Complete HP server hardware monitoring ZBX2

[omidta]

Some feedback

I imported HP template on zabbix 2.0.4 without any problem but there are just discovered 4 items in HP system and 12 Items in HP Eventlog and zabbix doesn't discover any item in HP SmartArray,HP Sensors and etc
Please Help
My servers are DL380G7 and DL360G4

[jimma1972]

omidta

You probably need to install the HP SmartArray software to the servers... see below from mmarkwitzz first post.

Go to http://support.hp.com and get the HP Smart Array Configuration Utility and HP Smart Array Configuration Utility CLI for your server model from the downloads section. After installation you must run the Setup HP Array Configuration utility and selec the local application mode.
This step is optional, but allows monitoring the physical disks temperature, if the disk/controller firmware supports it (you can always try to upgrade the firmware).

When I didnt get feedback from the servers its was because the proper software was not installed on them.

[omidta]

omidta

You probably need to install the HP SmartArray software to the servers... see below from mmarkwitzz first post.



When I didnt get feedback from the servers its was because the proper software was not installed on them.

Thank you
my problem's been solved . Zabbix agent wasn't in Active mode.Your template is quite useful for me thanks.

[djizno]

Error creating Graph

Hi,

Using Zabbix server 2.0.5.

Launching manualy your perl script throw me an error creating graph : {"jsonrpc":"2.0","error":{"code":-32602,"message":"Invalid params.","data":"Missing items for graph \"HP Physical disk \"ALL\" temperature\"."},"id":4}

Can you help me ?

Regards

[owlmind]

unexpected end of string while parsing JSON string,

/etc/cron.hourly/zabbix
unexpected end of string while parsing JSON string, at character offset 5 (before "ONT-WEIGHT: bold; FO...") at /etc/zabbix/hp_sensor_all_temperature.pl line 62
unexpected end of string while parsing JSON string, at character offset 5 (before "ONT-WEIGHT: bold; FO...") at /etc/zabbix/hp_phydisk_all_temperature.pl line 62


What wrong?

[WarBull]

zabbix/hp_sensor_all_temperature.pl line 62

You need to correct path on this file on line 344.
For example, my line is:
$url = "http://127.0.0.1/zabbix/api_jsonrpc.php";

[WarBull]

It can not work!

Dear collegues, i do not mind! Dont work it. I have HP Proliant DL 380 G5
Windows 2003 R2 x64 SP2.

Zabbix 2.0 on Server and on Client. I correct a config file:

#hp discovery
#
UserParameter=hp.discovery[*],cscript "C:\Program Files (x86)\Zabbix Agent\zabbix_hp_discovery.vbs" //Nologo "$1"
#hp wbem
UserParameter=hp.wbem[*],cscript "C:\Program Files (x86)\Zabbix Agent\zabbix_hp_wbem.vbs" //Nologo "$1" "$2" "$3"
#hp system
UserParameter=hp.system[*],cscript C:\Program Files (x86)\Zabbix Agent\zabbix_hp_system.vbs" //Nologo "$1"
#eventlog query
UserParameter=eventlog.query[*],cscript "C:\Program Files (x86)\Zabbix Agent\zabbix_win_eventlog.vbs" //Nologo "$1" "$2"
# allow weird chars in userparameters arguments
UnsafeUserParameters=1

Install: cp014513_HP_WBEM_x64.exe v. 2.8.8.0
HP Array Configuration Utility (64-bit)
HP Array Configuration Utility CLI (64-bit)

If i puts this line in CMD - hp.system[*],cscript C:\Program Files (x86)\Zabbix Agent\zabbix_hp_system.vbs" //Nologo "$1"
have not errors.

But my user parameters dont work! I have not information about WBEM and HP in Zabbix_agentd.log.
And on my template on zabbix server i have not new issues!
Help me please!

[nick0909]

Awesome template! This is the first time I have really seen discovery used to what I would call the fullest extent. However, I am trying to figure out how the temperature triggers are being set. Right now I have several servers that are reporting high temps in zabbix but HP SIM does not seem concerned.

It looks like the zabbix triggers are using the highest temp reading and then saying anything within 80% of that is too high? Or am I wrong on that? I am trying to figure out the best way to edit the template so I can get rid of these alarms but still have useful alarms if needed.

Also I see that it is sensing the temp status and if it = 6 (shutdown) it alerts. But if a temp sensor hits 6 doesn't the system shut down right away, and the value would never be reported to Zabbix because of that? Or is there a delay where it might get picked up by the agent before shutdown?

Thanks!

[angel_of_devil]

I followed the instruction without any errors. But nothing happens with this template, can you tell me how to check this template working successfully?

[mmarkwitzz]

I followed the instruction without any errors. But nothing happens with this template, can you tell me how to check this template working successfully?

To be perfectly honest with you, I don't remeber very well. I switched workplaces and we use SCOM now.

You should run cscript zabbix_hp_discovery.vbs PHYDISKS

If all is ok, you should get a JSON with the discovered disk. If you get an error instead of the JSON make sure you installed the HP WBEM providers for your server.

If you DO get the JSON, make sure you have properly activated ACTIVE checks in zabbix.conf.

"The template included uses active checks. Your zabbix agent must be configured for active checks otherwise the server won't receive any data.. " -- google it.

If you are certain you have active checks enabled, enable debug logging and restart the daemon and check the logs if the discovery is actually run.

[mmarkwitzz]

Awesome template! This is the first time I have really seen discovery used to what I would call the fullest extent. However, I am trying to figure out how the temperature triggers are being set. Right now I have several servers that are reporting high temps in zabbix but HP SIM does not seem concerned.

It looks like the zabbix triggers are using the highest temp reading and then saying anything within 80% of that is too high? Or am I wrong on that? I am trying to figure out the best way to edit the template so I can get rid of these alarms but still have useful alarms if needed.

Also I see that it is sensing the temp status and if it = 6 (shutdown) it alerts. But if a temp sensor hits 6 doesn't the system shut down right away, and the value would never be reported to Zabbix because of that? Or is there a delay where it might get picked up by the agent before shutdown?

Thanks!

Nick,
I no longer have a zabbix environment available so I cannot view my own templates unfortunately. If I remember correctly I compare the current temperature with 90% of the maximum allowed temperature. I have some issues with these too because it's hard to get a correct value for each sensor.

"{HP Server:hp.wbem[sensor,temp,{#SENSORNO}].avg(1800)}>90/100 * {HP Server:hp.wbem[sensor,maxtemp,{#SENSORNO}].last(1800)}"

The best thing you could do it replace the hardcoded 80% value with a marco {$PROCENT} and override the value on each of your hosts to your liking.

"{HP Server:hp.wbem[sensor,temp,{#SENSORNO}].avg(1800)}>{$PROCENT}/100 * {HP Server:hp.wbem[sensor,maxtemp,{#SENSORNO}].last(1800)}

This is better than nothing, but you would still have ONE value for all sensors.


Regarding the =6 issue.. HP documentation is lacking at best. I can only speculate that the server would shutdown itself and the trigger is useless. But just to be safe, let's assume it does not shutdown. At lest we get an alert. Luckily I did not have any overheating scenarios until how.

[angel_of_devil]

To be perfectly honest with you, I don't remeber very well. I switched workplaces and we use SCOM now.

You should run cscript zabbix_hp_discovery.vbs PHYDISKS

If all is ok, you should get a JSON with the discovered disk. If you get an error instead of the JSON make sure you installed the HP WBEM providers for your server.

If you DO get the JSON, make sure you have properly activated ACTIVE checks in zabbix.conf.

"The template included uses active checks. Your zabbix agent must be configured for active checks otherwise the server won't receive any data.. " -- google it.

If you are certain you have active checks enabled, enable debug logging and restart the daemon and check the logs if the discovery is actually run.

Dear mmarkwitzz,

I'm sure about active checks on host configuration, the parameter "ServerActive" have been set to 127.0.0.1
I have installed requirement app as your guide:

But when I run script, it told me that HP WBEM Provider haven't installed

I don't have any problems while installing HP Packages, and there's no restart requirement after installing them. I haven't restart server yet because it has many virtual servers on it.

My Zabbix server and agent: both 2.0.2 version
Host: Windows 2012 Server Data Center version
And there's one thing I don't understand: there's no zabbix_agentd.log file in my Zabbix Agent folder, although I have enable Log in zabbix_agentd.conf

Please tell me how to solve this, tks a lot

[mmarkwitzz]

Dear mmarkwitzz,

I'm sure about active checks on host configuration, the parameter "ServerActive" have been set to 127.0.0.1
I have installed requirement app as your guide:

But when I run script, it told me that HP WBEM Provider haven't installed

I don't have any problems while installing HP Packages, and there's no restart requirement after installing them. I haven't restart server yet because it has many virtual servers on it.

My Zabbix server and agent: both 2.0.2 version
Host: Windows 2012 Server Data Center version
And there's one thing I don't understand: there's no zabbix_agentd.log file in my Zabbix Agent folder, although I have enable Log in zabbix_agentd.conf

Please tell me how to solve this, tks a lot

The config file should look something like this:

Server=<LOCAL IP>
ServerActive=<ZABBIX SERVER>:10051
StartAgents=1
Hostname=HOSTNAME
LogFile=C:\Program Files\Zabbix agent\zabbix_agentd.log
LogFileSize=10
UnsafeUserParameters=1
Timeout=30
DebugLevel=3

So your config file is wrong.

[angel_of_devil]

The config file should look something like this:

Server=<LOCAL IP>
ServerActive=<ZABBIX SERVER>:10051
StartAgents=1
Hostname=HOSTNAME
LogFile=C:\Program Files\Zabbix agent\zabbix_agentd.log
LogFileSize=10
UnsafeUserParameters=1
Timeout=30
DebugLevel=3

So your config file is wrong.

Thank you so much. Now I understand about wrong thought of active checks configuration. I tested on a Windows 2008 R2 host and it worked successfully. The remaining problem is having no zabbix_agentd.log file in the log file location as set in zabbix_agentd.conf

I will check again on the Windows 2012 Server host later with the error of the HP WBEM haven't installed although I installed.

Thanks again

[angel_of_devil]

Everything's ok now. Log file zabbix_agentd.log appear too. Thank you so much, mmarkwitzz