UCS C200 M2 HDD LED Issue

[Bell Kun]

Dear all,

My company has purchased a new server for replacing the old one. The new one is model UCS C200 M2.

This server is containing a raid controller LSI 1064E for RAID 0/1 setting.

I did configure the RAID 1 and installed the Windows Server OS already.

But I found that the Harddisk in Bay 1 is light-on with blinking amber at harddisk fault LED and blinking green at harddisk activity LED at the same time (See attached photo).

Is this condition normal? and what it would like to indicate from this condition? Please advice.

If there is fault happened on the harddisk, I need to call Cisco for h/w replacement ASAP.

[mipetrin]

Hi Bell,

According to the C200 Front Panel LED status descriptions, Amber would indicate that the drive has failed.

Best would be to open a TAC case, upload the show tech from your C200, which they can verify the status and provide you with a replacement (if necessary).

Thanks,

Michael

[yeopaul]

Hi Michael,

I just want to check, I am using UCS-C210M2-VCD2, I am trying to replace the harddrives after it failed. I swapped the harddisk but it shows "unconfigured good", how do I configure the drives? the CIMC is running version 1.43.

regards

paul

[Keny Perez]

Paul,

In 1.4(3x) you need to reboot the server and according to your controller model's form factor, you will either need to press "Ctrl+c/m/h" you can see that during post when the system lists the controller and the disks present.

After that (if your controller is a PCIe RAID controller) you will see something similar to the image below after you presss "Ctrl+H" during post:

You may go to "Physical View" (on the left), click on the disk marked as "unconfigured good" (like the blue ones in the screenshot) and then click on "Replace PD missing" and then "Make online".

If the controller is not a PCIe card, please let me know what controller model you use and I will tell you how to recover the disk.

-Rate ALL helpful anwers.

-Kenny

[yeopaul]

Hi Kenny,

Thanks a lot for the reply, this is helpful. If i refer to the storage, it shows MegaRAID® SAS 9261-8i, I did a search on the web and I have this:

• Connect up to 128 SATA or SAS drives with eight internal 6 Gb/s SATA and SAS ports
  
• Fit into rack-mounted servers with low-profile form factor and side-mount connectors
  
• Bandwidth to support critical applications with PCIe® 2.0 connectivity
  
• Balance protection and performance for critical applications with RAID levels 0, 1, 5, 6, 10, 50, and 60
  

1) I have a total of 3 harddisks failed, and the Raid level shows 6, is it ok if i replace 2 at one time, or I should do it 1 by 1?

2) Also, is there a way for me to tell how long the rebuild will take, or any progress indicator?

regards,

paul

[Keny Perez]

Paul,

In the first post I see that you menitoned "I swapped the harddisk but it shows "unconfigured good".  It looks like you did the change already... Anyways, with RAID 6, the array supports the failure of up to 2 HDDs, having 3 might be the reason why it shows up as "off line" in CIMC.

1) IMHO, if you have 2 bad ("failed" state) disks in an array config for RAID 6 and the disk can't be recovered, there is no much of difference if you replace one by one or both at the same time, cause the amount allowed to fail at the same time is 2. In the other hand, if one is "failed" and the other has a Predictive Failure or something not that critical, I would change the one with the "Failed" state first and then the other.  In your case, if you have 3 bad disks and all of them are part of the same array (RAID 6), data is either gone or corrupted cause the double parity set for RAID 6 is not enough to maintain the integrity of the data, so 3 disk failed in your array might mean "data loss"

2)There is no way to tell how much time the rebuild will take but it may take between 24 and 72hrs depending on the amount of data to be synched up and the % of resources used by the server to complete the task.  If I am not wrong, you should be able to go (in the WebBIOS) to "Controller Properties" in the left pane and see something that says "Rebuild Rate", which is set to 30% by default but can be increase/decreased.  

Caveat:  If you increase the "Rebuild Rate" the server will use more resources to accomplish the task of synchronizing the disks, affecting server's performance.

From WebBIOS though, you can see if the rebuild rate is started and its percentage completed so far, however ( I have not tested) I understand that when the server is rebooted the rebuild restarts, too bad if that is the case cause that means either you leave the server down until the rebuild is complete so you can monitor from WebBIOS or you reboot the server, bring it up and wait until the disks LEDs are all green and the array stops showing as degraded (expected during the rebuild).

Let me know if there is any other question I can help with.

-Rate ALL helpful anwers.

Kenny

[yeopaul]

Hi Kenny,

I will start doing it today, your post really helps me a lot!

regards,

paul

[Keny Perez]

Paul,

Here to help anytime

-Kenny

[ellisonk]

Any reason why I would not have the "Replace missing PD" radio button in my WebBIOS?

[michaelpalmieri1976]

Thanks for this great post.   I had this same issue on 7/30/14.   The server was beeping, I had no way to silence the Alarm for failed hard drive. Its a production server and reboot was not an option as the server was in a degraded state.  I opened a case with TAC, had replacement sent next day.  I put in the replacement disk drive and it took about 8 hours to rebuild in a RAID 1 array.    

What I do not like about the CIMC is that you cannot manage the RAID controller, CIMC only reports what it sees off the RAID controller......However, I was able to download the Management utilities for my Raid controller  by doing a Google search on the card detected by CIMC.     Loaded up the Raid Management software  once downloaded from mfg web site.  I  was able to work with the RAID controller while the server was online through the operating system (Windows 2008R2) Vs having to reboot using the RAID utilities during the POST sequence.

Here are some helpful Articles;

http://www.lsi.com/products/raid-controllers/pages/megaraid-sas-9260-4i.aspx#tab/tab4

https://supportforums.cisco.com/discussion/11282696/ucs-c200-m2-hdd-led-issue

[Keny Perez]

Hello,

With the new servers that support CIMC 1.5 and on, the RAID controller can also be managed through CIMC without having to reboot

Check the answer from smcquerr  here : https://supportforums.cisco.com/discussion/11977411/ucs-c220-series-server  ; I showed there how to do it rebooting the server and he showed how to do it without the reboot

HTH,

Don't forget to rate helpful answers

-Kenny