Unsolved
This post is more than 5 years old
11 Posts
0
13176
February 6th, 2008 14:00
Hard drive issues with IT assistant and Server Administrator
So I'm new to Dell servers.. I've installed the OpenManage IT Assistant (8.1) on Server1, and the Server Administrator (5.3.0) on Server2. I've configured an alert action to alert me on EVERYTHING (filter I created was for all severities, all categories, all the available devices and groups.
Status polling is set to run every 2 minutes (Discovery and Inventories are once a week)
So I pull a power cord from Server2... Within 2 minutes the server in IT Assistant shows up with a red X next to it, Hardware logs show the power redundancy has been lost, and I get an e-mail alert. So far so good. I put the power cord back, server returns to green, I get an email alert telling me it's all good.
Next I pull a hard drive... Nothing happens... The server remains green in IT Assistant. No alert is sent. The hardware logs DO show that the drive was removed (Storage Drive 1: Drive Slot sensor for Storage, drive removed), but again no alerts, no red X by the server.
Now here's the really weird thing... I go into Server Administrator, and it shows everything as green!. I go into Storage, look at the physical disks, and it shows both of them as online with a green checkmark!
I check to make sure I pulled a drive from the right server, and I did...
So what I am doing wrong? Or is IT Assistant/Server Administrator incapable of detecting a missing hard drive?? That's a big deal...
Ron Rosenkoette
11 Posts
0
February 6th, 2008 17:00
Okay, now (after about three hours), Server2 is now yellow in IT assistant. Still no alert was sent out. I go into Server Administrator, and it shows the Virtual disk as degraded, but when I click on the link to show Physical drives, it still shows them BOTH as online and green...
Any help?
DELL-Jeff M
2 Intern
•
793 Posts
0
February 6th, 2008 22:00
Thanks for posting! SNMP communication is not 100% reliable, especially if you are running status polling out of your ITA system every 2 minutes. Anytime a SNMP (UDP protocol) packet collides with a normal TCP/IP packet, it will die and the message will not get through. That will only check to see if the server is pingable. Bring that back to the default near-slow and once per hour check. Also, I have seen issues with a filter that hits on everything, I suggest at least leaving out the Other and Unknown categories.
I recommend not pulling out a hard drive for testing purposes, that will sometimes cause array corruption and problems in the future.
OMNavigator
291 Posts
0
February 7th, 2008 11:00
Are you sure you installed the OpenManage Storage system on the server being monitored? To install it, when you install Server Administrator (OMSA), make sure you select the "Storage" component (select Custom install).
Since you are able to receive normal hardware related Alerts (like Power supply redundancy lost), but not storage related, I would check for some problem (or missing) in the storage component in the installation.
Also, just for reference, hardware logs are NOT sent to IT Assistant (ITA) management station as soon as the event occurs. They are instead queried and fetched by ITA, when the user requests it (by selecting the appropraite tab in the user interface). SNMP alerts, on the other hand, will be sent to ITA system whenever the corresponding event occurs.
Ron Rosenkoette
11 Posts
0
February 7th, 2008 11:00
Thanks for the responses...
How can I check to see if the Storage component is installed? This is on a VMware ESX server (using the Linux agents).
Ron Rosenkoette
11 Posts
0
February 7th, 2008 17:00
Bala Chandrasek
57 Posts
0
February 7th, 2008 17:00
OMNavigator
291 Posts
0
February 8th, 2008 11:00
Launch the web interface of OMSA in the browser. If you have storage component installed, you will see a node "Storage" in the tree on the left side.
Alternately, you could also execute the command-line interface "omreport -?" or "omreport storage -?" to find out if OMSS is installed on the system or not.