Date: Thu, 4 Apr 2002 13:35:00 -0800 From: Carl Reisinger <cer@mirapoint.com> To: freebsd-scsi@freebsd.org Subject: Any ideas what the following ahc errors mean? Message-ID: <200204042132.APW11519@mail.mirapoint.com>
next in thread | raw e-mail | index | archive | help
I am running FreeBSD 4.1-stable. The motherboard is a Intel STL2 with a 1.4GHz processor. On-board is a Adaptec AIC 7899G chipset, the channel being used is channel A of that chipset. The SCSI bus has three devices a StorageTek L40 (robotics), Quantum DLT8000 and a Seagate Viper 200 (LTO). Bus length is about 10 feet. I have been writting massive amounts of data to both drives until I hit the physical EOM, but just writting and have not seen any problems (of course the filemarks do not get written and the close returns a ENXIO error, but this is to be expected). Until this test I just closed the devices and started again. Now I am writting 128 KB records (yes, this required a change to the driver) to the LTO until I reach logical EOM (the first time a ENOSPC gets returned). I then issue a MTREW with the following results: 2002 Apr 04 15:17:27 rd12 /kernel: ahc0:A:2: no active SCB for reconnecting targ et - issuing BUS DEVICE RESET 2002 Apr 04 15:17:27 rd12 /kernel: SAVED_SCSIID == 0x27, SAVED_LUN == 0x0, ARG_1 == 0xff ACCUM = 0x80 2002 Apr 04 15:17:27 rd12 /kernel: SEQ_FLAGS == 0x0, SCBPTR == 0x0, BTT == 0xff, SINDEX == 0x31 2002 Apr 04 15:17:27 rd12 /kernel: SCSIID == 0x0, SCB_SCSIID == 0x27, SCB_LUN == 0x0, SCB_TAG == 0xff, SCB_CONTROL == 0x40 2002 Apr 04 15:17:27 rd12 /kernel: SCSIBUSL == 0x0, SCSISIGI == 0xc6 2002 Apr 04 15:17:27 rd12 /kernel: SXFRCTL0 == 0x88 2002 Apr 04 15:17:27 rd12 /kernel: SEQCTL == 0x10 2002 Apr 04 15:17:27 rd12 /kernel: ahc0: Dumping Card State at SEQADDR 0x1a8 2002 Apr 04 15:17:27 rd12 /kernel: SCSISEQ = 0x12, SBLKCTL = 0xa, SSTAT0 0x2 2002 Apr 04 15:17:27 rd12 /kernel: SCB count = 20 2002 Apr 04 15:17:27 rd12 /kernel: Kernel NEXTQSCB = 14 2002 Apr 04 15:17:27 rd12 /kernel: Card NEXTQSCB = 14 2002 Apr 04 15:17:27 rd12 /kernel: QINFIFO entries: 2002 Apr 04 15:17:27 rd12 /kernel: Waiting Queue entries: 2002 Apr 04 15:17:27 rd12 /kernel: Disconnected Queue entries: 2002 Apr 04 15:17:27 rd12 /kernel: QOUTFIFO entries: 2002 Apr 04 15:17:27 rd12 /kernel: Sequencer Free SCB List: 0 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 2002 Apr 04 15:17:27 rd12 /kernel: Pending list: 2002 Apr 04 15:17:27 rd12 /kernel: Kernel Free SCB list: 9 8 7 15 16 17 18 19 0 1 2 3 4 5 6 13 12 11 10 2002 Apr 04 15:17:27 rd12 /kernel: ahc0: Bus Device Reset on A:2. 0 SCBs aborted I have the EOTMODEL set to 1 filemark. Interesting note: it appears the tape is not being rewound. If I issue a read after the ioctl returns the drive issues a CONDITION CHECK with a sense key of 8 and asc/ascq of 0/5. Which is a BLANK CHECK (EOD encountered). Anyone have an idea? Carl Reisinger 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?200204042132.APW11519>