Hello,
I have a problem with what I believe might be related to the aic79xx driver.
I'm running 2.4.33.3 kernel with default aic79xx driver version 1.3.10, no options.
There're 2 drives in software RAID 1, hostRAID disabled:
root@mail2:~# cat /proc/scsi/scsi
Attached devices:
Host: scsi0 Channel: 00 Id: 00 Lun: 00
Vendor: COMPAQ Model: BD3008A4C6 Rev: HPB4
Type: Direct-Access ANSI SCSI revision: 03
Host: scsi0 Channel: 00 Id: 01 Lun: 00
Vendor: COMPAQ Model: BD3008A4C6 Rev: HPB4
Type: Direct-Access ANSI SCSI revision: 03
root@mail2:~# cat /proc/scsi/aic79xx/0
Adaptec AIC79xx driver version: 1.3.10
Adaptec AIC7902 Ultra320 SCSI adapter
aic7902: Ultra320 Wide Channel A, SCSI Id=7, PCI-X 67-100Mhz, 512 SCBs
Allocated SCBs: 64, SG List Length: 102
Serial EEPROM:
0x17c8 0x17c8 0x17c8 0x17c8 0x17c8 0x17c8 0x17c8 0x17c8
0x17c8 0x17c8 0x17c8 0x17c8 0x17c8 0x17c8 0x17c8 0x17c8
0x09f4 0x0146 0x2807 0x0010 0xffff 0xffff 0xffff 0xffff
0xffff 0xffff 0xffff 0xffff 0xffff 0xffff 0x0430 0xb3f7
Target 0 Negotiation Settings
User: 320.000MB/s transfers (160.000MHz RDSTRM|DT|IU|RTI|QAS, 16bit)
Goal: 320.000MB/s transfers (160.000MHz DT|IU|RTI|QAS, 16bit)
Curr: 320.000MB/s transfers (160.000MHz DT|IU|RTI|QAS, 16bit)
Transmission Errors 0
Channel A Target 0 Lun 0 Settings
Commands Queued 8762259
Commands Active 3
Command Openings 29
Max Tagged Openings 32
Device Queue Frozen Count 0
Target 1 Negotiation Settings
User: 320.000MB/s transfers (160.000MHz RDSTRM|DT|IU|RTI|QAS, 16bit)
Goal: 320.000MB/s transfers (160.000MHz DT|IU|RTI|QAS, 16bit)
Curr: 320.000MB/s transfers (160.000MHz DT|IU|RTI|QAS, 16bit)
Transmission Errors 0
Channel A Target 1 Lun 0 Settings
Commands Queued 8784600
Commands Active 2
Command Openings 30
Max Tagged Openings 32
Device Queue Frozen Count 0
...
The machine is a mail server that was recently migrated to the new hardware
and now the load average is constantly between 10 and 20 and sometimes
climbs higher. top and vmstat shows lots of processes in uninterruptible
sleep, which as far I understand means that the processes wait while the driver
code to complete:
root@mail2:~# vmstat 1
procs -----------memory---------- ---swap-- -----io---- -system-- ----cpu----
r b swpd free buff cache si so bi bo in cs us sy id wa
0 12 0 316004 100400 1026312 0 0 117 55 180 183 4 4 92 0
1 3 0 337004 100400 1026336 0 0 24 2052 526 1481 4 2 94 0
0 14 0 334832 100404 1026432 0 0 0 1528 510 1059 4 2 94 0
0 8 0 336312 100408 1026424 0 0 0 1868 494 1198 6 4 90 0
I wonder if anyone has seen anything like this and/or can advice any cure?
Do you think upgrading the driver to the latest provided by adaptec (2.0.20 I guess)
should help? Any other buttons to push?
There's another server running alongside with similar IO load on 2 SATA drives in software
RAID1 and load average never gets above 0.5
Please let me know if you need more info.
Thanks a lot,
Michael
-
To unsubscribe from this list: send the line "unsubscribe linux-scsi" in
the body of a message to majordomo@xxxxxxxxxxxxxxx
More majordomo info at http://vger.kernel.org/majordomo-info.html
