix coding style.

configuration.triggers.list.php: 184 and 191 too looooong rows

defines.inc.php:426 unnecessary tab
I think tabs is better, please don't use spaces in defines.inc.php
And in 683-693 fix style

CEvent.php:54 and 63 row, please remove ","
Again spaces... :105, 130...
Please div long rows (231) and in other places (743, 796)

forms.inc.php:531 and 569 please remove ","
And ";" from 92 and 459 lines (not your mistake)
Same problem in CTimePeriodValidator.php:46

configuration.action.edit.php:321, 448 long rows

actions.inc.php:594 please remove ","

CTrigger.php: 1002 unnecessary row break

jelisejev RESOLVED in r34568.

oleg.egorov CLOSED

(8) [F] Impossible acknowledge trigger from Monitoring->overview

oleg.egorov CLOSED

(9)[F] Monitoring->Dashboard->Last 20 issue missing Ack column

jelisejev CLOSED.

(10)[F] If open trigger event for "acknowledge"

Argument 1 passed to CMacrosResolverHelper::resolveTriggerName() must be an array, null given, called in /var/www/oleg/ZBXNEXT-1575/frontends/php/acknow.php on line 144 and defined [acknow.php:144 → CMacrosResolverHelper::resolveTriggerName() in /var/www/oleg/ZBXNEXT-1575/frontends/php/include/classes/macros/CMacrosResolverHelper.php:74]

jelisejev I cannot reproduce this problem.

oleg.egorov
http://192.168.3.4/oleg/ZBXNEXT-1575/frontends/php/acknow.php?eventid=328&triggerid=13485&backurl=dashboard.php&sid=95fa56691a407b13

EventId = 328 - don't exist...

Same problem in trunk.

jelisejev RESOLVED in r34727.

oleg.egorov CLOSED

(11)[F] Missing Ack column in Monitoring->Events

oleg.egorov CLOSED

(12)[F] In tr_events.php lost "Acknowledges table".
Monitoring->Events->Source "Trigger"->Time

oleg.egorov CLOSED

(13)[F] If we have in event.acknowledge (CEvent.php:769)

Request like this:

{
        "eventids": [40, 41, 42, 43, 44, 45, 46, 47...],
        "message": "Problem resolved."
}

And events is not in "allowed".

Then we have too much SQL queries.

I think better collect not allowed events and then in one query check all ID's.

jelisejev The additional API request will be executed only once, no matter how many valid or invalid IDs there will be. After that it will fail with an error

oleg.egorov CLOSED

(14)[A] Implement changes to the alert API described in the specs.

jelisejev RESOLVED in r34624-34687.

oleg.egorov CLOSED

(15)[F] Removed

define('ITEM_STATUS_NOTSUPPORTED',	3);

from defines.inc.php

ITEM_STATUS_NOTSUPPORTED used in c20ImportFormatter.php in 138 and 170 lines.
Please revert "define".

jelisejev RESOLVED in r34571.

oleg.egorov CLOSED

Server side successfully tested. Please review my changes in r33719:33794, r33805, r33858:34018.

(16) [D] Please document the new macros related to ZBXNEXT-1663 when this gets into trunk.

and

should be mentioned in

https://www.zabbix.com/documentation/2.2/manual/introduction/whatsnew220
https://www.zabbix.com/documentation/2.2/manual/appendix/macros/supported_by_location

You can ask me for details on where this is supported or just ask me to document it.

<Sasha> RESOLVED

dimir CLOSED

(17) When creating a new item or LLD rule, the "Enabled" checkbox is not checked by default.

jelisejev RESOLVED in r34688.

oleg.egorov CLOSED

(18) Review my minor changes in r34668

jelisejev Good. CLOSED.

(19) Administration->Notifications

Fatal error: Maximum execution time of 600 seconds exceeded in /var/www/oleg/ZBXNEXT-1575/frontends/php/report4.php on line 224

Don't work...

jelisejev RESOLVED IN r34701

oleg.egorov CLOSED

(20) [F] Host full clone

    Created: Application "admin_item" on "admin2".
    Created: Item "admin_item" on "admin2".
    Created: Trigger "admin_trigger" on "admin2".
    Cannot set "state" for item "admin_discovery".

jelisejev RESOLVED in r34731.

oleg.egorov CLOSED

TESTED.

Available in 2.1.0 r34766.

(21) Documentation needs to be updated.

martins-v Updated sections:

(with internal event info)
https://www.zabbix.com/documentation/2.2/manual/config/events
https://www.zabbix.com/documentation/2.2/manual/config/events/sources#internal_events
https://www.zabbix.com/documentation/2.2/manual/config/notifications/action
https://www.zabbix.com/documentation/2.2/manual/introduction/whatsnew220#notifications_on_unsupported_items

(regarding logic of Status column)
https://www.zabbix.com/documentation/2.2/manual/web_interface/frontend_sections/configuration/hosts/items
https://www.zabbix.com/documentation/2.2/manual/web_interface/frontend_sections/configuration/hosts/triggers
https://www.zabbix.com/documentation/2.2/manual/web_interface/frontend_sections/configuration/hosts/discovery

(action conditions for internal event actions added)
https://www.zabbix.com/documentation/2.2/manual/config/notifications/action/conditions

A new how-to section for receiving notification on unsupported items:
https://www.zabbix.com/documentation/2.2/manual/config/notifications/unsupported_item

Updated screenshots:

(trigger display without 'unknown' value)
https://www.zabbix.com/documentation/2.2/manual/web_interface/frontend_sections/reports/status_of_zabbix

(trigger display without 'unknown' value in Status of Zabbix)
https://www.zabbix.com/documentation/2.2/manual/web_interface/frontend_sections/monitoring/dashboard

(no 'unknown' column)
https://www.zabbix.com/documentation/2.2/manual/web_interface/frontend_sections/reports/availability

(Enabled field with a check-box)
https://www.zabbix.com/documentation/2.2/manual/config/items/item

(Filter screenshot has a State attribute)
https://www.zabbix.com/documentation/2.2/manual/web_interface/frontend_sections/configuration/hosts/items

Ok, the changes mentioned in the spec seem to be covered. If somebody has time to review these, that'd be great.

(22) The API docs and changelog needs to be updated.

jelisejev Updated:
https://www.zabbix.com/documentation/2.2/manual/api/changes_2.0_-_2.2
https://www.zabbix.com/documentation/2.2/manual/api/reference/action/object
https://www.zabbix.com/documentation/2.2/manual/api/reference/event/object
https://www.zabbix.com/documentation/2.2/manual/api/reference/alert/get

https://www.zabbix.com/documentation/2.2/manual/api/reference/event/get
https://www.zabbix.com/documentation/2.2/manual/api/reference/event/acknowledge

https://www.zabbix.com/documentation/2.2/manual/api/reference/item/object
https://www.zabbix.com/documentation/2.2/manual/api/reference/item/get
https://www.zabbix.com/documentation/2.2/manual/api/reference/item/getobjects

https://www.zabbix.com/documentation/2.2/manual/api/reference/discoveryrule/object
https://www.zabbix.com/documentation/2.2/manual/api/reference/discoveryrule/get

https://www.zabbix.com/documentation/2.2/manual/api/reference/trigger/object
https://www.zabbix.com/documentation/2.2/manual/api/reference/trigger/get

(23) would be great to do some basic performance analysis/testing - would the generation of new events have negative impact in some cases (like lots of items/triggers going into unsupported/unknown state).
in the light of our recent experience with lld, would a large amount of such events have a potential for db locking issues ?

was this issue closed so nobody would notice we didn't bother to document it properly ?
unclosed subissues : 21, 22, 23 (maybe more ? )

...and i do not think documentation work can properly start because we do not have a finished specification. unfortunately.

i was assured that the spec is complete, which sounds great. gotta read it

(25) [S] BLOCKER! Zabbix server crashes.

 31849:20130416:150308.603 In update_triggers_status_to_unknown() hostid:10084
 31849:20130416:150308.605 query [txnlev:0] [select distinct t.triggerid,t.description,t.expression,t.priority,t.type,t.value,t.state,t.error,t.lastchange from items i,functions f,triggers t,
hosts h where i.itemid=f.itemid and f.triggerid=t.triggerid and i.hostid=h.hostid and i.status=0 and i.state=0 and i.type in (0) and f.function not in ('nodata','date','dayofmonth','dayofweek
','time','now') and t.status=0 and h.hostid=10084 and h.status=0 and not exists (select 1 from functions f2,items i2,hosts h2 where f2.triggerid=f.triggerid and f2.itemid=i2.itemid and i2.hostid=h2.hostid and (f2.function in ('nodata','date','dayofmonth','dayofweek','time','now') or (i2.type not in (0) and (i2.type not in (0,1,4,6,12,16) or (i2.type in (0) and h2.available=1) or (i2.type in (1,4,6) and h2.snmp_available=1) or (i2.type in (12) and h2.ipmi_available=1) or (i2.type in (16) and h2.jmx_available=1)))) and i2.status=0 and i2.state=0 and h2.status=0) order by t.triggerid]
 31843:20130416:150308.655 One child process died (PID:31849,exitcode/signal:1). sig_exiting ...

wiper reviewed and successfully tested. CLOSED

sasha fixed in version pre-2.1.0 r35068.

(27) this probably solved the following, too - please, verify :

• ZBXNEXT-341
• ZBX-1003

sasha

• ZBXNEXT-341 - can be closed
• ZBX-1003 - isn't fixed

<richlv> closed ZBXNEXT-341, subissue CLOSED

(28) currently a problem exists. if we filter for "not supported", we also show disabled items without any indication why. we can :

a) implement some visualisation for such items;
b) reset unsupported state when enabling/disabling items;
c) do not show disabled-notsupported items in such case

it was decided to go with (c)
... and i also added this to the spec

jelisejev RESOLVED in svn://svn.zabbix.com/branches/dev/ZBXNEXT-1575.

<richlv> commit message said something about disabling one filter when another is in use (thanks for mentioning that ) - i believe spec is missing that detail

jelisejev done https://www.zabbix.org/mw/index.php5?title=Docs%2Fspecs%2FZBXNEXT-1575&diff=4369&oldid=4318

Eduards REOPEN
1. If status is "enabled" and is readonly and don't have previous position, and the state field is switched to "all", status field must stay as is - "enabled".
2. configuration.item.list.php become without lead 2 empty rows after copyright block.
3. + some js formating as discussed.

jelisejev RESOLVED in r36139 and r36143.
Eduards CLOSED, please review small fix r.36153

Tested!

Implemented in 2.1.0 r36160.

CLOSED.

Is there any documented configuration scenario to raise a trigger when some SNMP item becomes 'unsupported' so one can make a inter-devices link visualize a problem on a map? I've 2.1.2-alpha3 installed and can only create a 'send message' action based on internal event. We need to monitor OSFP states through many devices.

I've forgot to mention: both item and trigger are LLD-prototyped.

(29) The trigger error shouldn't be updated when expression was changed.

<richlv> this will be confusing now
fix version is "2.1.0", but there will be further fixes attached to this issue

sasha Moved to ZBX-7026. CLOSED

It caused a regression in ZBX-7452

Available in the development branch svn://svn.zabbix.com/branches/dev/ZBXNEXT-1574.

(3) [F] Triggers with value_flags=1 are not shown neither in "Monitoring" -> "Events" nor in "Monitoring" -> "Triggers".

Here is how the trigger looks when status is OK and it's visible in "Monitoring":

 triggerid | expression | description | url | status | value | priority | lastchange | comments |                                               error                                               | templateid | type | value_flags | flags
-----------+------------+-------------+-----+--------+-------+----------+------------+----------+---------------------------------------------------------------------------------------------------+------------+------+-------------+-------
     13546 | {12980}>10 | random>10   |     |      0 |     0 |        2 | 1358929802 |          | Received value [test] is not suitable for value type [Numeric (unsigned)] and data type [Decimal] |            |    0 |           0 |     0

Now it becomes "unknown" and disappears from "Monitoring" (only change is "value_flags"):

 triggerid | expression | description | url | status | value | priority | lastchange | comments |                                              error                                               | templateid | type | value_flags | flags 
-----------+------------+-------------+-----+--------+-------+----------+------------+----------+--------------------------------------------------------------------------------------------------+------------+------+-------------+-------
     13546 | {12980}>10 | random>10   |     |      0 |     0 |        2 | 1358929802 |          | Received value [foo] is not suitable for value type [Numeric (unsigned)] and data type [Decimal] |            |    0 |           1 |     0

Database changes successfully tested.

Perhaps ZBX-6080 should be closed after this issue is implemented.

(4) in events.inc.php
if (bccomp($startEvent['eventid'],$event['eventid']) == 0 && $nextevent = get_next_event($event, $events, true)) {

unnecessary 3rd param in get_next_event()

jelisejev RESOLVED in r33101.

tomtom CLOSED

(5) in CScreenEvents.php
$showUnknown = CProfile::get('web.events.filter.showUnknown', 0);

should be removed

jelisejev RESOLVED in r33101.

tomtom CLOSED

(6) in triggers.inc.php line:1456

statement: if ($state == -1) {

never executes now

jelisejev RESOLVED in r33109.

tomtom CLOSED

(7) in triggers.inc.php in calculate_availability()

if ($total_time == 0) {
	$ret['true_time'] = 0;
	$ret['false_time'] = 0;
	$ret['true'] = 0;
	$ret['false'] = 0;
}

Maybe $ret['true'] should be 100!?

jelisejev As far as I can see, this code should not be executed at all. But I suggest we leave it as is for now and deal with it in ZBX-5785.

tomtom CLOSED

(8) comments should be rewritten:

"// if disable trigger, unknown event must be created" in triggers.php

" * @param int $triggerValue TRIGGER_VALUE_FALSE, TRIGGER_VALUE_TRUE or TRIGGER_VALUE_UNKNOWN" in triggers.inc.php

"// set element to problem state if it has problem events, ignore unknown events" in maps.inc.php. At 2 places!

jelisejev RESOLVED in r33101. The last one should be removed under ZBX-6172.

tomtom CLOSED

(9) "return _('Unknown');" in trigger_value2str() should be rethought. Seems that it is unnecessary.

jelisejev I've left it as is because that's how all similar functions behave.

tomtom CLOSED

(10) "$hintTable->addRow(array(new CCol(SPACE, 'trigger_unknown'), _('Unknown')));" should be removed in monitoring.overwiew.php

probably associated class 'trigger_unknown' as well.

jelisejev Will be fixed under ZBX-6173. CLOSED.

tomtom CLOSED

(11) "$this->ok('Number of triggers (enabled/disabled)[problem/unknown/ok]');" in testPageStatusOfZabbix.php should be updated.

jelisejev RESOLVED in r33103.

tomtom CLOSED

(12) "$this->ok(array('Host', 'Name', 'Problems', 'Ok', 'Unknown', 'Graph'));" in testPageAvailabilityReport.php should be updated. At 2 places!

jelisejev RESOLVED in r33103.

tomtom CLOSED

DB changes and serverside tested. sasha please see my small changes in r33114 and r33122.

tomtom Frontend successfully TESTED.

There have been conflicts when updating to the latest trunk. Please review r33166.

Available in 2.1.0 r33169.

Updated the API docs:

the property list in https://www.zabbix.com/documentation/2.2/manual/api/reference/event/object
examples in https://www.zabbix.com/documentation/2.2/manual/api/reference/event/get
and the 2.2 changelog https://www.zabbix.com/documentation/2.2/manual/api/changes_2.0_-_2.2

(13) should be documented in whatsnew and upgrade notes, just have to figure out what are the important user-facing changes

(14) If events fro trigger has been deleted and trigger in the PROBLEM state (for example, when trigger is in PROBLEM state very long time), we can see the next errors:

Undefined index: eventid [dashboard.php:97 → make_system_status() → makeTriggersPopup() → getEventAckState() in /var/www/zabbix/trunk-clean/include/events.inc.php:298] 
Undefined index: objectid [dashboard.php:97 → make_system_status() → makeTriggersPopup() → getEventAckState() in /var/www/zabbix/trunk-clean/include/events.inc.php:298] 
Undefined index: acknowledges [dashboard.php:97 → make_system_status() → makeTriggersPopup() → getEventAckState() in /var/www/zabbix/trunk-clean/include/events.inc.php:299] 
Undefined index: acknowledges [dashboard.php:97 → make_system_status() → makeTriggersPopup() → getEventAckState() in /var/www/zabbix/trunk-clean/include/events.inc.php:307]

jelisejev RESOLVED in r33386.

tomtom REOPENED blocks.inc.php line: $triggers[$tnum]['clock'] = isset($trigger['event']) ? $trigger['event']['clock'] : $trigger['lastchange'];
$trigger['lastchange'] could be used always.

jelisejev RESOLVED in r33486.

tomtom CLOSED

Fixed in 2.1.0 r33493.

CLOSED.