raid superblock error using raidstart (level 0 software raid)

Hi,

I've read so may thing about what could cause this problem that I'm unsure
where to turn. Hope someone can shed some light, thanks for any response...

I have a 2 disk raid (and a 3 disk raid) that both get
"invalid raid superblock magic on mdn" when I start them or when I reboot.
see example below:

md: autorun ...
md: considering sde1 ...
md: adding sde1 ...
md: adding sdd1 ...
md: created md1
md: bind<sdd1,1>
md: bind<sde1,2>
md: running: <sde1><sdd1>
md: sde1's event counter: 00000004
md: sdd1's event counter: 00000004
md1: max total readahead window set to 496k
md1: 2 data-disks, max readahead per data-disk: 248k
raid0: looking at sdd1
raid0: comparing sdd1(245111616) with sdd1(245111616)
raid0: END
raid0: ==> UNIQUE
raid0: 1 zones
raid0: looking at sde1
raid0: comparing sde1(245111616) with sdd1(245111616)
raid0: EQUAL
raid0: FINAL 1 zones
raid0: zone 0
raid0: checking sdd1 ... contained as device 0
(245111616) is smallest!.
raid0: checking sde1 ... contained as device 1
raid0: zone->nb_dev: 2, size: 490223232
raid0: current zone offset: 245111616
raid0: done.
raid0 : md_size is 490223232 blocks.
raid0 : conf->smallest->size is 490223232 blocks.
raid0 : nb_zone is 1.
raid0 : Allocating 8 bytes for hash.
md: updating md1 RAID superblock on device
md: sde1 [events: 00000005]
(write) sde1's sb offset: 245111616
(write) sde1's sb offset: 245111616
md: sdd1 [events: 00000005]
(write) sdd1's sb offset: 245111616
[events: 00000000]
md: invalid raid superblock magic on md1 <<<<-----error
md: md1 has invalid sb, not importing!
md: no nested md device found
md: ... autorun DONE.


Periodically I get the following errors and the disk/raid crashes the system:


Apr 15 04:25:58 snood kernel: pdc-ultra:[warning] disk3 ATA timeoutat LBA
0x19000597
Apr 15 04:25:58 snood kernel: pdc-ultra:[warning] submit cam busy
Apr 15 04:26:51 snood kernel: scsi : aborting command due to timeout : pid
18925272, scsi1, channel 0, id 2, lun 0 Read (10) 00 19 00 05 97 00 00 08 00
Apr 15 04:26:51 snood kernel: pdc-ultra:[info] scsi abort success
Apr 15 04:26:58 snood kernel: pdc-ultra:[warning] disk3 ATA timeoutat LBA
0x19600597
Apr 15 04:26:58 snood kernel: pdc-ultra:[warning] submit cam busy
Apr 15 04:27:52 snood kernel: scsi : aborting command due to timeout : pid
18925365, scsi1, channel 0, id 2, lun 0 Read (10) 00 19 60 05 97 00 00 08 00
Apr 15 04:27:52 snood kernel: pdc-ultra:[info] scsi abort success
Apr 15 04:28:00 snood kernel: pdc-ultra:[warning] disk3 ATA timeoutat LBA
0x19d00597
Apr 15 04:28:00 snood kernel: pdc-ultra:[warning] submit cam busy
Apr 15 04:28:54 snood kernel: scsi : aborting command due to timeout : pid
18925486, scsi1, channel 0, id 2, lun 0 Read (10) 00 19 d0 05 97 00 00 08 00
Apr 15 04:28:54 snood kernel: pdc-ultra:[info] scsi abort success
Apr 15 04:28:59 snood kernel: pdc-ultra:[error] disk3 error at LBA 0x1a600597
cmd=0x25 status=0x51 error=0x40
Apr 15 04:29:02 snood last message repeated 2 times
Apr 15 04:29:08 snood kernel: pdc-ultra:[warning] disk3 ATA timeoutat LBA
0x1a600597
Apr 15 04:29:08 snood kernel: pdc-ultra:[warning] submit cam busy
Apr 15 04:29:58 snood kernel: scsi : aborting command due to timeout : pid
18925695, scsi1, channel 0, id 2, lun 0 Read (10) 00 1a 60 05 97 00 00 08 00
Apr 15 04:29:58 snood kernel: pdc-ultra:[info] scsi abort success
Apr 15 04:30:01 snood kernel: pdc-ultra:[error] disk3 error at LBA 0x1abc0597
cmd=0x25 status=0x51 error=0x40
Apr 15 04:30:02 snood kernel: pdc-ultra:[error] disk3 error at LBA 0x1abc0597
cmd=0x25 status=0x51 error=0x40
Apr 15 04:30:09 snood kernel: pdc-ultra:[warning] disk3 ATA timeoutat LBA
0x1abc0597
Apr 15 04:30:09 snood kernel: pdc-ultra:[warning] submit cam busy
Apr 15 04:31:00 snood kernel: scsi : aborting command due to timeout : pid
18925871, scsi1, channel 0, id 2, lun 0 Read (10) 00 1a bc 05 97 00 00 08 00
Apr 15 04:31:00 snood kernel: pdc-ultra:[info] scsi abort success
Apr 15 04:31:03 snood kernel: pdc-ultra:[error] disk3 error at LBA 0x1b400597
cmd=0x25 status=0x51 error=0x40
Apr 15 04:31:09 snood kernel: pdc-ultra:[error] disk3 error at LBA 0x1b4c0597
cmd=0x25 status=0x51 error=0x40
Apr 15 04:31:15 snood kernel: pdc-ultra:[warning] disk3 ATA timeoutat LBA
0x1b4c0597
Apr 15 04:32:06 snood kernel: scsi : aborting command due to timeout : pid
18926124, scsi1, channel 0, id 2, lun 0 Read (10) 00 1b 4c 05 97 00 00 08 00
Apr 15 04:32:06 snood kernel: pdc-ultra:[info] scsi abort success
Apr 15 04:32:10 snood kernel: pdc-ultra:[error] disk3 error at LBA 0x1c100597
cmd=0x25 status=0x51 error=0x40
Apr 15 04:32:13 snood last message repeated 2 times
Apr 15 04:32:19 snood kernel: pdc-ultra:[warning] disk3 ATA timeoutat LBA
0x1c100597
Apr 15 04:32:19 snood kernel: pdc-ultra:[warning] submit cam busy
Apr 15 04:33:09 snood kernel: scsi : aborting command due to timeout : pid
18926488, scsi1, channel 0, id 2, lun 0 Read (10) 00 1c 10 05 97 00 00 08 00
Apr 15 04:33:09 snood kernel: pdc-ultra:[info] scsi abort success
Apr 15 04:33:17 snood kernel: pdc-ultra:[warning] disk3 ATA timeoutat LBA
0x1cb00597
Apr 15 04:33:17 snood kernel: pdc-ultra:[warning] submit cam busy
Apr 15 04:34:11 snood kernel: scsi : aborting command due to timeout : pid
18926660, scsi1, channel 0, id 2, lun 0 Read (10) 00 1c b0 05 97 00 00 08 00
Apr 15 04:34:11 snood kernel: pdc-ultra:[info] scsi abort success
Apr 15 04:34:18 snood kernel: pdc-ultra:[error] disk3 error at LBA 0xb8eb07f
cmd=0x25 status=0x51 error=0x40
Apr 15 04:34:26 snood last message repeated 3 times
Apr 15 04:34:30 snood kernel: pdc-ultra:[error] disk3 error at LBA 0xe0405c7
cmd=0x25 status=0x51 error=0x40
Apr 15 04:34:31 snood kernel: pdc-ultra:[error] disk3 error at LBA 0xe0405c7
cmd=0x25 status=0x51 error=0x40
Apr 15 04:34:37 snood kernel: pdc-ultra:[warning] disk3 ATA timeoutat LBA
0xe0405c7
Apr 15 04:34:37 snood kernel: pdc-ultra:[warning] submit cam busy



I'm running suse 9.0/kernel 2.4.21-202-smp4G , with raidtools-1.00.3-119;
have 5/250gb maxtor Sata drives with 2 Promise S150 tx4 controllers (just
a sata controller, not raid) All disk partitions are typed as FD (linux
raid).

I've had the system crash at least 4 times in six months. Very frustrating.
In some cases it seemed like a real disk error, as above, but might the
Superblock warning show up later? Could the 2 problems be related? Between
crashes the raid works fine, backups, etc.... thanks again for any response,



Gary