On Mon, 2002-07-01 at 14:51, Edwin Fung wrote: > Hi, > > We have a 1.6 GHz Pentium with an Adaptec 2930 SCSI card, and we connect it > to an external Sony SDT 9000 4-mm tape drive. We have no problems with > doing "tar" or "dump" when we were running Red Hat 7.0, but when we upgrade > to Red Hat 7.1, we keep getting errors. Red Hat 7.1 does recognize the tape > drive, and when we use "tar" to copy files to the tape it actually does > some of the files in the beginning, but before too long (within 15 secs) > "tar" will quit. The following is a sample of the "tar" failures: > --snip-- > > From /var/log/messages, we got these: > > Jun 3 14:08:03 ewxterm kernel: st0: Block limits 1 - 16777215 bytes. > Jun 3 14:08:42 ewxterm kernel: st0: Error with sense data: [valid=0] Info > fld=0x0, Current st09:00: > sense key Illegal Request > Jun 3 14:08:42 ewxterm kernel: Additional sense indicates Invalid command > operation code > Jun 3 14:08:46 ewxterm kernel: (scsi0:0:6:0) Data Overrun during SEND_SENSE > operation. > Jun 3 14:08:52 ewxterm kernel: (scsi0:0:6:0) CHECK_CONDITION on > REQUEST_SENSE, returning an error. > Jun 3 14:08:52 ewxterm kernel: st0: Error 27070002 (sugg. bt 0x20, driver > bt 0x7, host bt 0x7). > Jun 3 14:24:03 ewxterm kernel: scsi : aborting command due to timeout : pid > 0, scsi0, channel 0, id 6, > lun 0 Read (6) 01 00 00 20 00 > Jun 3 14:24:05 ewxterm kernel: SCSI host 0 abort (pid 0) timed out - resetting > Jun 3 14:24:05 ewxterm kernel: SCSI bus is being reset for host 0 channel 0. > Jun 3 14:24:05 ewxterm kernel: (scsi0:0:6:0) Synchronous at 10.0 Mbyte/sec, > offset 15. > Jun 3 14:24:05 ewxterm kernel: (scsi0:0:6:0) CHECK_CONDITION on > REQUEST_SENSE, returning an error. > Jun 3 14:24:05 ewxterm kernel: st0: Error 27070002 (sugg. bt 0x20, driver > bt 0x7, host bt 0x7). >
These errors are probably because of improper termination of the SCSI bus, or trying to drive the device at a speed it doesn't support. Were there any hardware changes, including different cables, during the upgrade to 7.1? What is the total length of your SCSI bus and how many devices are there on it? The SDT-9000 is not an ultra-SCSI device; are there any ultra-SCSI devices on your bus? Is the tape drive being used to terminate the bus or is there another device or terminator? Adaptec lists AHA-2930U2, AHA-2930B, AHA-2930CU, and AHA-2930U adapters. Do you know which specific model you have? I would first check to be sure both ends of the SCSI bus are properly terminated. If you have any Ultra or Ultra-2 devices on this bus, make sure you meet the most stringent requirement. This tape drive supports a maximum burst rate of 5mb/s asynchronous and 10mb/s synchronous. If the speed is not being negotiated properly for the device, that can cause problems. I have only seen negotiation problems when there is a physical problem with the bus. > > As I mentioned, the tape drive works fine with Red Hat 7.0, but not Red Hat > 7.1. Any > suggestions/explanations will be much appreciated. Are you still able to run 7.0? Can you check your logs while doing a similar operation as the one that fails under 7.1? I believe there are two flavors of the Adaptec SCSI driver. Are you using the same one in 7.1 that you were in 7.0? Linus _______________________________________________ Redhat-list mailing list [EMAIL PROTECTED] https://listman.redhat.com/mailman/listinfo/redhat-list
