Cannot create raid group on newly added DAE

[khowatt]

Hello,

I have a FC5600 DPE (a Dell 650F) that is currently configured with a single raid group and 3 luns. I have added a DAE today with 10 drives (4x 18GB and 6x 36GB). I can see the drives, however I cannot bind any of them into a raid group. I am trying to create two raid5 groups; one consisting of the 18GB drives and the second with the 36GB drives, however I keep getting the error: "HOST_DUAL_BOARD_CREATE_RG_ERROR".

I do not have any software to manage these units, and have been getting by with the FCLI until now. One last note: I am only using SP-A currently - the switch connects to SP-A, and the EXP port of LCC-A is connected to the PRI port of LCC-A on the DAE.

I have attached some of the FCLI output below, followed by the error. Any help would be appreciated. Thanks!


fcli> ls
Logical Unit Summary:
Raid Dflt. Unit
LUN Group Owner Type Capacity Cache State Frus
--- ----- ----- ------ -------- ----- ------- ------------------
0 0 SP-A RAID-5 40.0 GB --- ENA 0 1 2 3 4 5 6 7 8
1 0 SP-A RAID-5 40.0 GB --- ENA 0 1 2 3 4 5 6 7 8
2 0 SP-A RAID-5 52.3 GB --- ENA 0 1 2 3 4 5 6 7 8
3 3 SP-A HotSpare 16.5 GB --- RDY 9

fcli> di -l

Fru Vendor Model Rev. Serial no. Capacity
==== ======== ================ ==== ============ =========
0. SEAGATE ST318304 CLAR18 DE84 3EL098JP 0x2141301
1. SEAGATE ST318304 CLAR18 DE84 3EL098YT 0x2141301
2. SEAGATE ST318304 CLAR18 DE84 3EL098JQ 0x2141301
3. SEAGATE ST318203 CLAR18 DE44 LRJ60226 0x2141301
4. SEAGATE ST318203 CLAR18 DE44 LRJ30137 0x2141301
5. SEAGATE ST318203 CLAR18 DE44 LRJ56232 0x2141301
6. SEAGATE ST318304 CLAR18 DE84 3EL098N1 0x2141301
7. SEAGATE ST318203 CLAR18 DE44 LRJ58711 0x2141301
8. SEAGATE ST318304 CLAR18 DE84 3EL098MN 0x2141301
9. SEAGATE ST318304 CLAR18 DE84 3EL098EB 0x2141301
10. SEAGATE ST318203 CLAR18 DE44 LRJ53425 0x2141301
11. SEAGATE ST318203 CLAR18 DE44 LRK78070 0x2141301
12. SEAGATE ST318203 CLAR18 DE44 LRK76331 0x2141301
13. SEAGATE ST318203 CLAR18 DE44 LRK76230 0x2141301
14. SEAGATE ST336605 CLAR36 DE18 3FP0NCNL 0x420C278
15. SEAGATE ST336607 CLAR36 DE06 3JA678KL 0x420C278
16. SEAGATE ST336607 CLAR36 DE06 3JA66YNZ 0x420C278
17. SEAGATE ST336607 CLAR36 DE06 3JA1Z6N6 0x420C278
18. SEAGATE ST336607 CLAR36 DE06 3JA1YY65 0x420C278
19. SEAGATE ST336704 CLAR36 DE84 3CD011KM 0x420C278


fcli> spstat

SP A LOOP ID 0x0 (0.)
PROM Revision: 3.23 Microcode Revision: 05.11.09
Statistics Logging: DISABLED PEER SP: PRESENT
Disk Write Caching: DISABLED R3 Write Buffering: DISABLED
WRITE CACHE: DISABLED READ CACHE: ENABLED
RAID OPTIMIZED: Mixed LUNs SP TYPE: DENALI
A: DP 00% TOTAL 1541 DIRTY 0000
B: TOTAL 1541
U: DP 00% TOTAL 0000
Requests Complete: 17
SPS A: --
SPS B: --

Press any key to continue....

slot : 0 1 2 3 4 5 6 7 8 9 | PSA PSB FAN
DPE-state: ENA ENA ENA ENA ENA ENA ENA ENA ENA RDY | FLT OK OK OK
Unit/Group : G00 G00 G00 G00 G00 G00 G00 G00 G00 H03 |
DAE1-state: UNB UNB UNB UNB UNB UNB UNB UNB UNB UNB | FLT OK OK
Unit/Group : ** ** ** ** ** ** ** ** ** ** |

06/09/2005 20:09:32

fcli> setgroup 2 -crg 1_0 1_1 1_2 1_3

Configuring RAID Group 2
RB error full 0x50220101
part category 0x5 code 0x5022 error num 0x22 status 0x1 type 0x1
Raid Group configuration ERROR: Category 0x5 Num 0x22 Type 0x1

Sunburst Code: HOST_DUAL_BOARD_CREATE_RG_ERROR
Status recv'd opcode 0x8D status 0x50220101

 

[maultier]

One of the problems is the 5.11.09 code you're running. 36gb is supported in 5.11.52 .. 72gb in 5.11.59. Perhaps upgrading the code will take care of the issue.

 

[maultier]

My mistake... I have mostly 5603 and that is the rev I gave you. For a 5600, the 5.11.09 is ok for 36gb. I'm not sure what is causing the bind error. Have you tried just binding the disks as individual units to see if that will work?

 

[maultier]

How about just binding the new raid 5 luns on the existing group (0) ??

 

[maultier]

Another thing that may be happening is that you have some residual raid group information out there. You should see info on rg's 0 and 3. Run the sg <rg#> -l command on Raid groups 1, 2 and any other numbers you are trying to assign and make sure you don't get any information back. If you do get info back on the groups you are trying to create, use to sg -crg command to remove those fragments. Then see if you can bind/build your raid group.

 

[khowatt]

Problem Solved!!

Thanks maultier for your posts. I tried removing any residual group info, and even went so far as to unbind everything. Still had no success in binding anything on the DAE.

One thing was haunting me: the HOST_DUAL_BOARD error. I was wondering if not having the LCC-B's (only the SP-A/LCC-A side was cabled) was causing the problems. As I don't have an extra FC DB9 cable around, I decided to pull SP-B from the DPE chassis and see what would happen.

After resetting the DPE, I was able to create raid groups and bind LUNs to my heart's content. I set up a bunch of LUNs, replaced SP-B, and was still able to access and create filesystems on the LUNs on the DAE, even after resetting the DPE. However, with SP-B back in place, I cannot bind any new LUNs.

Next week I'll get a cable and connect the LCC-B's and see if it improves the situation. No matter if it doesn't, but now I'm curious

 

[maultier]

Thank you for the the follow up post. I can only speculate that the broken path was the problem. Since the Clariion treats this as a fault condition, I'm guessing spb was brought into the mix somehow when the dae was added because now the path has to include the dae. Clariion runs data and command flow to the drives - odd first 9 to 1 on the dpe and then 9-1 on the dae. Then it goes back for the even drives to complete the loop.
Congratulations on getting the problem solved!