Solved

Trying to monitor a Dell PERC 5/i controller on a white box VMware ESXi 4.1 server

Posted on 2010-11-11
11
4,683 Views
Last Modified: 2012-06-21
Hi Experts,

I have spent lots of time Googling and researching previous attempts at solving this problem, all to no avail.

Let me begin by detailing my hardware setup:

White Box Server:
ASUS Z8NA-D6C Server Motherboard (latest BIOS)
2 x Intel Xeon E5506 Processors
24GB RAM
3 x Dell PERC 5/i controllers using the latest firmware from Dell's website (Hard Drives are only connected to controller #3, however)
Physical Hard Drives: 8 x Western Digital 600GB Velociraptors
Virtual Disk Config:
-----> 2 in RAID 1 (mirrored),
-----> 4 in RAID 5,
-----> 1 Hot Spare,
-----> 1 Connected directly to motherboard's IDE controller.

Versions of ESXi I tried:
1.  "Generic" ESXi 4.1 build 260247 from VMWare's website
2.  Dell-Customized ISO of ESXi 4.1 build 260247 from Dell's website

The Dell website states clearly that the customized ISO does NOT contain the OpenManage Server Administrator, and as recommended in various other threads, I tried to install the Dell OpenManage Offline Bundle and VIB for ESXi.

Following the instructions, I downloaded and installed the VMWare VSphere CLI application (on another workstation on the network), copied the zip (downloaded above) to it's local directory (subfolder "bin") and submitted the following command and got the output below:

vihostupdate.pl --server 192.168.0.236 -i -b bin\OM-SrvAdmin-Dell-Web-6.3.0-2075.VIB-ESX41i_A00.8.zip --username root --password ""
Please wait patch installation is in progress ...
No matching bulletin or VIB was found in the metadata.No bulletins for this platform could be found. Nothing to do.

Now, there are steps to follow after installing the OpenManage Server components (which I would appreciate assistance with), but I need to get past this first.

And of course, if there is a completely different route to be able to monitor LSI-based RAID cards within ESXi 4.1, I am completely open to it.  Within Windows, I have thus far been using LSI MegaRAID Storage Manager to monitor and manage PERC cards and arrays on my other non-virtualized white box servers.

Thanks in advance!
0
Comment
Question by:waqqas31
  • 6
  • 3
  • 2
11 Comments
 
LVL 28

Expert Comment

by:bgoering
ID: 34112772
The vanilla CIM agents in ESXi should be able to provide information. Do you have vCenter Server? If so you can configure alerts there. If not there are opensource systems capable of CIM monitoring. You might take a look at Nagios (http://www.nagios.org/)

Good Luck
0
 

Author Comment

by:waqqas31
ID: 34112819
Hi bgoering,

No, I do not have vCenter Server.  If I'm not mistaken, ESXi 4.1 cannot be monitored using vCenter.  That and it's an investment I can't afford to make at this juncture.

I am trying to read up on Nagios.  If you can point me to a quickstart guide to using Nagios, I would appreciate it.  All the information I've found so far seems quite convoluted.

Thanks!
0
 
LVL 28

Assisted Solution

by:bgoering
bgoering earned 150 total points
ID: 34113327
Take a look at http://nagios.sourceforge.net/docs/3_0/quickstart.html and http://nagios.sourceforge.net/docs/nagios-3.pdf. Its actually been a while since I played with Nagios.

ESX/ESXi provides hardware health infromation via CIM providers. You can either use the vanilla ones that come with ESX, or in some cases vendors like Dell, HP, IBM, etc. provide CIM providers specific to their hardware. The CIM functionality in Nagios is provided via plugins you can get at Nagios Exchange. See http://exchange.nagios.org/index.php?option=com_mtree&task=search&Itemid=74&searchword=ESX for some ESX related plugins.
0
 
LVL 47

Assisted Solution

by:dlethe
dlethe earned 150 total points
ID: 34113415
You have velociraptors configured to the PERC/5?   Are you aware that this combination has known issues, is not qualified (at least by LSI and WD/Seagate), and puts your data at risk?

The velociraptor has incorrect RAID timing and is simply not acceptable behind that controller. You risk data loss. Did Dell sell you those disks to work with that controller?  If so, you are getting burned.


0
 

Author Comment

by:waqqas31
ID: 34114273
@bgoering:

Thanks for the links.  I am reading up and implementing as we speak.  It might take me a day to have some feedback, since I have to set up a Linux system first.  I was hoping for a Windows-based approach, but this is better than nothing.

@dlethe:

Thanks a bunch for the tips.  

I was not aware of the issues you mentioned.  Is it isolated to the PERC 5 or is it the same with a PERC 6 or other LSI-based controllers as well?  Luckily for me, this is my only whitebox with this combination, and I am only using this one to solve the issue at hand.  When I will prepare to use this server for production purposes, I will change the controller/hard drive combination to something not known to have such issues.  The Velociraptors, like all the other components, were bought from regular oem vendors, not Dell, etc.

@everyone:

In the meanwhile, if anyone knows of a Windows-based solution, please share!

Thanks and I will report back asap
0
How your wiki can always stay up-to-date

Quip doubles as a “living” wiki and a project management tool that evolves with your organization. As you finish projects in Quip, the work remains, easily accessible to all team members, new and old.
- Increase transparency
- Onboard new hires faster
- Access from mobile/offline

 
LVL 47

Expert Comment

by:dlethe
ID: 34114531
The problem manifests itself with the LSI-MPT chipset, unfortunately.   You're simply going to have to use those disks elsewhere.  
0
 

Accepted Solution

by:
waqqas31 earned 0 total points
ID: 34139179
Gents,

I have solved the problem!

After downloading the VIB bundle from dell.com mentioned above, you need to unpack the ZIP file and then unpack the metadata.zip inside of it.  You will then find a vmware.xml file.  Will find the following section in the file:

<systemReqs>
          <swPlatform locale="" version="4.1.*" productLineID="embeddedEsx"/>
          <swPlatform locale="" version="4.1.*" productLineID="esx"/>
          <hwPlatform model="" vendor="Dell"/>
          <hwPlatform model="" vendor="Dell Inc."/>
          <hwPlatform model="" vendor="Dell Computer Corporation"/>
          <maintenanceMode>false</maintenanceMode>
</systemReqs>

Delete the 3 lines that contain vendor="Dell... in it.

I also modified this section:

<platforms>
      <softwarePlatform locale="" version="4.1.*" productLineID="embeddedEsx"/>
</platforms>

to

<platforms>
      <softwarePlatform locale="" version="4.1.*" productLineID="esx"/>
      <softwarePlatform locale="" version="4.1.*" productLineID="embeddedEsx"/>
</platforms>

But I don't think it was necessary (and I didn't try it without modifying it first).

After deleting the 3 "Dell" lines, repackage metadata.zip then the original ZIP file.

I made sure the larger zip file retained the name of the original download from Dell and that the repackaged zip files did not introduce any new "path" or "folder" information.

I then proceeded to install the VIB file as per the instructions on Dell's website and it went through without a hitch!

I then set a password for the root account on the ESXi host.

I then downloaded, extracted and installed Dell OpenManage Server Administrator Managed Node 6.3.0 on a VM running Windows Server 2008 R2 Enterprise on the ESXi server itself (i.e. even this will work) and being a non-Dell machine, it only allowed me to install the web interface, which is the only thing we actually need.  I fired up the application and I was greeted with a login screen where I entered the IP of the ESXi host, username (root) with the password and chose to ignore certificate warnings.  I was then presented with a wealth of information, including the ability to view and edit all storage adapters and their advanced settings.

I am now a happy camper :)

Dell OMSA is also available for Linux as well, so this is almost a platform independent solution.
0
 

Author Comment

by:waqqas31
ID: 34139340
@dlethe:

Is there a particular source or website you consult to stay up-to-date on issues such as the one you mentioned about the LSI-MPT chipset and Velociraptor drives?  I Googled to little avail.

Also, out of curiosity, how recently did you hear about this particular issue.  I did find one piece of information which seemed to be addressing the very problem you were describing:  link to forum

Is it possible that the problem has now been resolved?

Thanks again!

0
 

Author Comment

by:waqqas31
ID: 34176034
For what it's worth, here's an alternate but extremely basic solution to only "monitor" PERC arrays (which also works for HP Smart Array controllers like P400, P410i, P212, etc, btw).

1. Download the HP Customized VMware ESXi 4.1 CD.
2. Download the HP Offline Bundle for ESXi 4.1.
3. Download the VMware VSphere Client 4.1.
4. Download the VSphere CLI 4.1 application.
5. Install ESXi.  Set the root password (to other than blank).  Perform Step 9 below.  If everything displays (like in the picture), you're done, if not, continue from Step 6.
6. Modify the HP Offline Bundle similarly to the Dell Offline Bundle procedure above: unzip twice until you can edit the vmware.xml file.  Remove any lines in the <systemReqs> section specifying that the vendor is HP (3 lines).  Save and put the bundle file back together.
7. Install the CLI app and use the vihostupdate.pl command (script) to install the HP Offline Bundle (mimick the command above).
8. Restart the ESXi host.
9. Install the VSphere Client and log into the ESXi host.  Select the Configuration Tab.  Select Health Status in the left hand column then expand the tree on the right hand side.  Expand the Storage submenu.  You can now see entries for PERC cards as well as Smart Array cards and every disk connected to them.  You cannot do any sort of "operations" here, but you can still see how many drives are connected, how many hot spares are configured, whether an array is optimal, degraded or offline, whether a drive is rebuilding or not, etc.  In short, you can still determine if you need to act without having to shut down ESXi just to reboot into the RAID BIOS or ACU.  If you were to replace a dead drive with a fresh one and the controller automatically would start rebuilding, this would also be reflected.
 2 PERC cards and 1 HP Smart Array P410i
0
 
LVL 47

Expert Comment

by:dlethe
ID: 34176406
@waqqas31
I have an NDA with LSI and do a lot of storage/RAID development work for them, as well as numerous LSI OEMs, and drive manufacturers, so that is why I say things that you won't find on a website.   But stuff I have mentioned is typically buried in release notes of controllers (or source code of LINUX drivers).

Since the PERC is LSI kit with LSI firmware (and a few changes I won't reveal due to NDA), then one can go to LSI site and probably find confirmation of all of this by looking at the HCL lists.  Or look at HCL lists of Intel, IBM, and HP, who are some of the other vendors that rebrand the same kit.  

Reasons why things are as they are just aren't discussed outside of insiders, because many times if you take a HP architect, for example, they simply don't DVT non-qualified disks because HP won't authorize their use in the first place, so root causes are not well understood.  I'm in a rather unique position as my code caters to OEMs and non-OEMs, and appliance manufacturers, and they are the ones that pay for such testing and want to know all compatibility issues.
0
 

Author Closing Comment

by:waqqas31
ID: 34179050
Guys,

I've tried to award points to both of you.  I hope it goes through ok!

Thanks!
0

Featured Post

Threat Intelligence Starter Resources

Integrating threat intelligence can be challenging, and not all companies are ready. These resources can help you build awareness and prepare for defense.

Join & Write a Comment

Suggested Solutions

Data center, now-a-days, is referred as the home of all the advanced technologies. In-fact, most of the businesses are now establishing their entire organizational structure around the IT capabilities.
David Varnum recently wrote up his impressions of PRTG, based on a presentation by my colleague Christian at Tech Field Day at VMworld in Barcelona. Thanks David, for your detailed and honest evaluation!
Teach the user how to install log collectors and how to configure ESXi 5.5 for remote logging Open console session and mount vCenter Server installer: Install vSphere Core Dump Collector: Install vSphere Syslog Collector: Open vSphere Client: Config…
This tutorial will walk an individual through the process of installing the necessary services and then configuring a Windows Server 2012 system as an iSCSI target. To install the necessary roles, go to Server Manager, and select Add Roles and Featu…

744 members asked questions and received personalized solutions in the past 7 days.

Join the community of 500,000 technology professionals and ask your questions.

Join & Ask a Question

Need Help in Real-Time?

Connect with top rated Experts

13 Experts available now in Live!

Get 1:1 Help Now