As a business, we run Dell servers. For years we used Supermicro by ordering the chassis from one place, usually Boston (the firm not the city) and the components elsewhere. Often this would work well but SM would often change the memory requirements between a version of the same chassis and you’d find this out by getting to the datacentre with a nasty surprise. Also, the builds of several machines took time as well as all the waste from packaging. We switched to Dell to remove any surprises like that and cut down on waste.

With that, we could use Dell’s OMSA (OpenManage) to monitor all aspects of servers and allow us to remotely monitor. This was far better than anything SM had at the time, although that may have changed since. We’re talking about at least 10 years ago.

If a drive dies, we know within a few minutes via SNMP monitoring and alerts setup in PRTG. The same thing if the RAID array just magically becomes degraded. Until recently, we’ve had many flawless years of this.

That is until Dell pushed through an upgrade from OMSA 8 to OMSA 9 overnight. This resulted in any server that had an older SAS 6iR card entering a failure mode for all storage related queries. On checking with omreport, the storage card appeared completely missing.

[root@xxxx ~]# omreport storage controller
No controllers found

Many hours spent removing and installing OMSA without any luck. I found various posts about kernel patches but all were irrelevant to the running version.

After spending some time debugging the issue, I managed to pinpoint it to the storage card so I had a better point to work out what to do. I found the old repo of OMSA on the Dell site at Although some of the version numbers change from 7.4.0 to 14.04.0, I decided 15.07.00 was a safe bet.

Firstly I removed any trace of OMSA

yum remove srvadmin-* smbios* libsmbios -y
rm -f -R /opt/dell/
rm -f /etc/yum.repos.d/dell-system-update.repo

Then I added the older repo to yum

wget -q -O – | bash

and then installed normally.

yum install srvadmin-all

Standard process then applies after then to run OMSA and active with SNMP and start on boot so I won’t provide those steps.

There were a few issues with CentOS 7 on the odd server with various smbios and libsmbios errors with yum refusing to allow it due to version mismatches. I manually installed with rpm -ivh and then the respective URL from their RPM list. The below is one whole text block.

rpm -ivh

Then the standard yum install.

Categories: Server Stuff


I work in technology and communications, often for too many hours of the day. Can often be found in datacentres all over looking at the flashing lights of storage arrays.

1 Comment

Richard · 27th June 2020 at 10:23 pm

This is very helpful, unfortunately it doesn’t seem to work on Centos 8.

Leave a Reply

Avatar placeholder

Your email address will not be published. Required fields are marked *