Date: Tue, 25 Aug 1998 02:38:43 -0700 (PDT) From: asami@cs.berkeley.edu (Satoshi Asami) To: scsi@FreeBSD.ORG Subject: parity errors Message-ID: <199808250938.CAA01303@silvia.hip.berkeley.edu>
next in thread | raw e-mail | index | archive | help
Hi, One of our machines started hanging (or crashing) occasionally. This is the message that was printed out to the serial console just before the disk array (ccd of 13 disks) locked up. === (da8:ahc0:0:8:0): WRITE(10). CDB: 2a 0 0 cc a7 9f 0 0 30 0 (da8:ahc0:0:8:0): ABORTED COMMAND asc:47,0 (da8:ahc0:0:8:0): SCSI parity error (da1:ahc0:0:1:0): parity error during Data-In phase. === I could still log in to the machine, but any process (bash) that did anything with the array will lock up. Also, this is what was on the console before the machine hung before. (It was a hard hang, needed a reset.) === (da2:ahc0:0:2:0): WRITE(10). CDB: 2a 0 0 82 74 ef 0 0 40 0 (da2:ahc0:0:2:0): ABORTED COMMAND asc:47,0 (da2:ahc0:0:2:0): SCSI parity error (da3:ahc0:0:3:0): parity error during Data-In phase. (da3:ahc0:0:3:0): SCB 0xb - timed out while idle, LASTPHASE == 0x1, SCSISIGI == 0xe6 SEQADDR == 0xb (da3:ahc0:0:3:0): Queueing a BDR SCB (da3:ahc0:0:3:0): Bus Device Reset Message Sent Bus Device Reset Completed. (da3:ahc0:0:3:0): no longer in timeout ahc0: Bus Device Reset delivered. 1 SCBs aborted === It does seem to come back up after a reboot/power cycle and work fine for a while (a few days to a week) though. So, two questions. I assume the SCSI parity error means bus integrity problems (enclosure went bad or something)? Also, if it's going to come back after a reboot, can it be done without actual rebooting? (Like sending a bus reset....under no cases have we actually power-cycled the disk enclosures.) Satoshi To Unsubscribe: send mail to majordomo@FreeBSD.org with "unsubscribe freebsd-scsi" in the body of the message
Want to link to this message? Use this URL: <https://mail-archive.FreeBSD.org/cgi/mid.cgi?199808250938.CAA01303>