Topic: Software raid messages on a hardware raid system

[edform]

I've got a Compaq ML370 fitted with a 5300 raid controller with a lot of memory and 5 15000rpm 36gig drives.

I'm using it to test SME 7B5 at the moment.

Occasionally I get two emails from the system like this...

This is an automatically generated mail message from mdadm running on bigdog
A DegradedArray event had been detected on md device /dev/md1.

Faithfully yours, etc.

And...

This is an automatically generated mail message from mdadm
running on bigdog

A DegradedArray event had been detected on md device /dev/md2.

Faithfully yours, etc.

I know that such messages are generated when software raid systems run into trouble, but can the system supervise hardware raid as well?

Ed Form

[CharlieBrady]

I've got a Compaq ML370 fitted with a 5300 raid controller with a lot of memory and 5 15000rpm 36gig drives.

I'm using it to test SME 7B5 at the moment.

Occasionally I get two emails from the system like this...

As you (and everyone else) has been asked multiple times, please raise all issues with 7.0beta in the bug tracker at http://smeserver.sourceforge.net/. Please search before posting, as somebody else may already have raised the issue.

[edform]

As you (and everyone else) has been asked multiple times, please raise all issues with 7.0beta in the bug tracker at http://smeserver.sourceforge.net/. Please search before posting, as somebody else may already have raised the issue.

My apologies - it's done.

[Reinhold]

...since I could not find the bug report on sourceforge...

It's not a bug it's a feature.
- Your RAID-ARRAY shows as a single Disk to SME.
- SME7 is set up to always run as Raid1 (2 disks).
- If SME7 finds only one disk at install, it will set the 2nd Disk to missing and the SME-software-RAID1 to degraded
- There comes your messages:
You have only one (SME7)-disk and the SME7-Software Raid1 is degraded!
... fully correct in that context.

Regards
Reinhold

P.S.: Ed: Unless you put some proprietary software from the Raid controller manufacturer into SME how could it possibly know about the special abilities of your controller?
You could always switch off the mdadm message ...

[edform]

It's not a bug it's a feature.
- Your RAID-ARRAY shows as a single Disk to SME.
- SME7 is set up to always run as Raid1 (2 disks).
- If SME7 finds only one disk at install, it will set the 2nd Disk to missing and the SME-software-RAID1 to degraded
- There comes your messages:
You have only one (SME7)-disk and the SME7-Software Raid1 is degraded!
... fully correct in that context.

The system reports degrade events on two disk entities, /dev/md1 and /dev/md2.

Ed Form

[Reinhold]

Yes - it should!

SME7 uses 2 md's made up of hdx1 and hdx2 ...
... both are degraded if one hd is missing
(but still only half of what it used to break before in 6.x )

for details:

Code: [Select]

mdadm --query --detail /dev/md[12]

Regards
Reinhold

[edform]

mdadm --query --detail /dev/md[12]

I ran this command and found that the two dev objects are different. One is the battery-backed static ram on the raid controller card and the other is the disk-pack of the raid itself...

# mdadm --query --detail /dev/md[12]
/dev/md1:
        Version : 00.90.01
  Creation Time : Mon Oct 31 10:13:30 2005
     Raid Level : raid1
     Array Size : 101888 (99.50 MiB 104.33 MB)
    Device Size : 101888 (99.50 MiB 104.33 MB)
   Raid Devices : 2
  Total Devices : 1
Preferred Minor : 1
    Persistence : Superblock is persistent

    Update Time : Wed Nov  2 02:01:05 2005
          State : clean, degraded
 Active Devices : 1
Working Devices : 1
 Failed Devices : 0
  Spare Devices : 0


    Number   Major   Minor   RaidDevice State
       0     104        1        0      active sync   /dev/cciss/c0d0p1
       1       0        0       -1      removed
           UUID : 76835fd9:297138ae:a8c5dd6a:6db4ac08
         Events : 0.291
/dev/md2:
        Version : 00.90.01
  Creation Time : Mon Oct 31 10:11:52 2005
     Raid Level : raid1
     Array Size : 107426304 (102.45 GiB 110.00 GB)
    Device Size : 107426304 (102.45 GiB 110.00 GB)
   Raid Devices : 2
  Total Devices : 1
Preferred Minor : 2
    Persistence : Superblock is persistent

    Update Time : Wed Nov  2 13:05:33 2005
          State : clean, degraded
 Active Devices : 1
Working Devices : 1
 Failed Devices : 0
  Spare Devices : 0


    Number   Major   Minor   RaidDevice State
       0     104        2        0      active sync   /dev/cciss/c0d0p2
       1       0        0       -1      removed
           UUID : 6ac62cb0:37104dee:c9cd3f28:7d9dff10
         Events : 0.37537

[Reinhold]

Ed,

I ran this command and found that the two dev objects are different.
One is the battery-backed static ram on the raid controller card and the other is the disk-pack of the raid itself...
...
/dev/md1: /dev/cciss/c0d0p1  104.33 MB
/dev/md2:  /dev/cciss/c0d0p2  110.00 GB

???
- Looks (almost) perfectly "2 partitions" to me...
- What gives you the impression that you have
md1 e.q. "104.33 MB battery-backed static ram" as md-device
(the 110Gb surely ain't RAM!) ?

Regards
Reinhold

P.S.: Note: Software Raid is PARTITION BASED not disk based ! So usually you will group partitions from different disks!

[migelraton]

I get the same message and decided perhaps I should add a second hd.  Now, i'm relativle new at this so could someone tell me how to tell sme to use them both so I can stop getting the messages.