Hello, Here I want to share my experience about the JNI installation in E3500. One of my memebers was stuck while connecting E3500 to Hitachi disk array through Brocade switch. Machine is E3500(one I/O board with two JNI 1473 HBA and one wide SCSI sbus card), Solaris 8 with all latest patches. There is no internal fiber disks and machine is booting from external disk(connected to onboard scsi interface) mirrored to the another external disk(connected to the SCSI sbus card). After he installed the volume manager, mirror the disks, restart, the system will crash. He has to boot to single user mode and break the mirror, redo mirror, restart, crash again. The messages is: === WARNING:v 7[AjFiT14] :Bunsc E6r1ror okn oSyopstaemon Bus ein privileged mode from CPU6 Data access at TL=0, errID 0x00000076.7b0808c4 AFSR 0x00000000.84000000<PRIV,BERR> AFAR 0x000001c5.20400020 AFSR.PSYND 0x0000(Score 05) AFSR.ETS 0x00 Fault_PC 0x78030178 UDBH 0x0000 UDBH.ESYND 0x00 UDBL 0x0000 UDBL.ESYND 0x00 panic[cpu6]/thread=2a100145d20: [AFT1] errID 0x00000076.7b0808c4 BERR Error(s) See previous message(s) for details === Going through the core file, I found this crash is attributed to the JNI driver(JNIC146x). Upgrading the driver to the latest version(v5.3.0.3) with the latest Fcode(3.9) doesn't help a lot. Finally, I changed the Sbus card layout on E3500 I/O board, two JNI HBAs share one sbus and the SCSI card is on a separate sbus. After that, everything is fine. I add 50M/s load to this system and it is doing well. Does anyone else have the same experience? thanks, carl _______________________________________________ sunmanagers mailing list sunmanagers@sunmanagers.org http://www.sunmanagers.org/mailman/listinfo/sunmanagersReceived on Thu Mar 11 09:37:17 2004
This archive was generated by hypermail 2.1.8 : Thu Mar 03 2016 - 06:43:27 EST