I have two firewire drives that are the same brand. One old one (one year), and one new one. I have Suse 9.1 and Mac OSX systems. The old drive can be seen on the mac and the linux box. The new drive can only be seen on the mac box. The new drive does not show up in a Yast hardware scan or /etc/fstab. Here are the relevant messages from /var/log/messages: Sep 16 23:45:20 Cosmos3 kernel: ieee1394: Error parsing configrom for node 0-00:1023 Sep 16 23:45:20 Cosmos3 kernel: ieee1394: Node changed: 0-00:1023 -> 0-01:1023 Sep 16 23:45:25 Cosmos3 kernel: ieee1394: Node resumed: ID:BUS[0-00:1023] GUID[00d04b43e0091193] Sep 16 23:45:25 Cosmos3 kernel: scsi3 : SCSI emulation for IEEE-1394 SBP-2 Devices Sep 16 23:45:26 Cosmos3 kernel: ieee1394: sbp2: Logged into SBP-2 device Sep 16 23:45:26 Cosmos3 kernel: ieee1394: Node 0-00:1023: Max speed [S400] - Max payload [2048] Sep 16 23:45:26 Cosmos3 kernel: Vendor: WDC WD16 Model: 00BB-00DWA0 Rev: 15.0 Sep 16 23:45:26 Cosmos3 kernel: Type: Direct-Access ANSI SCSI revision: 06 Sep 16 23:45:26 Cosmos3 kernel: SCSI device sda: 312581808 512-byte hdwr sectors (160042 MB) Sep 16 23:45:26 Cosmos3 kernel: sda: asking for cache data failed Sep 16 23:45:26 Cosmos3 kernel: sda: assuming drive cache: write through Sep 16 23:45:26 Cosmos3 /etc/hotplug/block.agent[10013]: try 1 while waiting for /block/sda's bus_id 3:0:0:0 Sep 16 23:45:26 Cosmos3 kernel: sda: unknown partition table Sep 16 23:45:26 Cosmos3 kernel: Attached scsi disk sda at scsi3, channel 0, id 0, lun 0 Sep 16 23:45:26 Cosmos3 kernel: Attached scsi generic sg0 at scsi3, channel 0, id 0, lun 0, type 0 Sep 16 23:45:27 Cosmos3 /etc/hotplug/block.agent[10013]: new block device /block/sda What do I do to fix this? Thanks, Jerome
I'm also trying to diagnose a problem with 9.1 detecting all my firewire drives. Just curious, are the two firewire drives daisy-chained together, or does each drive go to a separate firewire port on the Linux box? Mark -----Original Message----- From: Jerome Lyles [mailto:susemail@hawaii.rr.com] Sent: Friday, September 17, 2004 5:55 AM To: suse-linux-e@suse.com Subject: [SLE] New Firewire drive not seen I have two firewire drives that are the same brand. One old one (one year), and one new one. I have Suse 9.1 and Mac OSX systems. The old drive can be seen on the mac and the linux box. The new drive can only be seen on the mac box. The new drive does not show up in a Yast hardware scan or /etc/fstab. Here are the relevant messages from /var/log/messages: Sep 16 23:45:20 Cosmos3 kernel: ieee1394: Error parsing configrom for node 0-00:1023 Sep 16 23:45:20 Cosmos3 kernel: ieee1394: Node changed: 0-00:1023 -> 0-01:1023 Sep 16 23:45:25 Cosmos3 kernel: ieee1394: Node resumed: ID:BUS[0-00:1023] GUID[00d04b43e0091193] Sep 16 23:45:25 Cosmos3 kernel: scsi3 : SCSI emulation for IEEE-1394 SBP-2 Devices Sep 16 23:45:26 Cosmos3 kernel: ieee1394: sbp2: Logged into SBP-2 device Sep 16 23:45:26 Cosmos3 kernel: ieee1394: Node 0-00:1023: Max speed [S400] - Max payload [2048] Sep 16 23:45:26 Cosmos3 kernel: Vendor: WDC WD16 Model: 00BB-00DWA0 Rev: 15.0 Sep 16 23:45:26 Cosmos3 kernel: Type: Direct-Access ANSI SCSI revision: 06 Sep 16 23:45:26 Cosmos3 kernel: SCSI device sda: 312581808 512-byte hdwr sectors (160042 MB) Sep 16 23:45:26 Cosmos3 kernel: sda: asking for cache data failed Sep 16 23:45:26 Cosmos3 kernel: sda: assuming drive cache: write through Sep 16 23:45:26 Cosmos3 /etc/hotplug/block.agent[10013]: try 1 while waiting for /block/sda's bus_id 3:0:0:0 Sep 16 23:45:26 Cosmos3 kernel: sda: unknown partition table Sep 16 23:45:26 Cosmos3 kernel: Attached scsi disk sda at scsi3, channel 0, id 0, lun 0 Sep 16 23:45:26 Cosmos3 kernel: Attached scsi generic sg0 at scsi3, channel 0, id 0, lun 0, type 0 Sep 16 23:45:27 Cosmos3 /etc/hotplug/block.agent[10013]: new block device /block/sda What do I do to fix this? Thanks, Jerome -- Check the headers for your unsubscription address For additional commands send e-mail to suse-linux-e-help@suse.com Also check the archives at http://lists.suse.com Please read the FAQs: suse-linux-e-faq@suse.com
On Friday 17 September 2004 01:54 am, Jerome Lyles wrote:
for /block/sda's bus_id 3:0:0:0 Sep 16 23:45:26 Cosmos3 kernel: sda: unknown partition table Sep 16 23:45:26 Cosmos3 kernel: Attached scsi disk sda at scsi3, channel 0, id 0, lun 0 Sep 16 23:45:26 Cosmos3 kernel: Attached scsi generic sg0 at scsi3, channel 0, id 0, lun 0, type 0 Sep 16 23:45:27 Cosmos3 /etc/hotplug/block.agent[10013]: new block device /block/sda
Looks like it was found to me... Its on scsi bus 3.
Sometimes SuSE does not scan all scsi buses. Often it only
scans the first two busses, (numbered bus 0 and 1.)
You can check by doing (as root)
cat /proc/scsi/scsi
If you only see two busses listed, then do (as root again)
echo "scsi add-single-device 2 0 0 0" > /proc/scsi/scsi
That will make it visible at least. Then you can rescan the
scsi bus and it should find the drive... As root:
rescan-scsi-bus.sh 2
Sometime you have to hunt for rescan-scsi-bus.sh .
Here is the one I use on SuSE 8.2
norvio:~ # cat bin/rescan-scsi-bus.sh
#!/bin/bash
# Skript to rescan SCSI bus, using the
# scsi add-single-device mechanism
# (w) 1998-03-19 Kurt Garloff
On Saturday 18 September 2004 09:38 pm, John Andersen wrote:
On Friday 17 September 2004 01:54 am, Jerome Lyles wrote:
for /block/sda's bus_id 3:0:0:0 Sep 16 23:45:26 Cosmos3 kernel: sda: unknown partition table Sep 16 23:45:26 Cosmos3 kernel: Attached scsi disk sda at scsi3, channel 0, id 0, lun 0 Sep 16 23:45:26 Cosmos3 kernel: Attached scsi generic sg0 at scsi3, channel 0, id 0, lun 0, type 0 Sep 16 23:45:27 Cosmos3 /etc/hotplug/block.agent[10013]: new block device /block/sda
Looks like it was found to me... Its on scsi bus 3.
Sometimes SuSE does not scan all scsi buses. Often it only scans the first two busses, (numbered bus 0 and 1.) You can check by doing (as root) cat /proc/scsi/scsi If you only see two busses listed, then do (as root again) echo "scsi add-single-device 2 0 0 0" > /proc/scsi/scsi
Look at this (as root): # cat /proc/scsi/scsi Attached devices: Host: scsi4 Channel: 00 Id: 00 Lun: 00 Vendor: WDC WD16 Model: 00BB-00DWA0 Rev: 15.0 Type: Direct-Access ANSI SCSI revision: 06 This is not the output I expected. Is it what you expected. Should I still run echo "scsi add-single-device 2 0 0 0" > /proc/scsi/scsi? Thanks, Jerome
On Friday 17 September 2004 10:54, Jerome Lyles wrote:
I have two firewire drives that are the same brand. One old one (one year), and one new one. I have Suse 9.1 and Mac OSX systems. The old drive can be seen on the mac and the linux box. The new drive can only be seen on the mac box. The new drive does not show up in a Yast hardware scan or /etc/fstab.
Here are the relevant messages from /var/log/messages: [snip] Sep 16 23:45:26 Cosmos3 kernel: sda: unknown partition table [snip]
This is the pertinent line. It found the device and set it up as a SCSI device. However the partition table makes no sense. I've seen this a few times. What I think the problem is, is that the device has a file system directly on /dev/sda. Ordinarily in Linux (and particularly with the autodetecting USB and firewire stuff) it expects a partition table defining /dev/sda1, /dev/sda2, etc. Then looks for filesystems on the /dev/sda#'s. The fudge way is to drop to a root console, and do the following: mkdir /mnt/tmp mount /dev/sda /mnt/tmp ls -la /mnt/tmp This is quite likely to work, although you'll need some files on the device to see. Don't forget to run: umount /mnt/tmp before you disconnect, or you'll risk corruption. This would have to be repeated everytime you plug the device in. A more permanent solution is to backup any files on the new disc to the Mac OSX, plug in the device in to Linux, then do: fdisk /dev/sda Now create some partitions, save and exit, create appropriate filesystems, then disconnect the drive. Now plug it back into Linux, and it should all automount. HTH -- Steve Boddy
participants (4)
-
Jerome Lyles
-
John Andersen
-
Mark Hilbush
-
Stephen Boddy