Home > Unassigned > IBM: Mode select for redundant controller page 2C received

IBM: Mode select for redundant controller page 2C received

Problem:
This event can appear on IBM storages when some host is trying to access unmapped disk. Here is an example of event:

Date/Time: 29.4.2011 10:53:38
Sequence number: 318286
Event type: 300D
Event category: Command
Priority: Informational
Description: Mode select for redundant controller page 2C received
Event specific codes: 0/0/0
Component type: Controller
Component location: Enclosure 85, Slot 1
Logged by: Controller in slot A

Raw data:
4d 45 4c 48 03 00 00 00 4e db 04 00 00 00 00 00
0d 30 38 00 92 7c ba 4d 00 00 00 00 62 00 00 00
29 00 00 00 03 00 00 00 22 00 00 00 22 00 00 00
08 00 00 00 55 00 00 00 01 00 00 00 01 00 00 00
0a 00 00 00 01 00 00 00 00 00 00 00 00 00 00 00
00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00
00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00
01 00 00 00 00 00 00 04 7c 00 00 00 20 00 1a 06
00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00
00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00
20 00 1a 86 00 02 00 00 05 00 81 00 00 00 00 00
00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00
00 00 00 00 20 00 1a 86 00 00 00 00 00 00 00 00
00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00
00 00 00 00 00 00 00 00 0a 00 1a 86 00 00 00 00
00 00 00 00 00 00 00 00

Solution:

One solution is to send data to IBM support and they will answer you with name of server or list of servers which are trying to access unmapped disk or disks. Let’s look how to find out the name of server by your self.

You need to download support data from your storage and store it somewhere on disk. Support data are store as ZIP file, so you can easy extract and open files contained in ZIP file.

One of the files is file called majorEventLog.txt which contains events from storage. When you look into it you can see all events which you saw in Storage Manager. And of course our “2C” events. Just note which controller had problem (line Logged by).

The biggest file is called stateCaptureData.txt. Let’s open it and look for string “Mode Select 2C”. Here is log more verbose:

08:33:10.217 tVkiLog vkiCE ffff CE_NOTE 04/29/11-08:33:10 (nonRwTask2): NOTE: Mode Select 2C – mode is Take Vols(1) (0), force is True and host Id is 118 (ssql08_a_e,

Just make note of “host Id” value. It’s 118. So let’s look in same file for string “tditnall” for correct controller (Controller A or Controller B). In the list below string “Executing tditnall(0,0,0,0,0,0,0,0,0,0) on controller” you see list of hosts sorted by Host Id (column tditn):

118 4 :ssql08_a_ 20000000c9926cbe 10000000c9926cbe : 0 19425 169

To get full info about name of the host use “Port Name” value (in our case 10000000c9926cbe) to search through file. You can find following info under section “Initiator PORT Objects”:

ssql08_a_e x805e Host FCP 10000000c9926cbe ssql08_lv persist 3 posc

So now we know it’s host name ssql08_lv defined on controller A. Let’s do a rescan on machine (or reboot) and events will disappear.

Word after:

Steps can differ based on firmware versions, but I’m sure this will help you to find right way to solve problem.

I don’t know why IBM is not just providing name of the server causing problem in events. It would be much more easy to solve problem quicker without bothering IBM with, now, easy tickets 🙂

Please, leave me comment if this article was helpful for you. Thank you.

  1. October 28th, 2011 at 08:19 | #1

    How much of an exciting piece of writing, continue creating companion

  2. Leo
    December 22nd, 2011 at 17:03 | #2

    Thanks a lot. I have that problem, log shows every 2-5min an event 300D and using your guide I see that there events are generated by almost all host. Also there is a permanent event of “Logical drive not on preferred path due to ADT/RDAC failover” (4011).

  3. Cristina
    March 20th, 2012 at 14:10 | #4

    Very helpfull, thank you, it even applies for Oracle arrays.

  4. Phil
    November 12th, 2012 at 21:22 | #5

    Very helpful. This applies to all SAS/iSCSI arrays OEM’ed from LSI – Dell (MD3000 & MD3200), HP (StorageWorks P2000), IBM (DS3500), Oracle (Sun Storage 2530-M2), and any number of other companies.

  5. November 12th, 2012 at 22:13 | #6

    Thank you 🙂

  1. No trackbacks yet.