Hello everyone...
Here's the goal of the P.O.C....
Monitoring the status of the local Proliant DL360/G6 hard disk, while this system runs VmWare ESXi5.
It's important because the Proliant are booting on the local disk(s), that are either RAID 0 or RAID 5, and monitoring the status of the LOCAL disks is important.
The status of the Virtual Machines is not a problem, as they're already monitored under Zabbix.
The proliant have been patched with the latest version of HP Service Pack for Proliant (SPP), and now have iLO2 (Firmware 2.07, 08/17/2011)
SSH is activated on iLO, and on the ESXi5 kernel
ESXi Shell is activated on the ESXi host
ESXi has been downloaded directly from HP, software version is HP-ESXi-5.0.0-20110904001 (STANDARD)
Now, I would like to be able to monitor
-As much as possible about the above ESXi host.
-The local hard disks on the ESXi host (Proliant DL360/G6)
Here's what I found.
--------------------------------------------
I can "speak" with the Proliant iLO2 using
ssh admin@IP_of_iLO show -all /system1/drives
<enter password>
Then I can see "more or less" the status of the drives in the bay
If drive status=Ok , then it means all is OK
If drive status=Spun Down , then it means that one of the HD of the RAID5 has been removed, but I can't see which drive has been removed,at the same time the ESXi is still working.
If drive status=Rebuilding, then it means that one drive has been replaced / re-inserted, and that the raid5 is...rebuilding
But, through the iLO, I couldn't find anything more, like disk size, %free (on each HD and on the RAID), and the status of the HD
--------------------------------------------
I can "speak" with the ESXi using
ssh login@IP_of_ESXi
Then I can use df / du, but these commands do NOT show the status of a hard disk, and don't even show if a disk is removed !!
--------------------------------------------
Of course, using the local copy of the vSphere Client (directly dnloaded from the ESXi host), on a WinXP desktop, I can see
-The datastore (RAID5), total size, and free size.
Then through the "configuration" tab, I can see
-The status of the HP Smart Array P410i Controller
-The status of the lofical volume (RAID5)
-The status of each disk, and events on them.
--------------------------------------------
I enabled the SNMP according to the following method
# SNMP configuration
sed -i -e 's/false/true/g' /etc/vmware/snmp.xml
sed -i -e 's/<communities></<communities>MY_RW_Community</g' /etc/vmware/snmp.xml
sed -i -e 's/<targets></<targets>192.168.1.250@161 MY_RW_Community</g' /etc/vmware/snmp.xml
i.e. found the configuration file in /etc/vmware/snmp.xml (made a backup copy, just in case), changed false --> true
So, now I have something looking like
<config>
<snmpSettings>
<communities>public</communities>
<enable>true</enable>
<targets>
IP_of_ZABBIX_Server@161 public
</targets>
</snmpSettings>
</config>
And then restarted the ESXi
But I cannot access it via SNMP :-(
--------------------------------------------
--------------------------------------------
So, here are the questions.
- How can I monitor the iLO2 and the ESXi with Zabbix (1.8.5)
- How can I turn SNMP agent ON on the iLO and on the ESXi, and I don't want to receive traps, I want to be able to poll SNMP on these system.
Any ideas welcome
Here's the goal of the P.O.C....
Monitoring the status of the local Proliant DL360/G6 hard disk, while this system runs VmWare ESXi5.
It's important because the Proliant are booting on the local disk(s), that are either RAID 0 or RAID 5, and monitoring the status of the LOCAL disks is important.
The status of the Virtual Machines is not a problem, as they're already monitored under Zabbix.
The proliant have been patched with the latest version of HP Service Pack for Proliant (SPP), and now have iLO2 (Firmware 2.07, 08/17/2011)
SSH is activated on iLO, and on the ESXi5 kernel
ESXi Shell is activated on the ESXi host
ESXi has been downloaded directly from HP, software version is HP-ESXi-5.0.0-20110904001 (STANDARD)
Now, I would like to be able to monitor
-As much as possible about the above ESXi host.
-The local hard disks on the ESXi host (Proliant DL360/G6)
Here's what I found.
--------------------------------------------
I can "speak" with the Proliant iLO2 using
ssh admin@IP_of_iLO show -all /system1/drives
<enter password>
Then I can see "more or less" the status of the drives in the bay
If drive status=Ok , then it means all is OK
If drive status=Spun Down , then it means that one of the HD of the RAID5 has been removed, but I can't see which drive has been removed,at the same time the ESXi is still working.
If drive status=Rebuilding, then it means that one drive has been replaced / re-inserted, and that the raid5 is...rebuilding
But, through the iLO, I couldn't find anything more, like disk size, %free (on each HD and on the RAID), and the status of the HD
--------------------------------------------
I can "speak" with the ESXi using
ssh login@IP_of_ESXi
Then I can use df / du, but these commands do NOT show the status of a hard disk, and don't even show if a disk is removed !!
--------------------------------------------
Of course, using the local copy of the vSphere Client (directly dnloaded from the ESXi host), on a WinXP desktop, I can see
-The datastore (RAID5), total size, and free size.
Then through the "configuration" tab, I can see
-The status of the HP Smart Array P410i Controller
-The status of the lofical volume (RAID5)
-The status of each disk, and events on them.
--------------------------------------------
I enabled the SNMP according to the following method
# SNMP configuration
sed -i -e 's/false/true/g' /etc/vmware/snmp.xml
sed -i -e 's/<communities></<communities>MY_RW_Community</g' /etc/vmware/snmp.xml
sed -i -e 's/<targets></<targets>192.168.1.250@161 MY_RW_Community</g' /etc/vmware/snmp.xml
i.e. found the configuration file in /etc/vmware/snmp.xml (made a backup copy, just in case), changed false --> true
So, now I have something looking like
<config>
<snmpSettings>
<communities>public</communities>
<enable>true</enable>
<targets>
IP_of_ZABBIX_Server@161 public
</targets>
</snmpSettings>
</config>
And then restarted the ESXi
But I cannot access it via SNMP :-(
--------------------------------------------
--------------------------------------------
So, here are the questions.
- How can I monitor the iLO2 and the ESXi with Zabbix (1.8.5)
- How can I turn SNMP agent ON on the iLO and on the ESXi, and I don't want to receive traps, I want to be able to poll SNMP on these system.
Any ideas welcome
Comment