HP P2000 G3 monitoring (Zabbix 2.0)

[elvar]

Template update with disk and enclosures status monitoring support

Woot! Can't wait to try it out.

Thanks!

[avk]

I updated this version.
./hp-msa.pl fqdn lld returned data, but stats returned 0 and LATEST DATA was no longer change.
I installed the old version of the script. ./hp-msa.pl fqdn lld returned data and stats returned 0 and LATEST DATA was started change.

[dish]

I updated this version.
./hp-msa.pl fqdn lld returned data, but stats returned 0 and LATEST DATA was no longer change.
I installed the old version of the script. ./hp-msa.pl fqdn lld returned data and stats returned 0 and LATEST DATA was started change.

Discovery rule have 1 day delay, try decrease (600 for example) it and disconnect and clear template from the host.

You can comment string in file hp-msa.pl
# die "Can not remove file $sendFile!" unless(unlink ($sendFile));
and check data in file /var/tmp/zabbixSenderHPP2000_hostip...

[elvar]

I updated this version.
./hp-msa.pl fqdn lld returned data, but stats returned 0 and LATEST DATA was no longer change.
I installed the old version of the script. ./hp-msa.pl fqdn lld returned data and stats returned 0 and LATEST DATA was started change.

Yes I'm having problems with the new version as well.

[elvar]

Template update with disk and enclosures status monitoring support

I'm seeing the following with this version...

hp-msa.pl["{HOST.DNS1}","stats"]] error: Received value [] is not suitable for value type [Numeric (unsigned)] and data type [Decimal]

Edit1: My discovery item shows this error - "Value should be a JSON object."

[MyL0]

ssl

One problem:

LWP::Protocol::https::Socket: SSL connect attempt failed because of handshake problems error:140943FC:SSL routines:SSL3_READ_BYTES:sslv3 alert bad record mac at /usr/local/share/perl/5.14.2/LWP/Protocol/http.pm line 41.

Solution:
Insert "use Net::SSL;" on top of script (install by cpan or other)

and modify

my $ua = LWP::UserAgent->new;

by

my $ua = LWP::UserAgent->new(ssl_opts => { verify_hostname => 0 });


But my problem is the stat = 0 ...

Thanks

[elvar]

I managed to get an updated version working however, I do not think Dish's updated version of the original script & template will work due to what I believe is the inability to mix and match macro names (thanks to Volter for pointing this out) such as in the example at https://www.zabbix.com/documentation...evel_discovery. In their example they are all the same down each column.

Attached is an updated version of the original script provided by OP which just contains the extra enclosure and disk elements. I then added the following prototype item...

Name: $1 $2 health
Type: Zabbix trapper
Key: hp.p2000.stats[{#HP_P2000_TYPE},{#HP_P2000_ID},health]
Type of Information: Text

Regards,

[AmPheT]

Hi,

Thanks for your script, i tried it today and it works fine except i have some trouble with controllers and enclosure status.

I can see the items so that means discovery is ok but no data for those, the others items like disk, speed etc.. are ok. I have no errors from zabbix server logs, no timeout, i am monitoring two differents enclosure with two differents firmware and same for both.

Thanks for your help.

[vymysleny]

Import problem

Hello
When I imported template to zabbix 2.2 , 7 application folder was created , but only one item. As I saw on one screen, there have to be lot of paarameters (and according XML too). Is there any problem with importing 2.0 template to 2.2 zabbix.

[nilldot]

from log>> timeout while executing a shell

HP P2000 MSM SAS:hp-msa.pl["{HOST.CONN}","lld"]" became not supported: timeout while executing a shell script


HP P2000 MSM SAS:hp-msa.pl["{HOST.CONN}"]" changed: time-out while executing a shell script



I though something wrong with {HOST.X} part, thus replaced this part with IP address, still no luck.

Any help?
Thanks

Env.
Zabbix 2.2.6
CentOS release 6.5 (Final)

####
Update


Timeout were sorted out by increasing timeout value to 30 in the zabbix config file. Though, I still can see some timeouts.

But all my graphs are empty!

On emore thing, I can't see file /var/tmp/zabbixSender* being created!

[ViktorJJ]

Hp msa 2000 g2

Just a note:
to get authenthicated on MSA G2 arrays (2312, 2324, etc.) you need add a cookie to request:

$req->header('Cookie' => "wbiusername=$USERNAME; wbisessionkey=$sessionKey" );

and there is no performance data in G2 arrays API.

[winsash]

stats results 0

Hello all!

My system is:
• OS: Ubuntu 14.04
• Zabbix Server 2.4.3
• Zabbix-Sender 2.4.3
• Zabbig-Get 2.4.3

When I run pl-script with key “stats”, it returns result “0”.

When I comment string

die "Can not remove file $sendFile!" unless(unlink ($sendFile)); 

and run pl-script again, I see file zabbixSenderHPP2000_%hostname%_ in folder /var/tmp and this file contains data.

When I try manually use zabbix-sender like this:

zabbix_sender -c /etc/zabbix/zabbix_agentd.conf -i /var/tmp/zabbixSenderHPP2000_%hostname%_ 

I have these results:

info from server: "processed: 0; failed: 55; total: 55; seconds spent: 0.000757"
sent: 55; skipped: 0; total: 55 

It log files in zabbix server I have no errors about P2000 items.

Whan can it be?

Thank you!

P.S. Sorry for my bad English.

[adambmedent]

I am running into the same issue.

[root@zabbix externalscripts]# ./hp-msa.pl cloudst1a stats
0

I can also comment out.

die "Can not remove file $sendFile!" unless(unlink ($sendFile));

Then I end up with a /var/tmp/zabbixSenderHPP2000_cloudst1a_3719 which has good info. Just not sure what the issue is.

[crispytown]

I had the same issues with the files being created but not importing the data.

info from server: "processed: 0; failed: 55; total: 55; seconds spent: 0.000757" 
sent: 55; skipped: 0; total: 55

What I did to get it to work was make sure in the configuration for the host... that the "Host Name" is the spelled the same as the Agent interfaces "DNS Name". See Capture.jpg

[adambmedent]

I had the same issues with the files being created but not importing the data.

info from server: "processed: 0; failed: 55; total: 55; seconds spent: 0.000757" 
sent: 55; skipped: 0; total: 55

What I did to get it to work was make sure in the configuration for the host... that the "Host Name" is the spelled the same as the Agent interfaces "DNS Name". See Capture.jpg

I gave it a shot but I still just get a value of 0.