warning message " Patrol Read aborted on PD 07(e0xfc/s3) " in UCS

[fibinfaizal]

My Applications (CUCM, CUCXN and UCCX) are running on UCS. I am not a data center guy and not good with UCS. 

When I went through storage logs in CIMC, I found a warning message " Patrol Read aborted on PD 07(e0xfc/s3) ".

What does this mean and what actions needs to be taken?

Thanks

[Qiese Dides · Accepted solution]

Hi Patrol Read Aborted could mean a number of things. However, it relates to the raid controller having a issue.

What is your UCS Blade Model?

You are most likely hitting one of these bugs;

https://tools.cisco.com/bugsearch/bug/CSCul22968/?reffering_site=dumpcr

https://tools.cisco.com/bugsearch/bug/CSCuj10535/?reffering_site=dumpcr

I would recommend to look at these bugs and see if the fit your situation. If possible open a case with TAC so they can determine if it is one of these bugs or a faulty raid controller.

HTH,

Qiese Dides

[fibinfaizal]

product id: UCS C220 M4S

[Qiese Dides]

Hi Fibin,

Can you make sure your ENIC / FNIC drivers are up to date. If you follow the guide below you can see the Cisco recommended value also.

http://www.cisco.com/web/techdoc/ucs/interoperability/matrix/matrix.html

How to obtain ENIC and FNIC driver versions ( ~# vmkload_mod -s [enic|fnic])

- SSH into the ESXi CLI (via putty or the KVM console (type alt-F1 in the KVM console)*

- enter the command vmkload_mod –s enic | grep Version for the enic driver version

- enter the command vmkload_mod –s fnic | grep Version for the fnic driver version

*Note:

- Make sure SSH is enabled: Login to ESXi > Troubleshooting Options > Enable SSH

- To escape the ESXi CLI within the KVM console type alt-F2, then regular F2 to enable the console

Regards,

Qiese Dides

[fibinfaizal]

I have two UCS servers. Primary and secondary.. Though both are of same model and having same versions, this warning appears to be only for the primary one. 

I've opened a TAC case and sent them show tech support files and screenshot of the warning from cimc.

Will update all once i get an update from them.. 

Thanks a ton for your advice.

[Qiese Dides]

No problem at all. Please keep us updated I'm interested in the solution as well :).

Happy Holidays.

[fibinfaizal]

Here are the findings from TAC.

We are hitting the below bug:

https://tools.cisco.com/bugsearch/bug/CSCul22968/?reffering_site=dumpcr

You were correct. :) 

Here is the Workaround:
Manually kick off a patrol read with MegaCLI to alter the schedule to not coincide with the consistency check
MegaCli -AdpPR -Start -aALL

or

Manually adjust consistency check schedule via MegaCLI or WebBIOS:

** MegaCLI -AdpCcSched -Dsbl|-Info|
{-ModeConc | -ModeSeq [-ExcludeLD
-LN|-L0,1,2] [-SetStartTime yyyymmdd
hh ] [-SetDelay val ] }
-aN|-a0,1,2|-aALL

There are two tests that the RAID controller runs, patrol read and consistency check ,if they are scheduled at the same time, the system can display errors like on this case. 

A Patrol Read (PR) periodically verifies all sectors of physical disks that are connected to a controller, for all hot spare drives.

Consistency check  (CC)  checks RAID volume integrity.

It won’t effect on the production but the server will keeps giving you such these errors .

[Qiese Dides]

Glad to hear we got the problem resolved!

Happy Holidays,

Qiese Dides