Server Clustering
cancel
Showing results for 
Search instead for 
Did you mean: 

kernel: sym53c896-0: CL380, Linux RH 7.3

Rodion
Occasional Contributor

kernel: sym53c896-0: CL380, Linux RH 7.3

Proliant CL380
CR3500 Shared Storage Controller.
RAID 5 on Shared Storage.
Two server node.
Linux RH 7.3
KERNEL-2.4.20-28.7.i386
PSP v 7.00 (Proliant Support Pack).
Informix 9.21. UC3.

When I try to read data from shared storage disks,I catch errors on both servers.
On the first server:

kernel: sym53c896-0: SCSI phase error fixup: CCB already dequeued.
kernel: sym53c896-0: handling phase mismatch from SCRIPTS.

And this errors on the second server:
kernel: sym53c896-0: handling phase mismatch from SCRIPTS.
kernel: sym53c896-0:2: ERROR (a0:0) (40-67-20) (0/7) @ (mem 8:00000000).
kernel: sym53c896-0: regdump: ca 10 c0 07 47 00 02 0a 20 40 82 67 80 00 07 00 90 c3 ff c6 08 00 00 00.
kernel: sym53c896-0:2: message c sent on bad reselection.
kernel: sym53c896-0: SCSI phase error fixup: CCB already dequeued.

1 REPLY
Serviceguard for Linux
Honored Contributor

Re: kernel: sym53c896-0: CL380, Linux RH 7.3

So the CR3500 is not (offically?) support on Linux. I'd check the driver source and see if there any indication in the comments that it is written to support a shared environment.

It was designed for Microsoft MSCS. This SW uses SCSI reserve to reserve a LUN for one server and keep the other from accessing it. If you are trying to access the same LUN for 2 servers, it may not be able to handle it.