In the trunk such items are not displayed on Latest Data page at all.

Fixed for 1.8 in /branches/dev/ZBX-4586.

RESOLVED.

Added item's application validation to the Item.update and Item.create API methods.

Successfully tested!

Please review my changes in r25102.

Merged to 1.8 r25112.

My mistake, it seems that cloning works correctly in the trunk, but I've added the same application validation I've implemented in 1.8. It's available in /branches/dev/ZBX-4586-trunk.

(1) A 'host' field is not used
class.citemgeneral.php:119,133

<pavels> RESOLVED.
<Sasha> CLOSED

(2) class.citemgeneral.php:284 An error message violate coding guidelines.

<pavels> RESOLVED.
<Sasha> CLOSED

(3) Errors, when creating new item with applications:

• in_array() expects parameter 2 to be array, null given [api/classes/class.citemgeneral.php:283]
• Undefined index: name [api/classes/class.citemgeneral.php:284]
• Application with ID "292" is not available on host "" [CItem.create -> CItem.checkInput -> CItemGeneral.checkInput]

<pavels> RESOLVED.
<Sasha> CLOSED

Merged to trunk r25140.

CLOSED.

svn://svn.zabbix.com/trunk 25028

Redhat EL6.1 / Postgres 8.4.7 / Zabbix 1.9.8
for both Master and proxies

Related issue: ZBX-3739

Fixed in the development branch svn://svn.zabbix.com/branches/dev/ZBX-4577-trunk.

In version 1.8.х such error is eliminated.

Tested. Great performance improvement in the DBadd_condition_alloc() function!

Fixed in version pre-1.9.9, r25120.

really fixed it ? why does my zabbix_proxy has this error ? version:2.0.8 ... I retry to import the database , but it doesn't helpful.

The file is class.ctemplate.php

See also ZBX-4573

Fixed in ZBX-4519.

When executing such SQL statement on MySQL DB appears the following error:

ERROR 1064 (42000): You have an error in your SQL syntax; check the manual that corresponds to your MySQL server version for the right syntax to use near '*) AS cnt,MIN(e.clock) AS minn, MAX(e.clock) AS maxx FROM events e WHERE e.objec' at line 1

BUT when executing this SQL query under PostgreSQL DB all is fine:

postgres=# select count(e.*) AS cnt,MIN(e.clock) AS minn, MAX(e.clock) AS maxx FROM events e WHERE e.objectid=13025 AND e.object=0 AND clock>=1327057886 AND clock<=1327057886

postgres-# ;
cnt | minn | maxx
-------------
0 | |
(1 row)

This problem appears for both report modes: "By host" and "By trigger template".

Broken in r24440 (ZBXNEXT-905)

Fixed in the development branch svn://svn.zabbix.com/branches/dev/ZBX-4570

(1) Please review my changes in r24992. If everything is ok - you can merge.

<Sasha> CLOSED

TESTED.

Fixed in version pre-1.9.9, revision 24995.

I think we should remove this code as suggested. I am working on it.

Fixed in dev branch branches/dev/ZBX-4564, ready to test.

<zalex> Agent and Server of dev branch tested. All is fine. Seems ready to code review and merge.

Great! Tested successfully!

Implemented in revision 25024.

Note: Fixed in pre-1.8.11 r25022, pre-1.9.9 r25023

trunk, revision 24812

(1) If we change the trigger back, from host:agent.ping.last(0) to template:agent.ping.last(0) the templated triggers 'flag' and 'value_type' values are reset.

RESOLVED.

<Sasha> REOPENED
The "Multiple problem events" flag now "blinks".
Steps to reproduce:

• create trigger prototype and save
• open it: Multiple problem events[checked]; save
• open it again: Multiple problem events[unchecked]; save
• open it again: Multiple problem events[checked]; save

<pavels> RESOLVED.

<Sasha> CLOSED

(2) There is no information message about removed trigger. Audit record too.

<pavels> We've decided not to add audit records, but I've added the info messages. RESOLVED.

<richlv> why audit records won't be added ?

<pavels> I fear, that their current implementation may cause performance problems if a large amount of hosts is affected.

<Sasha> CLOSED

(3) Please review my changes in r24944:24945

<pavels> Thnx, CLOSED.

(4) No audit records for operations Create and Update.

<pavels> RESOLVED.
<Sasha> CLOSED Please review my changes in r25019.

<pavels> Thanks, everything OK.

There was a conflict when updating to the latest trunk in r25030 class.ctrigger.php. Please review.

<Sasha>

• db_items and db_trigger should be renamed to dbItems and dbTrigger
• trunk/frontends/php/api/classes/class.ctrigger.php:1511,1513,1538,1539 - indentation should be fixed
• trunk/frontends/php/api/classes/class.ctrigger.php:2103 - translation has disappeared (S_INCORRECT_DEPENDENCY => _('Incorrect dependency'))

<pavels> RESOLVED.

<Sasha> CLOSED Please review r25033.

Merged to trunk r25039.

CLOSED.

Reopened: See ZBX-4598 for more details.

<pavels> The new bugs will be fixed in ZBX-4598, so I'll close this issue.

Solaris
SunOS 5.10 Generic_142910-17 i86pc i386 i86pc

Thanks for the report. We've discovered this problem recently in another issue and have fixed it there but it's not merged to trunk yet so we will move the fix here.

We decided to add special define ZBX_IS_UINT64_MAX_LEN for is_uint64_n() function instead of using INT_MAX.

Fixed in development branch svn://svn.zabbix.com/branches/dev/ZBX-4555 .

Successfully tested! Please review my change in r24885 before merge.

Fixed in pre-1.9.9 r24893 .

We've recently discussed a possible implementation of a new autoloader that would use spl_autoload_register. Hopefully, it will be done in the near future, then we could use your patch.

By the way, thank's a lot about your feedback. We really need some perspectives on how to make the PHP code more extendable and reusable for other developers.

Updated autoloader which checks if the file exists prior attempting to load it. Improves compatibility with other autoloaders, as the zabbix autoloader doesn't throw an error; allowing other class loaders to catch up.

By the way, I'd need a pointer with SVN:

I simply do a svn diff which spits out a patch very nicely. However, as I can't commit to the zabbix SVN, how can I mark the patch as "done"? I.e. if I now do another change which would go to another area of zabbix, a svn diff would also include the contents of this patch, which is not exactly good.

With git, I simply could commit locally, and then create a new diff.

usually it's "handled manually".

so if there's a patch that depends on another patch and you wouldn't want to create a "monster patch", you would do a diff on two trees and just mention that this patch now depends on that one. "diff -u", of course

A new autoloader has been implemented in svn://svn.zabbix.com/branches/dev/ZBX-4538.

We've also introduced a new Z class to serve as a wrapper for the autoloader registration and some other stuff in the future.

RESOLVED.

(1) [GUI]
I have changed some things, we can discuss that

<pavels> Please review my changes in r24914.

<Vedmak> Please review mine in rev. 24929.

<pavels> CLOSED.

I haven't looked at the changes yet - is it possible to view the change on the web without svn checkout ?

there is no official web frontend for zabbix svn (yet)

I have looked at this now and I have a few questions:

There's ZBase (which seems to be the base class for the Zabbix application), but then there's the "Z" class. I'm not sure what's the reason behind this, but it looks a bit odd, as ZBase::getInstance() returns a new Z() instance. This looks a bit like an approach for 3rd-parties to include their own functionality, but would require people to overwrite the class.z.php with their own version. Did I get that right, or is there something I've missed?

I'm also not a very big fan of extremely short method names. I'd prefer "getInstance" instead of "i", because when you read "getInstance", it's obvious what it does, while "i" is not.

A picky one: Line 77 of class.zbase.php is longer than 120 characters

Overall the code looks good, I have nothing left to complain. I can also confirm that the auto loader works from ExtZabbix without modifications.

Yes, the Z class was introduced to provide an ability to extend the core Zabbix functionality. To do that you can either overwrite class.z.php, or insert your own class between Z and ZBase.

I don't think that providing an empty Z class is the right approach, and overwriting files from a distribution also isn't a good idea, because the class.z.php would probably get overwritten when updating.

A better approach would be to specify the class to use within a configuration file, and then return an instance of the configured class instead of returning an instance of ZBase.

Yes, this may not be the most correct approach. But if you're modifying something in Zabbix you'll probably set up your own repository and then all of the conflicts can be easily resolved.

Merged to trunk r24954.

CLOSED.

Ubuntu Dev Precise (12.04)

Fixed in the development branch svn://svn.zabbix.com/branches/dev/ZBX-4535

Just for record: as I see this dev branch already includes fix from ZBX-3475. Great!
So ZBX-3475 can be closed after this one.

Great fix! Even more order in our autoconfiguration process! Just tiny formatting change in r25079.

<Sasha> CLOSED

Fixed in versions pre-1.8.11 r25085 and pre-1.9.9 r25087.

Tested on Linux.

might be another items.lastvalue manifestation (although i can't find the previous issues on this right now)

what database are you using?

Cannot reproduce with PostgreSQL.

Check a structure of a 'items' table. Size of fields 'lastvalue' and 'prevvalue' should be 255.

Mysql.
Yes, lastvalue and prevvalue columns of items table have type "varchar(255)".

Is it possible that it depends of 8-bit value?
In my case, triggers "

<1" and "

<1" returns problem, but "

<1" returns OK.
I can open my zabbix_agentd for your server if it can help.

Please attach screenshot of "web.page.get[fidonet.org.ua]" item configuration.

web.page.get[fidonet.org.ua] item configuration

Confirmed: The given problem happens if in the first 255 characters is present the CR (0х13) character.

Fixed in the development branch svn://svn.zabbix.com/branches/dev/ZBX-4526

The priority is changed to the "Critical".

Successfully tested. Please fix ChangeLog message, it contains typo.

Fixed in version pre-1.8.11, revision 24752.

Sébastien, welcome to the translation team !
Here is many similar issues with suggestions like your and I believe that they all will be fixed according to the ZBXNEXT-790 for Zabbix 2.0.[1-9]
Additionally I would like to let you know about issues ZBX-3271 and ZBX-4402 and see to a page http://www.zabbix.org/wiki/Translating_Zabbix

Regards,
Oleksiy

i'm not quite sure it's a good idea to fix it this way in 1.8 branch... this introduces lots of new translation strings - even if more correct, this will mean additional work to translators who might already have translated those partial strings (and who would not want to translate a bunch of new ones for 1.8 as they work on trunk already).

zalex, your opinion ?

Of course, I'm agree with Richlv for 1000%

Three hours ago (when Sanja assigned the issue himself) I checked the source of dev branch copied from and it was .... 1.8. I wanted to ask - WHY?
Why to spend time for 1.8 branch?
But I decides to be silent -> maybe it's a case to training/solve a not hard task or so.

I'm agree that would be better to resolve this issue for the 2.0 only.

It introduces 57 new strings. May be few more will be added.

Fixed for 1.8 in development branch svn://svn.zabbix.com/branches/dev/ZBX-4523

Fixed for trunk in development branch svn://svn.zabbix.com/branches/dev/ZBX-4523-trunk

(1) that's at least 57 untranslated strings, introduced very lately in the branch lifecycle. based on zalex's opinion, i'd suggest not to do this for 1.8

<Sasha> Let's not do it in version 1.8. Too many untranslated new phrases. CLOSED

(2) trunk branch seems to add new S_ strings

I'm personally not sure worth to merge it to trunk as it is at the moment. Maybe would be better to discard these changes and to work only according to the ZBXNEXT-790 ?

<sanja> RESOLVED

(3) [GUI]
Please review my changes in rev. 24783.

I reviewed 1.8 branch, don't forget to remove it CLOSED

(4) [GUI]
Please review my changes in rev. 24819.

<sanja> CLOSED

Fixed in trunk, r24888

interestingly, this was broken a long, long time ago in :

------------------------------------------------------------------------
r15624 | Vedmak | 2010-11-22 16:27:37 +0200 (Mon, 22 Nov 2010) | 2 lines

• ZBX-2577 fixed template trigger update affects downstream triggers
  when fixing, please, add unittests
  <Slava>As i understand it's doesn't work in version 1.8.3 too ...
  <Slava> RESOVED r 24758

(1) unittests ?
<Slava>RESOLVED

(2) when adding dependencies to a host trigger, templates are still shown. is that intended ?
<Slava> RESOLVED in r 24776

<richlv> CLOSED

(3) in the dev branch, when creating a trigger :

reset() expects parameter 1 to be array, null given [api/classes/class.ctrigger.php:1635]
Created: Trigger "xxx" on "".
<Slava>This bug was created in ZBX-2737. RESOLVED in r 24776
<Sasha> CLOSED Fixed in r24780 in ZBX-2737

(4) create a new trigger, add a dependency on some template trigger. choose an item from a host afterwards. such a trigger can be saved. looks like api level check is only on trigger update, not on creation... there shouldn't be duplicate code to check this as well
<Slava>RESOLVED r 24861
<CLOSED>

(5) i believe the following might violate coding style :

+ $hostName=explode(':',$trigger['expression']);
+ info(_s('Created: Trigger "%1$s" on "%2$s".', $trigger['description'], str_replace('{','',reset($hostName))));

<Sasha> CLOSED Fixed with merge from a trunk r24805

(6) for what purposes we receive all fields from the "hosts" table?
frontends/php/api/classes/class.ctrigger.php:2211
<Slava>RESOLVED r 24861

<richlv> something rather fishy was done to this comment, REOPENED
<Slava>RESOLVED r 24875

<richlv> ...
i don't know what was resolved in svn, but the problem still there is with this comment...

<richlv> it should be doublechecked what was done in svn, as the second problem didn't involve code at all. comment is fixed now, original issue can be reviewed

<Sasha> REOPENED
Please review all options in this API call.

<Sasha> CLOSED

(7) one more SQL? We cannot use the existing?
<Slava>RESOLVED r 24861

<richlv> something rather fishy was done to this comment, REOPENED
<Slava>RESOLVED r 24875

<richlv> ...
i don't know what was resolved in svn, but the problem still there is with this comment...

<richlv> it should be doublechecked what was done in svn, as the second problem didn't involve code at all. comment is fixed now, original issue can be reviewed

<pavels> I've remade the fix, so last commit is irrelevant now.

<Sasha> CLOSED

(8) [GUI] rollback all changes in configuration.triggers.edit.php
<Slava> RESOLVED r24829
<Sasha> CLOSED please review a small change in r24866
<Slava>CLOSED

I've remade the fix. Please review.

RESOLVED.

Excellent! Successfully tested!

Merged to trunk r25026.

CLOSED.

(1) while fixing this, please, also remove extra space in "WHERE EXISTS ( SELECT"

<sanja> RESOLVED

<pavels> CLOSED.

(2) additionally, even though there were these errors, frontend message was "Template [dep2] deleted." - shouldn't be

<pavels> This may be a bit tricky to fix now because we'll have to make changes to all of the SELECT queries on a global level. I suggest we leave it as is for now.

<sanja> "Template [dep2] deleted." changed to standart message.

<pavels> CLOSED.

Fixed in development branch svn://svn.zabbix.com/branches/dev/ZBX-4519

TESTED.

Changes from a development branch has been merged to a trunk incorrectly. Please fix it.

<sanja> RESOLVED

<Sasha> TESTED

(3) Message about a deleted template does not correspond Zabbix syntax and naming guidelines.
Message about a deleted host too.

<sanja> RESOLVED

<Sasha> CLOSED Please review my changes in r24878.

TESTED

Trunk (v1.9.9)

Closing duplicated issue.

Reopened incorrectly closed issue

dev branch: svn://svn.zabbix.com/branches/dev/ZBX-4513

(1) The new files should have svn:eol-style set to "native"

<Vedmak> RESOLVED

<pavels> CLOSED.

(2) You'll also need to replace the references to config.php in testFormAdministrationGeneralGUI.php and testFormAdministrationGeneralHousekeeper.php

<Vedmak> RESOLVED

<pavels> CLOSED.

(3) When validating the mapping value field it says "Only positive or negative number". Let's change it to "Only integers" because it also allows zeros. "

<richlv> what about "Only positive or negative numbers or zeroes" ? longish, but we won't have to make all our users learn what an integer is

<pavels> Maybe "Only numbers" or "Only whole numbers"?

<richlv> "Only whole numbers" sounds ok to me

<Vedmak> RESOLVED

<pavels> Since Rich didn't like the word "integer", I've also changed the placeholder to "Whole number" in r24813. Ok with you?

<Vedmak> Sure. CLOSED

(4) Let's leave the testing of the select menu in the *CheckLayout test to make sure it works correctly.

<Vedmak> RESOLVED

<pavels> CLOSED.

TESTED, but please close (3) before merging.

commit: svn://svn.zabbix.com/trunk 24814

this resulted in a regression : ZBX-5344

fix versions missing

i believe it was fixed in 1.9.9, please change if that's not correct

Linux, Debian

Discovery can work very slow for big IP addresses range with so big count of checks and many of unavailable hosts.
Maybe need to wait day or two ?

I'm aware of that, but I have this configuration for about 6 months now. So I think that is more than enough waiting.

If you have a better way of configuring the discovery, please tell me.

Confirmed: We always add discovered host to group "Discovered hosts".

That is a good thing. But I'd like to remove it again.

Because it also shows up as a host group with errors on the dashboard for example.

Fixed in the development branch svn://svn.zabbix.com/branches/dev/ZBX-4507

It's interesting how I finally managed to reproduce the problem only when I split "add host" and "remove from discovered hosts" operations into 2 different actions. Within one action I wasn't able to reproduce it. Yes, but it's true that server was adding the host back to group every time "add host" operation was executed.

Tested. Please review my changes in r24932, r24933.

<Sasha> CLOSED

Fixed in version pre-1.8.11, revision 24989.

Just a note: Fixed in version pre-1.9.9, revision 24993 too.

This will require some API updates, so I've started a specs draft https://zabbix.org/wiki/Docs/specs/ZBX-4503

See also ZBX-4083.

Additional notes from ZBX-4083

"1. The new macro shouldn't be added. The existing macro should be updated.
2. The unchanged macros shouldn't generate auditlog records and SQL statements."

RESOLVED for trunk in /branches/dev/ZBX-4505.

(1) [GUI]
In class.cusermacro.php:523 "$" is missing in message parameter reference.

<pavels> RESOLVED.

<Vedmak> CLOSED

(2) [GUI]
In deleteGlobal method we cannot expect only arrays as we usually support single entity as well..

<pavels> RESOLVED.

<Vedmak> CLOSED

(3) [GUI]
After DB class methods we should not check if it returned false, because all methods of this class throws exceptions on errors.

<pavels> RESOLVED.

<Vedmak> CLOSED

(4) [GUI]
In class.cusermacro.php:563 we try to interact with not defined variable "$dbGmacros".

<pavels> RESOLVED.

<Vedmak> CLOSED

(5) [GUI]
With usermacro.updateGlobal api call we can create two identical macros.

<pavels> RESOLVED.

<Vedmak> CLOSED

(6) [GUI]
Please review my changes in rev. 24642.

<pavels> CLOSED.

Merged to trunk r24661.

CLOSED.

(7) [DOCS]

Updated docs. Global macro update and delete calls are now executed by "globalmacroid" instead of "macro".
http://www.zabbix.com/documentation/2.0/manual/appendix/api/usermacro/deleteglobal
http://www.zabbix.com/documentation/2.0/manual/appendix/api/usermacro/updateglobal

<Vedmak> CLOSED

fixed in svn://svn.zabbix.com/branches/dev/ZBX-4502 along with minor coding style fixes and using direct gettext string for template page

TESTED.

merged to trunk in rev 24496.

resolved in svn://svn.zabbix.com/branches/dev/ZBX-4484

TESTED.

merged to trunk in rev 24287.

I've added a server-side icon resizing feature we use to properly scale icons.

RESOLVED.

(1) [GUI]
There are 3 places we do "imagealphablending($source, false); imagesavealpha($source, true);" for image, maybe we can do it in one place?

<pavels> I've implemented a separate function for it, but we have to do it even if we don't resize the image. RESOLVED.

<pavels> CLOSED.

There have been some conflicts when updating to the latest trunk, plus I've fixed another small bug. Please review.

REOPENED.

<Vedmak> CLOSED

Merged to trunk r24444.

CLOSED.

(1) rev 24503 did :
+ for ($i = 1; $i <=2; $i++) {

is suspect this might not match the coding stye
<Slava> RESOLVED r 24511

• for ($i = 1; $i <=2; $i++) {
  + for ($i = 1; $i <= 2; $i++) {

<richlv> CLOSED

Server side TESTED.

for the record, this is supposed to allow template triggers to depend on host triggers and should also fulfill ZBXNEXT-835

(2) this should be documented in the trigger dependency docs

http://www.zabbix.com/documentation/2.0/manual/config/triggers/dependencies

<Sasha> Rich, thanks! CLOSED

Tested, somebody should review my changes in r24536 and r24547 and then it can be merged into upstream.

<Sasha> CLOSED

Available in version pre-1.9.9, rev. 24570.

Reopened to remove 2.0 from "Fix version"

Closed again.

dev branch: svn://svn.zabbix.com/branches/dev/ZBX-4477

(2) Please review my changes in r24487.

Please review my changes in r24611.

<pavels> CLOSED.

I've merged some changes from ZBX-4504, please review them in r24619 and r24618. If everything is ok, you can merge.

(3) [GUI]
Please review my changes in rev. 24659.

<pavels> Fixed a couple of typos in r24662. CLOSED.

Merged: svn://svn.zabbix.com/trunk 24681

v1.9.9 (rev #24084)

Confirmed with PostgreSQL r24096.

Broken by ZBX-3104 r23677.

RESOLVED.

Successfully tested! Please review my changes in r24224.

<pavels> CLOSED.

Merged to trunk r24231.

CLOSED.

1.9.9 (trunk), 1.8.x

(1) Don't use concatenation in messages, use the _s() function instead.
<Slava>RESOLVED r 24281

<pavels> Not sprintf() but _s(). It's our standard gettext wrapper for strings with arguments.
<Slava> RESOLVED r 24506

<pavels> CLOSED.

(2) The input parameters that we use in SQL queries should always be escaped. Use the zbx_dbstr to escape the name in get_valuemap()
<SlaVa>RESOLVED r 24281

<pavels> CLOSED.

(3) We use upperCamelCase for function names. Plus, I suggest we rename the get_valuemap function to getValuemapByName($name) and make the $name parameter obligatory.
<SlaVa>RESOLVED r 24281

<pavels> One more thing, all the tables in SQL queries must use an alias. Even if there's only one table.
<Slava> RESOLVED r 24506

<pavels> CLOSED.

(4) [GUI] Please review my changes in r24304
<Slava>CLOSED

(5) [GUI] Cannot update value map
ERROR: Cannot add value map. Map with name "Service state" already exist
<Slava>RESOLVED r 24359

<pavels> Still not working. REOPENED.
<Slava> RESOLVED r24490

<pavels> I thing you've meant
isset($_REQUEST['valuemapid']) && bccomp($_REQUEST['valuemapid'], $prevMap['valuemapid']) == 0
intead of
isset($prevMap['valuemapid']) && bccomp($_REQUEST['valuemapid'], $prevMap['valuemapid']) == 0
<Slava>Resolved r 24498

<pavels> CLOSED.

(6) it should be "already exists"
<Slava>fixed
it should be also verified whether the message should have the trailing dot or not (according to the guidelines)
<SlaVa>https://zabbix.org/wiki/Docs/specs/syntax :
2 . Main status messages like "Host added" (displayed on green/red background at the top of the page) should never end with a dot.
RESOLVED

<pavels> CLOSED.

(5) [GUI] "Cannot add or update value map" -->For this on 30.12.2011 has been created separate issue ZBX-4498

<pavels> This fix hasn't been merged yet.
<Slava> fixed in r 24515

(6) Please review my changes in r24512.
<Slava>CLOSED

TESTED.

Please close (6) before merging. Also, after updating to the latest trunk, please, fix (5).

<Slava> fixed in r 24515

(7) merge to trunk has lots of coding guideline violations and modifies several unrelated changelog entries. i would strongly suggest reverting it, reviewing all modified code for coding style and doublechecking what it does to the changelog

edit : ok, changelog probably was intentional. why wasn't it mentioned in the commit message ?
<Slava> Sorry. I must recommit this with detailed description ?

<richlv> well, this time i'd say that it would be enough to fix formatting for all modified lines - but in future, describing all changes in the commit message would be highly appreciated
<Slava> You are so kind. RESOLVED r 24542

<richlv> 24542 has a commit message "updated to latest trunk "...
<Slava>Formating error starts after conflict resolving in 'valuemap.inc.php' . Now i re-resolve this conflicts.

<pavels> CLOSED.

<Slava> Pavel, please review my confict resolving in 24542 'valuemap.inc.php'

<pavels> CLOSED. Before merging to trunk, please review my changes in r24578 and r24579.
<Slava> CLOSED

fixed in trunk in 24575

Fix for 1.8 is TESTED.

Please review my changes before merging.
<Slava> CLOSED

fixed in 1.8 r 24593

Reopened to replace 2.0 to 1.9.9 in the Fix version/s

Closed again.

Zabbix-1.9.9-23907
CentOS 6.0
MySQL 5.5.19

Fixed in development branch svn://svn.zabbix.com/branches/dev/ZBX-4467

(2) as discussed with zalex, string change that only changed [] to "" should also be reflected in all translations that had those strings translated. it should be made so that these translations are not lost (both msgid and msgstr must be properly changed). it should not be changed for languages that didn't have those strings translated.
if in doubt, consult translation maintainers : https://zabbix.org/wiki/Translating_Zabbix

<pavels> But he didn't change only of those strings. Or am I missing something?

<richlv> this was my failure, commented on the wrong issue. added (30) to ZBX-2737 instead, CLOSED here

(3) Since we have a nice _n() function, let's use it in graphs.inc.php:378 and :382.
<sanja> RESOLVED

<pavels> CLOSED.

Merged to trunk r25017.

CLOSED.

(4) regression. set ZBX_MAX_PERIOD to 10 years, have 5 years worth of data. in a graph, click "all" - it goes to two years. try this in rev 25016 - it goes to whatever amount of data you have

<pavels> RESOLVED in /branches/dev/ZBX-4467.

<richlv> functionally seems to work as expected again, CLOSED

(5) Will be nice to crop slider till max period size in event when user make it bigger.

<pavels> I've talked to Alexander about that, and we've decided to fix this as a separate issue later. CLOSED.

Tested

The fix for the maximum period has been merged to trunk r25792.

CLOSED.

In cases like this the "Fix version/s" should contain two values: 1.9.9 and 1.9.10 because both releases have changes related to this issue.
REOPENED

closed again with two values

dev branch: svn://svn.zabbix.com/branches/dev/ZBX-4458

(1) Please review my commit r24162.

<Vedmak> CLOSED

Update to my commit 24163 and then you can merge.

TESTED.

MERGED: svn://svn.zabbix.com/trunk 24167

(2) i don't like the following :
return confirm('<?php echo _('Are you sure you want to delete'); ?> 'removedCount' <?php echo _('macro(s)') ?>?');

we are composing a message for translation from 2 strings, which makes it hard to translate for many languages.
if we can't use placeholders in js, we should use the S_ thingie, which should allow placeholders

<Vedmak> How S_ thingie can help with placeholders?
I think we can't do this without any js gettext implementations.

<richlv> right, got confused
CLOSED

dev branch: svn://svn.zabbix.com/branches/dev/ZBX-4456

(1) If an error occurred, then all of the values we've entered are lost. We should probably keep them in the form.

<Vedmak> RESOLVED

<pavels> CLOSED.

(2) If we leave the time field empty, the error message looks like this "Error adding service time ":". Should be from 00:00 to 24:00.". The ":" part is kind of ugly.

<Vedmak> RESOLVED

<pavels> CLOSED.

(3) Fixed a typo in r24212. Please review.

<Vedmak> CLOSED

Fixed a bracket in r24273 =)

TESTED.

merged: svn://svn.zabbix.com/trunk 24279

Fixed for 1.8.

RESOLVED.

Merged to trunk r24126.

Merged to 1.8 r24315.

CLOSED.

I believe it is no longer the case in latest trunk (rev 25169) . Can we close it?

Yep, seems it was fixed.

RESOLVED.

Merged to trunk revision 24165.

CLOSED.

Trunk rev #24012.

RESOLVED.

(1) Cannot update a proxy (same error message)
Steps to reproduce:

• create active proxy
• open its and change a type to Passive
• save

<pavels> RESOLVED in 24059.
<Sasha> REOPENED frontends/php/api/classes/class.cproxy.php:385 The interface can be only one. What for here 'foreach'?

<pavels> Right. Anyway, I think I'll fix it a bit differently.

<pavels> RESOLVED.

<Sasha> REOPENED frontends/php/api/classes/class.cproxy.php:403,451 It is necessary to remove and from these places 'foreach'.

<pavels> RESOLVED.
<Sasha> CLOSED

(2) The interface remains not deleted if we change type to "active"

After that we can't make it passive again:
Host cannot have more than one default interface of the same type. [CProxy.update -> CHostInterface.create ->
CHostInterface.checkMainInterfacesOnCreate -> CHostInterface.checkMainInterfaces]

<pavels> RESOLVED.
<Sasha> REOPENED the active proxy shouldn't have an interface

<pavels> RESOLVED.
<Sasha> CLOSED

(3) Formatting and coding style:
frontends/php/api/classes/class.chostinterface.php:866 the field 'status' is not used anywhere
frontends/php/include/views/administration.proxy.edit.php:32 unused variable $interfaces
frontends/php/include/views/administration.proxy.edit.php:77:80 formatting

<pavels> RESOLVED.
<Sasha> CLOSED

(4) The proxy should have a main interface

<pavels> RESOLVED.
<Sasha> CLOSED

(5) Formatting and coding style:
frontends/php/api/classes/class.cproxy.php:359 'main' not boolean variable; true -> INERFACE_PRIMARY
frontends/php/api/classes/class.cproxy.php:450 in such cases we should use API_OUTPUT_REFER

<pavels> CLOSED.

Tested successfully. Before merge please fix (5)

Merged to trunk r24139.

CLOSED.

RESOLVED.

May this fix help for ZBX-4412 ?

<pavels> No, it doesn't affect the "last changed" column, only the duration.

TESTED. Please review my changes in r24043.

Merged to trunk r24057.

CLOSED.

Fixed in /branches/dev/ZBX-4427. RESOLVED.

(1)
I removed two .htaccess files, if it's ok, can be merged.

<pavels> OK. CLOSED.

Merged to trunk r24075. CLOSED.

(1) i believe this is the 3rd time for the same bug... unittests

<pavels> I've added a simple create/retrieve/delete unit test for the item API, plus a special test for this case ). RESOLVED.

RESOLVED.

(2) [GUI]
Incorrect formatting of "if" statements, and some logic improvements we discussed.

<pavels> RESOLVED.

<Vedmak> we decided to have next line in long "ifs" to be indented with two tabs.

<pavels> RESOLVED.

<Vedmak> CLOSED

Merged to trunk r24033. CLOSED.

When full-cloning a host there is a following error:
"Item uses host interface from non-parent host."

https://support.zabbix.com/secure/attachment/17695/cannot_add_host.jpg

<pavels> RESOLVED in /branches/dev/ZBX-4425 r24121. This was actually broken in ZBX-3771.

<sasha> It do not agree! It's broken in r24033. In r23994 all work fine!

<pavels> Item prototype cloning never supported interfaces, Prior to r24033 it, probably, just haven't shown any errors =)

<pavels> CLOSED.

(1) frontends/php/api/classes/class.cdiscoveryrule.php:1080 In all places, except calculated item and trigger expressions, should be used 'Visible host name'

<pavels> RESOLVED.
<Sasha> TESTED Please review my changes in r24152. I have corrected some more messages.

<pavels> CLOSED.

(2) Please review my changes in r24136

<pavels> CLOSED.

Tested successfully! Please close point (1) before merge

<pavels> There have been some conflicts when I merged the changes from trunk. Please review my commit r24156.

<Sasha> RESOLVED I have corrected incorrectly merged changes. Please review them.

<pavels> Thanks. CLOSED.

Merged the prototype interface fix to the trunk in r24203.

CLOSED.

reopen to change labels

RHEL CentOS 5.6, 6.0 Postgresql-8.4.7

Fixed in the development branch svn://svn.zabbix.com/branches/dev/ZBX-4424

Please review my changes in r23928.

<Sasha> Great! CLOSED

Fixed in version pre-1.8.10, revision 23950.

FreeBSD srv-zabbix01.trendeiras 8.2-RELEASE-p4 FreeBSD 8.2-RELEASE-p4 #5: Mon Dec 5 07:57:08 BRST 2011 [email protected]:/usr/obj/usr/src/sys/TREND amd64

FreeBSD inside vmware, typical instalation.

(root@srv-zabbix01 - ~ @14:29:40)
1: pkg_info |grep mysql
mysql-client-5.5.17 Multithreaded SQL database (client)
mysql-server-5.5.17 Multithreaded SQL database (server)
php5-mysql-5.3.8 The mysql shared extension for php

./configure --prefix=/usr/local --sysconfdir=/usr/local/etc/zabbix --enable-server --enable-agent --with-mysql --with-libcurl --with-net-snmp --with-ssh2 --with-ldap --with-openipmi

Fixed in the development branch svn://svn.zabbix.com/branches/dev/ZBX-4418

Tested.

Fixed in version pre-1.9.9, revision 23920.

Maybe it has been already fixed in a dev branch here:
https://support.zabbix.com/browse/ZBX-3366?focusedCommentId=51208&page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel#comment-51208

It will be fixed with ZBX-3366. I'm closing duplicated issue.

PostgreSQL 8.4, RHEL 5.5-based OS, Firefox 3.0.19

Trunk 23633 - search working.

Hi,

I can not reproduce the same problem in v1.9.9, rev #23645.

Can you please upgrade and try this revision ?

also - what theme is that ? does it work with english locale ? do you have any errors in the apache log ?

Rich, the theme is "Black & Blue". It doesn't work with English locale and default theme either. There are no any errors in the apache log.

I'll try http://www.zabbix.com/downloads/nightly/pre-zabbix-1.9.9.x-23657.tar.gz later today and post the result here.

Can this problem be related to Postgres? Some strange bugs are caused by Zabbix incompatibility with Postgres, for example, icons corruption after xml map import. Did anybody test search in trunk with Postgres backend?

Hi,

Just for the record:

For me search is working fine in v1.9.9 (rev #23645) when using PosgreSQL DB and using "Black & Blue" theme with English locale.

Need to debug this specific case more deeper.

Compilation of http://www.zabbix.com/downloads/nightly/pre-zabbix-1.9.9.x-23657.tar.gz fails.

<---cut--->
Making all in housekeeper
make[3]: Entering directory `/usr/src/zabbix_1.9.9/zabbix-1.9.9/src/zabbix_server/housekeeper'
x86_64-linux-gnu-gcc -DHAVE_CONFIG_H -I. -I../../../include -Wall -g -O2 -I/usr/include/postgresql -I/usr/include -I/usr/local/include -I/usr/lib/perl/5.10/CORE -I. -I/usr/include -I/usr/include -I/usr/include -I/usr/include -MT housekeeper.o -MD -MP -MF .deps/housekeeper.Tpo -c -o housekeeper.o housekeeper.c
housekeeper.c: In function ‘housekeeping_cleanup’:
housekeeper.c:135: error: ‘ids_alloc’ undeclared (first use in this function)
housekeeper.c:135: error: (Each undeclared identifier is reported only once
housekeeper.c:135: error: for each function it appears in.)
housekeeper.c:135: error: ‘ids_num’ undeclared (first use in this function)
housekeeper.c: In function ‘housekeeping_alerts’:
housekeeper.c:186: warning: ‘return’ with a value, in function returning void
housekeeper.c: In function ‘housekeeping_events’:
housekeeper.c:213: warning: ‘return’ with a value, in function returning void
housekeeper.c: In function ‘main_housekeeper_loop’:
housekeeper.c:328: error: void value not ignored as it ought to be
housekeeper.c:331: error: void value not ignored as it ought to be
make[3]: *** [housekeeper.o] Ошибка 1
make[3]: Leaving directory `/usr/src/zabbix_1.9.9/zabbix-1.9.9/src/zabbix_server/housekeeper'
make[2]: *** [all-recursive] Ошибка 1
make[2]: Leaving directory `/usr/src/zabbix_1.9.9/zabbix-1.9.9/src/zabbix_server'
make[1]: *** [all-recursive] Ошибка 1
make[1]: Leaving directory `/usr/src/zabbix_1.9.9/zabbix-1.9.9/src'
make: *** [all-recursive] Ошибка 1
<---cut--->

I have taken frontend and database data files from the above nightly build, and have the same bug, see new screenshot.
It is interesting that if nothing is found, Zabbix generates web page as expected.

Hi,

Compilation errors "error: void value not ignored as it ought to be" were fixed today in trunk branch, in rev #23663.
Compilation problems were fixed in ZBX-4298.

Concerning not working global search- please confirm the same behaviour (not working search) in the latest trunk revision.
As already told, I can not confirm the same in v1.9.9 (rev #23645) when using PosgreSQL DB and using "Black & Blue" theme with English locale.

Igor, I experience the same problem with search in 1.9.9. 23674, original light blue theme and English locale.

Sergey,

Do you have any relevant messages in the apache error log?

Ooops! I'm sorry, I have looked Apache logs on another virtual machine Yes, there actually is an error:
PHP Fatal error: Call to undefined function mb_stripos() in /usr/share/zabbix/include/func.inc.php on line 855

My PHP version is php-5.1.6-27.el5.x86_64.rpm. Apache is httpd-2.2.3-43.el5.centos.x86_64.rpm.

Sergey,

Seems that this error means that there are no "mbstring" module.

Can you please check that you have PHP "mbstring" module enabled ?

"Multibyte Support" should be enabled.

Please let us know about the results after that.

php-mbcstring-5.1.6-27.x86_64.rpm is installed.
mbstring.so in enabled in /etc/php.d/mbstring.ini
Zabbix installation wizard says that PHP mbstring is OK.

Please could you replace:

function mbstrings_available()

by

function mbstrings_available()

in include/locales.inc.php and tell us if it works or not. Thanks!

It looks like you mbstring module has some functions missing.

According to Apache log my mbstring module lacks 'mb_stripos' function.

I have found the following workaround for this problem;

function zbx_stripos($haystack, $needle, $offset = 0) {
if (defined('ZBX_MBSTRINGS_ENABLED')) {

• return mb_stripos($haystack, $needle, $offset);
  + return mb_strpos(mb_convert_case($haystack, MB_CASE_LOWER), mb_convert_case($needle, MB_CASE_LOWER), $offset);

The mb_stripos() function was implemented only in PHP 5.2. Maybe you have an older PHP installation. Anyway, since we support PHP 5.1.6, I've replaced it the way Sergey proposed.

<richlv> as per https://support.zabbix.com/browse/ZBX-4411?focusedCommentId=51255&page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel#comment-51255 , that's indeed 5.1.6

<pavels> RESOLVED.

just a note - we don't seem to use mb_stripos in 1.8, so no fix needed for it - although we might have to be a bit more careful when relying on new functions

Merged to trunk r23828.

CLOSED.

apache2.2

Version 1.9 has new key format. You need to change keys manually.

More info about these changes are available here: http://www.zabbix.com/documentation/2.0/manual/config/items/itemtypes/4_simple_checks

awwww. we just have to support 1.8 xml as an absolute minimum, otherwise most templates would be unusable.

db upgrade patch converts such keys - frontend xml legacy import code (for 1.8) must convert those keys in the exact same way

Fixed in development branch svn://svn.zabbix.com/branches/dev/ZBX-4407

(1) it changes (which is great) :
$host_db['profile_mode'] = isset($host_db['profile']) ? HOST_PROFILE_MANUAL : HOST_PROFILE_DISABLED;
$host_db['profile_mode'] = isset($host_db['profile']) ? HOST_INVENTORY_MANUAL : HOST_INVENTORY_DISABLED;

but maybe should also change profile_mode and "profile" ?

<pavels> I think $host_db['profile_mode'] should be changed to $host_db['inventory_mode'], but double check it please.

<sanja> RESOLVED

<richlv> i believe "isset($host_db['profile'])" should also be changed to 'inventory', but it should be doublechecked

<sanja> RESOLVED

<richlv>
// HOST PROFILES {{

}} HOST PROFILES

<sanja> RESOLVED

<pavels> It's great that we fixed the notice, but profile data import doesn't work at all. Please, have a look.
<richlv> i split this out as (6)

<richlv> this is where i'd like to refer to DeepC presentation slides again
it would help us all if we all would try to understand what the issue is and fix it fully, properly and finally. in this case we did an issue pingpong for "profile_mode" and "profile" separately. then another round for "HOST PROFILES" - in this case it was even visible in the diff context for the two previous changes.

as such i would like to suggest using svn diff, and attempt to fix the issue so nobody could come and point at things missed (like i'm doing now )
i was ready to split out remaining renames in another issue, but as pavel noted that profile import doesn't work, let's do it here instead.

so instead of pointing at specific instances, let's just remove any reference to host profiles (except for 1.8 xml format, of course). please note that we shouldn't nuke user profiles in any way

<pavels> Yeah, that's I meant by "double check it please" =) I haven't tested the fix since then, so I've noticed the problem just now. Anyway, CLOSED.

<richlv> as discussed, let's be pedantic enough and clear this up completely
as in, no references to host profile should be anywhere in the code, they all should refer to "inventory" except 1.8 xml fields

<sanja> Can not find any references! Maybe You have found some before reopening this issue?

<richlv> there might be more that i missed, but here's one file with them :

./include/maps.inc.php: $hostsToGetProfiles = array();
./include/maps.inc.php: // if we have icon map applied, we need to get profiles for all hosts,
./include/maps.inc.php: $hostsToGetProfiles[] = $selement['elementid'];
./include/maps.inc.php: 'hostids' => $hostsToGetProfiles,
./include/maps.inc.php: $profiles = getHostInventories();

<sanja> Fixed that and 1 more.
RESOLVED

(2) function convertOldSimpleKey seems to be defined both in class.ctriggerexpression.php and export.inc.php - that scares me a bit... wouldn't it be better to define it only once ?

<pavels> Yes, that's very very wrong. Since this conversion is required only when importing, I suggest we move it into a protected method of zbxXML class and only perform the conversion there. Plus, it shouldn't be done when creating new triggers in the trunk, only when importing old ones.

<sanja> RESOLVED

<pavels> CLOSED.

(4) Regarding in class.ctriggerexpression.php:235. I know we sometimes use references in foreach loops, but mostly in simple loops where we just need to perform some modifications on the arrays, like unsetting some keys. In more complex loops like this one, it's best not to use references, as it may cause unexpected problems.

<sanja> RESOLVED

<pavels> CLOSED.

(5) "Converts Simle key"
i also believe "$keyWithotPerf" is a typo
export.inc.php has lost trailing newline

<sanja> RESOLVED

<richlv> CLOSED

(6) <pavels> It's great that we fixed the notice, but profile data import doesn't work at all. Please, have a look.

<sanja> RESOLVED

<pavels> The new line character should be "\r\n" to work properly on mac and windows. And there are no name conflicts between host profile fields, so we don't need to combine them there. I've corrected it, please review r24209 and r24210. Otherwise, good.

<sanja> CLOSED

TESTED.

Before merging, please close (6).

we got a workflow problem here

rev 24216 merged this dev branch to trunk already, but rev 24221 did changes in the dev branch after that. these changes will be lost.

dev branch should be deleted, recreated, changes from rev 24221 reapplied and problem (1) fully fixed

<sanja> RESOLVED

<Sasha> REOPENED
in a new recreated branch no changes from a deleted ' svn di svn://svn.zabbix.com/branches/dev/ZBX-4407@24225 -c24221 ' branch

<sanja> RESOLVED

<Sasha> CLOSED

Pavels, please retest its branch. There are some changes.

TESTED again. Everything ok.

reopen to set proper "fix version"

Centos

Thanks for reporting this. It should be fixed.

Fixed the default zabbix_server.conf in version pre-1.8.11, r24572.

Confirmed this problem. The same error "ERROR: Cannot add discovery rule. Incorrect arguments passed to function [CDiscoveryRule.create -> CDiscoveryRule.checkInput -> CItemGeneral.checkInput]" appears also in the latest trunk revision #23583.

This was NOT broken by fixing issue ZBX-3510 in rev #23147.

It was broken in ZBX-3625 r23438. RESOLVED.

<Igor> This dev branch tested using Mysql DB, it is working as desired.
<Igor> Dev branch ZBX-4403 tested also with PostgreSQL DB, it is working as desired.

(1) The value type of the discovery rule always should be ITEM_VALUE_TYPE_TEXT.

<pavels> RESOLVED
<sasha> CLOSED

Merged to trunk revision r23661.

CLOSED.

Fixed in development branch svn://svn.zabbix.com/branches/dev/ZBX-4391

(1) [GUI]
Not all suffixes are working (http://www.zabbix.com/documentation/1.8/manual/config/suffixes)

<slava> RESOLVED

<Vedmak> as decided we need support up to Y suffix.
<SlaVa> RESOLVED
<Vedmak> CLOSED

(3) [GUI]
I think regular expression could be simplyfied to smth like: "/([0-9\.]+)[K|M|G]?/" also case is significant, as "K" is not the same as "k"
<slava> RESOLVED
<Vedmak> in fact regexp can be simplified even more: "/[0-9\.]+[KMG]?/"
<SlaVa> RESOLVED
<Vedmak> CLOSED

"Fixed in development branch svn://svn.zabbix.com/branches/dev/ZBX-4391"

<zalex> dev branch tested. works as expected.

(0) is not solved

(4) a trigger expression parser on server side do not support suffixes P (Peta), E (Exa), Z (Zetta) and Y (Yotta). Please, rollback a support of these suffixes on GUI side in the trigger expressions.

<Slava> RESOLVED

sasha Has been fixed in ^/trunk r24449. CLOSED

Fixed in 24150

(5) [GUI]
In function "convert" there is still code for converting to yotta etc.
<Slava>RESOLVED r 24185

<Vedmak> CLOSED

(6) [GUI]
Why do you use in test both "button_click" and "click" methods?
Also formatting in test file should be fixed.
<Slava>RESOLVED r 24198

<Vedmak> CLOSED

(7) new file, testTriggerExpressions.php, should have svn:eol-style svn property set to 'native'
<SlaVa>RESOLVED r 24236

<richlv> CLOSED

(8) [GUI]
In trunk branch all suffixes like Y, Z, etc. are still acceptable in trigger expressions, is it ok?
<SlaVa> No, it's not ok. RESOLVED r 24282

<Vedmak> CLOSED

(9) [GUI]
Suffixes are left in defines.inc.php and maybe somewhere else..
<Slava>RESOLVED

<Vedmak> CLOSED

fixed in r 24449

(10) Missing record for the trunk ChangeLog.
1.9.9 value is missing in the "Fix Version/s:"
Will be REOPENED.
<SlaVa>
Changes for 1.8.10rc1

..F....... ZBX-4391 fixed suffix support in trigger testing interface (slava)

<zalex> I've added the tag 1.9.9 to the "Fix Version/s:"
I'm closing it. ChangeLog processing will be discussed separately.
CLOSED