Hardware

Hi there,

I have recently inherited administration of some v880s with A1000s attached to them, and am seeing some I/O performance issues.

I�ve got one V880, with Solaris 8 2/02, and RAID manager 6.22 on it, and 2 A1000s hanging off it, off separate SCSI connections from the same PCI card.

I�ve run Sun Install Check to ensure that my patch revisions are OK, and they seem to be fine.

Each A1000 is on a separate target, configured as RAID 5, and RAID manager reports them as optimal, and each is configured as a single LUN consisting of 10 disks, with 2 hot spares.

However, �format� sees them both differently��

One of them is seen as a single device (c3t5d0), and another is seen as several devices (c2, t0 � t15, d0), as below:

AVAILABLE DISK SELECTIONS:
0. c1t0d0 <SUN72G cyl 14087 alt 2 hd 24 sec 424> /pci@8,600000/SUNW,qlc@2/fp@0,0/ssd@w500000e01049d6a1,0
1. c1t1d0 <SUN72G cyl 14087 alt 2 hd 24 sec 424> /pci@8,600000/SUNW,qlc@2/fp@0,0/ssd@w500000e01049bd91,0
2. c1t2d0 <SUN72G cyl 14087 alt 2 hd 24 sec424>
/pci@8,600000/SUNW,qlc@2/fp@0,0/ssd@w500000e01049ad21,0
3. c1t3d0 <SUN72G cyl 14087 alt 2 hd 24 sec 424>
/pci@8,600000/SUNW,qlc@2/fp@0,0/ssd@w500000e01049ae51,0
4. c1t4d0 <SUN72G cyl 14087 alt 2 hd 24 sec 424>
/pci@8,600000/SUNW,qlc@2/fp@0,0/ssd@w500000e01049aa41,0
5. c1t5d0 <SUN72G cyl 14087 alt 2 hd 24 sec 424>
/pci@8,600000/SUNW,qlc@2/fp@0,0/ssd@w500000e01049d6f1,0
6. c2t0d0 <Symbios-StorEDGEA1000-0301 cyl 65533 alt 2 hd 64 sec 152>
/pseudo/rdnexus@2/rdriver@0,0
7. c2t1d0 <Symbios-StorEDGEA1000-0301 cyl 65533 alt 2 hd 64 sec 152>
/pci@9,600000/scsi@1,1/sd@1,0
8. c2t2d0 <Symbios-StorEDGEA1000-0301 cyl 65533 alt 2 hd 64 sec 152>
/pci@9,600000/scsi@1,1/sd@2,0
9. c2t3d0 <Symbios-StorEDGEA1000-0301 cyl 65533 alt 2 hd 64 sec 152>
/pci@9,600000/scsi@1,1/sd@3,0
10. c2t4d0 <Symbios-StorEDGEA1000-0301 cyl 65533 alt 2 hd 64 sec 152>
/pci@9,600000/scsi@1,1/sd@4,0
11. c2t5d0 <Symbios-StorEDGEA1000-0301 cyl 65533 alt 2 hd 64 sec 152>
/pci@9,600000/scsi@1,1/sd@5,0
12. c2t6d0 <Symbios-StorEDGEA1000-0301 cyl 65533 alt 2 hd 64 sec 152>
/pci@9,600000/scsi@1,1/sd@6,0
13. c2t8d0 <Symbios-StorEDGEA1000-0301 cyl 65533 alt 2 hd 64 sec 152>
/pci@9,600000/scsi@1,1/sd@8,0
14. c2t9d0 <Symbios-StorEDGEA1000-0301 cyl 65533 alt 2 hd 64 sec 152>
/pci@9,600000/scsi@1,1/sd@9,0
15. c2t10d0 <Symbios-StorEDGEA1000-0301 cyl 65533 alt 2 hd 64 sec 152>
/pci@9,600000/scsi@1,1/sd@a,0
16. c2t11d0 <Symbios-StorEDGEA1000-0301 cyl 65533 alt 2 hd 64 sec 152>
/pci@9,600000/scsi@1,1/sd@b,0
17. c2t12d0 <Symbios-StorEDGEA1000-0301 cyl 65533 alt 2 hd 64 sec 152>
/pci@9,600000/scsi@1,1/sd@c,0
18. c2t13d0 <Symbios-StorEDGEA1000-0301 cyl 65533 alt 2 hd 64 sec 152>
/pci@9,600000/scsi@1,1/sd@d,0
19. c2t14d0 <Symbios-StorEDGEA1000-0301 cyl 65533 alt 2 hd 64 sec 152>
/pci@9,600000/scsi@1,1/sd@e,0
20. c2t15d0 <Symbios-StorEDGEA1000-0301 cyl 65533 alt 2 hd 64 sec 152>
/pci@9,600000/scsi@1,1/sd@f,0
21. c3t5d0 <Symbios-StorEDGEA1000-0301 cyl 65533 alt 2 hd 64 sec 152>
/pseudo/rdnexus@3/rdriver@5,0
Specify disk (enter its number):

Now, c2t0d0s6 is mounted and being used for a SAS application to write to, and I�m having I/O issues on when running this SAS process (it analyses a lot of data and accesses that file system).

When this process is running, the I/O activity shows that the %w is 100%, %b is 100%, and we have a wait queue which reaches 38!!!! This is not good. The disk goes BLACK in Cockroft's SE ZOOM tool :-/

I�m not sure if this is caused by the different ways that Solaris sees the A1000s, or whether it is a different root cause?

Also, does anyonehave any idea why they�re seen in these different ways, despite the fact that they�re both configured in the same way?

Any help you can give would be GREATLY appreciated.