Hardware

Hi there,

I have a SFE25K with several domains. Some domains with Solaris 9 and others with Solaris 10. They have ISP2312 HBAs with firmware from 1.14.09 to 1.14.15, these are connected to a Director 4800 with the following version info:

Kernel: 2.4.19
Fabric OS: v5.1.0d
Made on: Tue Aug 8 18:35:50 2006
Flash: Thu Nov 23 23:17:35 2006
BootProm: 4.5.3

The firmware on the switch was recently upgraded without rebooting the switch.

/var/adm/messages on all domains record these messages every half hour:

Feb 15 13:23:39 myhost fp: [ID 517869 kern.info] NOTICE: fp(0): PLOGI to 610c13 failed state=Timeout, reason=Hardware Error
Feb 15 13:23:39 myhost fp: [ID 517869 kern.info] NOTICE: fp(0): PLOGI to 610c13 failed state=Timeout, reason=Hardware Error
Feb 15 13:23:39 myhost fctl: [ID 517869 kern.warning] WARNING: 184561=>fp(0)::PLOGI to 610c13 failed. state=c reason=1.
Feb 15 13:23:39 myhost fctl: [ID 517869 kern.warning] WARNING: 184561=>fp(0)::PLOGI to 610c13 failed. state=c reason=1.
Feb 15 13:53:39 myhost fp: [ID 517869 kern.info] NOTICE: fp(0): PLOGI to 610c13 failed state=Timeout, reason=Hardware Error
Feb 15 13:53:39 myhost fp: [ID 517869 kern.info] NOTICE: fp(0): PLOGI to 610c13 failed state=Timeout, reason=Hardware Error
Feb 15 13:53:39 myhost fctl: [ID 517869 kern.warning] WARNING: 184651=>fp(0)::PLOGI to 610c13 failed. state=c reason=1.
Feb 15 13:53:39 myhost fctl: [ID 517869 kern.warning] WARNING: 184651=>fp(0)::PLOGI to 610c13 failed. state=c reason=1.

In addition, every time we attach a new fiber from the hba to the switch in order to get new disk or tape devices, Solaris does not detect anything until a reconfiguration reboot (reboot -- -r). We tried cfgadm -c configure <dev>, luxadm -e forcelip <dev>, with not success.

Any advice would be appreciated.