Re: [opensuse] UUID=>0x and LVM
Hi,
Quoting "Boyd Stephen Smith Jr."
In Monday 15 June 2009, you wrote:
To see the UUIDs on the system as LVM sees them issue the command: pvscan -u
Booting from the system, this gives:
Couldn't find device with uuid ' lLWxB7-QmLl-FMXX-KQRW-EDH2-EHLL-o7D21b' Couldn't find device with uuid ' lLWxB7-QmLl-FMXX-KQRW-EDH2-EHLL-o7D21b' Couldn't find device with uuid ' lLWxB7-QmLl-FMXX-KQRW-EDH2-EHLL-o7D21b' PV /dev/sda4 with UUID ecQz2l-Yl5S-s1AR-delt-5zXF-Wnjs-P2y2dH VG system lvm2 [113.03 GB / 0 free] PV /dev/sdb1 with UUID 5P6jGx-ePjX-43Az-wyLL-wuF5-R4Vt-vP9Uff VG system lvm2 [465.76 / 0 free] PV unknown device with UUID lLWxB7-QmLl-FMXX-KQRW-EDHZ-EHLL-o7DZ1b VG system lvm2 [465.76 GB / 224.54 GB free] Total: 3 [1.02 TB] / in use: 3 [1.02 TB] / in no VG: 0 [0 ]
When I boot from the GParted live-CD:
PV /dev/sdb4 with UUID ecQz2l-Yl5S-s1AR-delt-5zXF-Wnjs-P2y2dH VG system lvm2 [113.03 GB / 0 free] PV /dev/sdc1 with UUID 5P6jGx-ePjX-43Az-wyLL-wuF5-R4Vt-vP9Uff VG system lvm2 [465.76 GB / 0 free] PV /dev/sda with UUID lLWxB7-QmLl-FMXX-KQRW-EDHZ-EHLL-o7DZ1b VG system lvm2 [465.76 GB / 224.54 GB free] Total: 3 [1.02 TB] /in use: 3 [1.02 TB] / in no VG: 0 [0 ]
No error at all. How does the GParted Live-CD do that?
This is something different. As the GParted Live-CD shows, the device has the correct UUID. This means that, for whatever reason, LVM is not scanning the correct device(s) to find a UUID. Either that, or the disk itself isn't being detected by the kernel, so the block devices (for the disk and partitions) are not being created.
When you boot the system normally, which device is the drive that LVM can't find?
sdc
You may simply be missing the correct kernel module for the disk controller.
How could I verify that?
Assuming you can determine which device LVM should be finding (perhaps /dev/sdc?), check your /etc/lvm/lvm.conf. Specifically, the "device" section. Mine looks like this: devices { dir = "/dev" scan = [ "/dev" ] filter = [ "r|/dev/.*/by-path/.*|", "r|/dev/.*/by-id/.*|", "a/.*/" ] cache = "/etc/lvm/.cache" write_cache_state = 1 sysfs_scan = 1 md_component_detection = 1 }
Mine is: filter = [ "r|/dev/.*/by-path/.*|", "r|/dev/.*/by-id/.*|", "a/.*/" ]
The critical lines are the "scan"
This one is also the same.
If your device is rejected (name
How could I find its name? Is it 'sdc'? Btw: Here is fdisk's output: Disk /dev/sda: 160.0 GB, 160041885696 bytes 255 heads, 63 sectors/track, 19457 cylinders Units = cylinders of 16065 * 512 = 8225280 bytes Disk identifier: 0x0007e41f Device Boot Start End Blocks Id System /dev/sda1 1 2612 20980858+ c W95 FAT32 (LBA) /dev/sda2 2613 4571 15735667+ 82 Linux swap / Solaris /dev/sda3 * 4572 4702 1052257+ 83 Linux /dev/sda4 4703 19457 118519537+ 8e Linux LVM Disk /dev/sdb: 500.1 GB, 500107862016 bytes 255 heads, 63 sectors/track, 60801 cylinders Units = cylinders of 16065 * 512 = 8225280 bytes Disk identifier: 0x000d7480 Device Boot Start End Blocks Id System /dev/sdb1 1 60801 488384001 8e Linux LVM Disk /dev/sdc: 500.1 GB, 500107862016 bytes 255 heads, 63 sectors/track, 60801 cylinders Units = cylinders of 16065 * 512 = 8225280 bytes Disk identifier: 0x00000000 Disk /dev/sdc doesn't contain a valid partition table Disk /dev/dm-0: 858.9 GB, 858993459200 bytes 255 heads, 63 sectors/track, 104433 cylinders Units = cylinders of 16065 * 512 = 8225280 bytes Disk identifier: 0x00000000 Disk /dev/dm-0 doesn't contain a valid partition table Disk /dev/dm-1: 21.4 GB, 21474836480 bytes 255 heads, 63 sectors/track, 2610 cylinders Units = cylinders of 16065 * 512 = 8225280 bytes Disk identifier: 0x00000000 Disk /dev/dm-1 doesn't contain a valid partition table Thanks a lot for your help! -- To unsubscribe, e-mail: opensuse+unsubscribe@opensuse.org For additional commands, e-mail: opensuse+help@opensuse.org
Philip Mötteli wrote:
You may simply be missing the correct kernel module for the disk controller.
How could I verify that?
lspci and lsmod. Find your disk controller in the lspci output, determine which module it needs and check if that is loaded (lsmod output). /Per -- Per Jessen, Zürich (16.9°C) -- To unsubscribe, e-mail: opensuse+unsubscribe@opensuse.org For additional commands, e-mail: opensuse+help@opensuse.org
Hi, Am 16.06.2009 um 09:57 schrieb Per Jessen:
Philip Mötteli wrote:
You may simply be missing the correct kernel module for the disk controller.
How could I verify that?
lspci
There is no 'lspci' in the fall back shell. But I did it afterwards on the Live-CD. That shouldn't make a difference: 00:00.0 Host bridge: Intel Corporation E7525 Memory Controller Hub (rev 0c) 00:02.0 PCI bridge: Intel Corporation E7525/E7520/E7320 PCI Express Port A (rev 0c) 00:03.0 PCI bridge: Intel Corporation E7525/E7520/E7320 PCI Express Port A1 (rev 0c) 00:04.0 PCI bridge: Intel Corporation E7525/E7520 PCI Express Port B (rev 0c) 00:1d.0 USB Controller: Intel Corporation 82801EB/ER (ICH5/ICH5R) USB UHCI Controller #1 (rev 02) 00:1d.1 USB Controller: Intel Corporation 82801EB/ER (ICH5/ICH5R) USB UHCI Controller #2 (rev 02) 00:1d.2 USB Controller: Intel Corporation 82801EB/ER (ICH5/ICH5R) USB UHCI Controller #3 (rev 02) 00:1d.3 USB Controller: Intel Corporation 82801EB/ER (ICH5/ICH5R) USB UHCI Controller #4 (rev 02) 00:1d.7 USB Controller: Intel Corporation 82801EB/ER (ICH5/ICH5R) USB2 EHCI Controller (rev 02) 00:1e.0 PCI bridge: Intel Corporation 82801 PCI Bridge (rev c2) 00:1f.0 ISA bridge: Intel Corporation 82801EB/ER (ICH5/ICH5R) LPC Interface Bridge (rev 02) 00:1f.1 IDE interface: Intel Corporation 82801EB/ER (ICH5/ICH5R) IDE Controller (rev 02) 00:1f.2 IDE interface: Intel Corporation 82801EB (ICH5) SATA Controller (rev 02) 00:1f.5 Multimedia audio controller: Intel Corporation 82801EB/ER (ICH5/ICH5R) AC'97 Audio Controller (rev 02) 01:00.0 PCI bridge: Intel Corporation 6700PXH PCI Express-to-PCI Bridge A (rev 09) 01:00.1 PIC: Intel Corporation 6700/6702PXH I/OxAPIC Interrupt Controller A (rev 09) 01:00.2 PCI bridge: Intel Corporation 6700PXH PCI Express-to-PCI Bridge B (rev 09) 01:00.3 PIC: Intel Corporation 6700PXH I/OxAPIC Interrupt Controller B (rev 09) 02:05.0 SCSI storage controller: LSI Logic / Symbios Logic 53c1030 PCI- X Fusion-MPT Dual Ultra320 SCSI (rev 07) 02:05.1 SCSI storage controller: LSI Logic / Symbios Logic 53c1030 PCI- X Fusion-MPT Dual Ultra320 SCSI (rev 07) 0b:03.0 RAID bus controller: Adaptec Serial ATA II RAID 1420SA (rev 01) 10:00.0 VGA compatible controller: S3 Inc. 86c325 [ViRGE] (rev 05) 10:02.0 FireWire (IEEE 1394): Texas Instruments TSB43AB22/A IEEE-1394a-2000 Controller (PHY/Link) 10:03.0 Ethernet controller: Intel Corporation 82541GI Gigabit Ethernet Controller (rev 05) 40:00.0 VGA compatible controller: nVidia Corporation NV44 [Quadro NVS 285] (rev a1)
and lsmod.
There was no 'lsmod' either. I hope, a 'modprobe -l' can help too: libata ata_piix usbcore ohci-hcd uhci-hcd edd linear dm-mod dm-snapshot mptscsih mptbase mptspi usbhid hid ff-memless scsi_transport_spi sd_mod scsi_mod processor fan thermal jbd ext3 mbcache
Find your disk controller in the lspci output
Should that be the 'LSI Logic / Symbios Logic 53c1030 PCI-X Fusion-MPT Dual Ultra320 SCSI (rev 07)'?
determine which module it needs
That should be the 'mptscsih' and 'mptbase'?
and check if that is loaded (lsmod output).
In order to be sure, is the 'modprobe -l', that I did, enough? Thanks -- To unsubscribe, e-mail: opensuse+unsubscribe@opensuse.org For additional commands, e-mail: opensuse+help@opensuse.org
Philip Mötteli wrote:
There was no 'lsmod' either. I hope, a 'modprobe -l' can help too:
Uh, on my system 'modprobe -l' gives me a list of all _available_ modules, not just the loaded ones.
Find your disk controller in the lspci output
Should that be the 'LSI Logic / Symbios Logic 53c1030 PCI-X Fusion-MPT Dual Ultra320 SCSI (rev 07)'?
determine which module it needs
That should be the 'mptscsih' and 'mptbase'?
I think so, yes.
and check if that is loaded (lsmod output).
In order to be sure, is the 'modprobe -l', that I did, enough?
Not as far as I can tell, no. Try 'cat /proc/modules' instead. /Per -- Per Jessen, Zürich (18.6°C) -- To unsubscribe, e-mail: opensuse+unsubscribe@opensuse.org For additional commands, e-mail: opensuse+help@opensuse.org
Hi, Am 16.06.2009 um 12:26 schrieb Per Jessen:
Philip Mötteli wrote:
Find your disk controller in the lspci output
Should that be the 'LSI Logic / Symbios Logic 53c1030 PCI-X Fusion- MPT Dual Ultra320 SCSI (rev 07)'?
and check if that is loaded (lsmod output).
Try 'cat /proc/modules' instead.
linear ehci_hcd uhci_hcd sd_mod usbcore dm_snapshot edd dm_mod ext3 mbcache jbd fan mptspi mptscsih mptbase scsi_transport_spi ata_piix libata scsi_mod thermal processor (Sometimes, there's a '0' and sometimes a '1' behind. I think, that doesn't matter?) It seems, they are loaded. Thanks -- To unsubscribe, e-mail: opensuse+unsubscribe@opensuse.org For additional commands, e-mail: opensuse+help@opensuse.org
In
Quoting "Boyd Stephen Smith Jr."
: In Monday 15 June 2009, you wrote:
Booting from the system, this gives:
Couldn't find device with uuid ' lLWxB7-QmLl-FMXX-KQRW-EDH2-EHLL-o7D21b' Couldn't find device with uuid ' lLWxB7-QmLl-FMXX-KQRW-EDH2-EHLL-o7D21b' Couldn't find device with uuid ' lLWxB7-QmLl-FMXX-KQRW-EDH2-EHLL-o7D21b' PV /dev/sda4 with UUID ecQz2l-Yl5S-s1AR-delt-5zXF-Wnjs-P2y2dH VG system lvm2 [113.03 GB / 0 free] PV /dev/sdb1 with UUID 5P6jGx-ePjX-43Az-wyLL-wuF5-R4Vt-vP9Uff VG system lvm2 [465.76 / 0 free] PV unknown device with UUID lLWxB7-QmLl-FMXX-KQRW-EDHZ-EHLL-o7DZ1b VG system lvm2 [465.76 GB / 224.54 GB free] Total: 3 [1.02 TB] / in use: 3 [1.02 TB] / in no VG: 0 [0 ]
When I boot from the GParted live-CD:
PV /dev/sdb4 with UUID ecQz2l-Yl5S-s1AR-delt-5zXF-Wnjs-P2y2dH VG system lvm2 [113.03 GB / 0 free] PV /dev/sdc1 with UUID 5P6jGx-ePjX-43Az-wyLL-wuF5-R4Vt-vP9Uff VG system lvm2 [465.76 GB / 0 free] PV /dev/sda with UUID lLWxB7-QmLl-FMXX-KQRW-EDHZ-EHLL-o7DZ1b VG system lvm2 [465.76 GB / 224.54 GB free] Total: 3 [1.02 TB] /in use: 3 [1.02 TB] / in no VG: 0 [0 ]
No error at all. How does the GParted Live-CD do that?
Are the LVM versions different?
This is something different. As the GParted Live-CD shows, the device has the correct UUID. This means that, for whatever reason, LVM is not scanning the correct device(s) to find a UUID.
When you boot the system normally, which device is the drive that LVM can't find?
sdc
You may simply be missing the correct kernel module for the disk controller.
How could I verify that?
If it is showing up as a device, you've got the right modules loaded.
If your device is rejected (name
How could I find its name? Is it 'sdc'?
Filtering is based on absolute pathname. So "/dev/sdc", which should be accepted and not rejected your rules.
Disk /dev/sdc: 500.1 GB, 500107862016 bytes 255 heads, 63 sectors/track, 60801 cylinders Units = cylinders of 16065 * 512 = 8225280 bytes Disk identifier: 0x00000000
Disk /dev/sdc doesn't contain a valid partition table
That's normal for a whole disk PV.
Disk /dev/dm-0: 858.9 GB, 858993459200 bytes Disk /dev/dm-1: 21.4 GB, 21474836480 bytes
Hrm, are either one of these claiming the /dev/sdc device as a component device? LVM might skip devices that are part of an active dm device. You might try issuing the command: pvscan -vv as root in the booted system. It might provide some information as to why LVM is not treating /dev/sdc as a PV. -- Boyd Stephen Smith Jr. ,= ,-_-. =. bss@iguanasuicide.net ((_/)o o(\_)) ICQ: 514984 YM/AIM: DaTwinkDaddy `-'(. .)`-' http://iguanasuicide.net/ \_/
Hi, Am 16.06.2009 um 20:31 schrieb Boyd Stephen Smith Jr.:
In
, Philip Mötteli wrote: Quoting "Boyd Stephen Smith Jr."
: In Monday 15 June 2009, you wrote:
Booting from the system, this gives:
Couldn't find device with uuid ' lLWxB7-QmLl-FMXX-KQRW-EDH2-EHLL-o7D21b' Couldn't find device with uuid ' lLWxB7-QmLl-FMXX-KQRW-EDH2-EHLL-o7D21b' Couldn't find device with uuid ' lLWxB7-QmLl-FMXX-KQRW-EDH2-EHLL-o7D21b' PV /dev/sda4 with UUID ecQz2l-Yl5S-s1AR-delt-5zXF-Wnjs-P2y2dH VG system lvm2 [113.03 GB / 0 free] PV /dev/sdb1 with UUID 5P6jGx-ePjX-43Az-wyLL-wuF5-R4Vt-vP9Uff VG system lvm2 [465.76 / 0 free] PV unknown device with UUID lLWxB7-QmLl-FMXX-KQRW-EDHZ-EHLL- o7DZ1b VG system lvm2 [465.76 GB / 224.54 GB free] Total: 3 [1.02 TB] / in use: 3 [1.02 TB] / in no VG: 0 [0 ]
When I boot from the GParted live-CD:
PV /dev/sdb4 with UUID ecQz2l-Yl5S-s1AR-delt-5zXF-Wnjs-P2y2dH VG system lvm2 [113.03 GB / 0 free] PV /dev/sdc1 with UUID 5P6jGx-ePjX-43Az-wyLL-wuF5-R4Vt-vP9Uff VG system lvm2 [465.76 GB / 0 free] PV /dev/sda with UUID lLWxB7-QmLl-FMXX-KQRW-EDHZ-EHLL-o7DZ1b VG system lvm2 [465.76 GB / 224.54 GB free] Total: 3 [1.02 TB] /in use: 3 [1.02 TB] / in no VG: 0 [0 ]
No error at all. How does the GParted Live-CD do that?
Are the LVM versions different?
Original system: LVM version: 2.02.25 (2007-04 27) Library version: 1.02.19 (2007-04-27) Driver version: 4.11.0 RescueCD: LVM version: 2.02.45 (2009-03-03) Library version: 1.02.31 (2009-03-03) Driver version: 4.14.0 They are in fact different. The problem is, they are running software that only runs until OpenSuSE 10.3.
This is something different. As the GParted Live-CD shows, the device has the correct UUID. This means that, for whatever reason, LVM is not scanning the correct device(s) to find a UUID.
When you boot the system normally, which device is the drive that LVM can't find?
sdc
You may simply be missing the correct kernel module for the disk controller.
How could I verify that?
If it is showing up as a device, you've got the right modules loaded.
No, it is not showing up in '/dev'. So there's probably a driver/kernel module problem. Though the scsi drivers seem to be loaded. Have you seen the other posts with Per Jessen?
Disk /dev/dm-0: 858.9 GB, 858993459200 bytes Disk /dev/dm-1: 21.4 GB, 21474836480 bytes
Hrm, are either one of these claiming the /dev/sdc device as a component device? LVM might skip devices that are part of an active dm device.
You might try issuing the command: pvscan -vv as root in the booted system. It might provide some information as to why LVM is not treating /dev/sdc as a PV.
Setting global/locking_type to 1 File-based locking selected. Setting global/locking_dir to /var/lock/lvm Wiping cache of LVM-capable devices Wiping internal VG cache Walking through all physical volumes /dev/sda: size is 312581808 sectors /dev/sda1: size is 41961717 sectors /dev/sda1: size is 41961717 sectors /dev/sda1: No label detected /dev/sda2: size is 31471335 sectors /dev/sda2: size is 31471335 sectors /dev/sda2: No label detected /dev/sda3: size is 2104515 sectors /dev/sda3: size is 2104515 sectors /dev/sda3: No label detected /dev/sda4: size is 237039075 sectors /dev/sda4: size is 237039075 sectors /dev/sda4: lvm2 label detected /dev/sdb: size is 976773168 sectors /dev/sdb1: size is 976768002 sectors /dev/sdb1: size is 976768002 sectors /dev/sdb1: lvm2 label detected /dev/sda4: lvm2 label detected /dev/sdb1: lvm2 label detected /dev/sda4: lvm2 label detected /dev/sdb1: lvm2 label detected /dev/sda1: No label detected /dev/sda2: No label detected /dev/sda3: No label detected /dev/sda4: lvm2 label detected /dev/sdb1: lvm2 label detected Couldn't find device with uuid 'lLWxB7-QmLl-FMXX-KQRW-EDHZ-EHLL- o7DZ1b'. /dev/sda4: lvm2 label detected /dev/sdb1: lvm2 label detected /dev/sda1: No label detected /dev/sda2: No label detected /dev/sda3: No label detected /dev/sda4: lvm2 label detected /dev/sdb1: lvm2 label detected Couldn't find device with uuid 'lLWxB7-QmLl-FMXX-KQRW-EDHZ-EHLL- o7DZ1b'. /dev/sda1: No label detected /dev/sda2: No label detected /dev/sda3: No label detected /dev/sda4: lvm2 label detected /dev/sdb1: lvm2 label detected /dev/sda4: lvm2 label detected /dev/sdb1: lvm2 label detected /dev/sda4: lvm2 label detected /dev/sdb1: lvm2 label detected /dev/sda1: No label detected /dev/sda2: No label detected /dev/sda3: No label detected /dev/sda4: lvm2 label detected /dev/sdb1: lvm2 label detected Couldn't find device with uuid 'lLWxB7-QmLl-FMXX-KQRW-EDHZ-EHLL- o7DZ1b'. /dev/sda4: lvm2 label detected /dev/sdb1: lvm2 label detected /dev/sda1: No label detected /dev/sda2: No label detected /dev/sda3: No label detected /dev/sda4: lvm2 label detected /dev/sdb1: lvm2 label detected Couldn't find device with uuid 'lLWxB7-QmLl-FMXX-KQRW-EDHZ-EHLL- o7DZ1b'. PV /dev/sda4 VG system lvm2 [113.03 GB / 0 free] PV /dev/sdb1 VG system lvm2 [465.76 GB / 0 free] PV unknown device VG system lvm2 [465.76 GB / 224.54 GB free] Total: 3 [1.02 TB] / in use: 3 [1.02 TB] / in no VG: 0 [0 ] Thanks! -- To unsubscribe, e-mail: opensuse+unsubscribe@opensuse.org For additional commands, e-mail: opensuse+help@opensuse.org
In <85F9AC9E-0DB0-4B13-96F1-091BF80E617C@econophone.ch>, Philip Mötteli wrote:
Am 16.06.2009 um 20:31 schrieb Boyd Stephen Smith Jr.:
In
, Philip Mötteli wrote: Quoting "Boyd Stephen Smith Jr."
: In Monday 15 June 2009, you wrote: When you boot the system normally, which device is the drive that LVM can't find?
sdc
If it is showing up as a device, you've got the right modules loaded.
No, it is not showing up in '/dev'.
Then you don't know it will show up as sdc. It might, but those names are assigned dynamically, so don't assume it will be sdc. I inferred from the fact that you have a name for the device that it was showing up in /dev. How was fdisk able to see /dev/sdc? Or was that /dev/sdc not actually the correct device?
Have you seen the other posts with Per Jessen?
Yes, but I assumed it was a red herring. Now I've got to re-read them. -- Boyd Stephen Smith Jr. ,= ,-_-. =. bss@iguanasuicide.net ((_/)o o(\_)) ICQ: 514984 YM/AIM: DaTwinkDaddy `-'(. .)`-' http://iguanasuicide.net/ \_/
Hi, Now, that I have a backup, I tried the 'pvcreate' command. As you probably would have expected, it refused to do anything, because it couldn't find sdc. Am 17.06.2009 um 18:04 schrieb Boyd Stephen Smith Jr.:
In <85F9AC9E-0DB0-4B13-96F1-091BF80E617C@econophone.ch>, Philip Mötteli wrote:
Am 16.06.2009 um 20:31 schrieb Boyd Stephen Smith Jr.:
In
, Philip Mötteli wrote: Quoting "Boyd Stephen Smith Jr."
: In Monday 15 June 2009, you wrote: When you boot the system normally, which device is the drive that LVM can't find?
sdc
If it is showing up as a device, you've got the right modules loaded.
No, it is not showing up in '/dev'.
Then you don't know it will show up as sdc. It might, but those names are assigned dynamically, so don't assume it will be sdc.
Here are the devices in '/dev': blog console device-mapper disk fb0 fd fd0 fd0u[360-1920] full hpet input kmem kmsg mapper mem null oldmem port psaux ptmx pts pty[q0-sf] random rtc sda sda1 sda2 sda3 sda4 sdb sdb1 snapshot stderr stdin stdout tty tty[0-sf] urandom usbdev[1.1_ep00-5.1_ep81] vcs vcsa vcsa1 zero All the other disks and partitions are showing up as usual. Also, these disks should use the same drivers to be accessible. Why doesn't it just not find this single disk? Perhaps, because it's a whole disk lvm?
How was fdisk able to see /dev/sdc?
Well, there's no fdisk in the fall-back shell of the system. So I call fdisk in the RescueCD. I don't know, how fdisk works, but at least the device nodes are already there.
Or was that /dev/sdc not actually the correct device?
GParted shows all the devices, including sdc. All show up in '/proc/ partitions' (so have a device node), except 'sdc'. It can't be a broken hard-drive, because the hard-drive is working under the RescueCD. I was even able to make a backup of all the data from there. So it must be something with the configuration on the system disk. What is used, to scan the SCSI bus during boot up? Thanks -- To unsubscribe, e-mail: opensuse+unsubscribe@opensuse.org For additional commands, e-mail: opensuse+help@opensuse.org
In <5EC9CC69-3A07-4CE5-A2C0-776B6BDF52E3@econophone.ch>, Philip Mötteli wrote:
Now, that I have a backup, I tried the 'pvcreate' command. As you probably would have expected, it refused to do anything, because it couldn't find sdc.
If the device shows up when booting from the GParted LiveCD and does show up when the system boots normally, you are most likely missing a kernel module. Boot into each and get a list of loaded modules. (Either lsmod or cat /proc/modules.) You should have at least one loaded during the GParted LiveCD boot that is not loaded in the normal boot. Load each of those in turn, looking for the one that makes a new block device appear and the PV be detected again. -- Boyd Stephen Smith Jr. ,= ,-_-. =. bss@iguanasuicide.net ((_/)o o(\_)) ICQ: 514984 YM/AIM: DaTwinkDaddy `-'(. .)`-' http://iguanasuicide.net/ \_/
Hi, Am 18.06.2009 um 17:15 schrieb Boyd Stephen Smith Jr.:
In <5EC9CC69-3A07-4CE5-A2C0-776B6BDF52E3@econophone.ch>, Philip Mötteli wrote:
Now, that I have a backup, I tried the 'pvcreate' command. As you probably would have expected, it refused to do anything, because it couldn't find sdc.
If the device shows up when booting from the GParted LiveCD and does show up when the system boots normally, you are most likely missing a kernel module.
Boot into each and get a list of loaded modules. (Either lsmod or cat /proc/modules.)
RescueCD: ipv6 207432 16 video 13472 0 output 1560 1 video mptspi 13396 0 mptscsih 22700 1 mptspi mptbase 63560 2 mptspi,mptscsih e1000 92864 0 ppdev 5592 0 parport_pc 16644 0 e752x_edac 9464 0 edac_core 22396 1 e752x_edac raid10 17380 0 raid456 113936 0 async_xor 2108 1 raid456 async_memcpy 1224 1 raid456 async_tx 1972 3 raid456,async_xor,async_memcpy xor 13012 2 raid456,async_xor raid1 17292 0 raid0 5836 0 sbp2 16608 0 ohci1394 23444 0 ieee1394 51760 2 sbp2,ohci1394 sl811_hcd 8180 0 usb_storage 66576 0 virtio_blk 4064 0 osst 39992 0 sym53c8xx 58576 0 atp870u 23496 0 hptiop 9624 0 ses 4820 0 enclosure 4264 1 ses aic79xx 97880 0 aic7xxx 88992 0 aic94xx 58660 0 ppa 8056 0 raid_class 3012 0 sym53c500_cs 5780 0 qlogic_cs 4056 0 qlogicfas408 3952 1 qlogic_cs aacraid 51640 0 imm 7928 0 parport 22940 4 ppdev,parport_pc,ppa,imm mvsas 17428 0 libsas 32772 2 aic94xx,mvsas 3w_xxxx 19928 0 initio 13564 0 gdth 71944 0 arcmsr 17200 0 stex 9032 0 tmscsim 15776 0 dc395x 25848 0 iscsi_tcp 6968 0 libiscsi_tcp 7780 1 iscsi_tcp 3w_9xxx 24812 0 a100u2w 7392 0 BusLogic 57900 0 libsrp 4708 0 libiscsi 19928 2 iscsi_tcp,libiscsi_tcp st 27656 0 ch 9812 0 scsi_transport_srp 3632 0 scsi_transport_spi 14072 4 mptspi,sym53c8xx,aic79xx,aic7xxx qla4xxx 44136 0 scsi_transport_iscsi 16892 3 iscsi_tcp,libiscsi,qla4xxx qla2xxx 178236 0 lpfc 221796 0 scsi_transport_fc 25188 2 qla2xxx,lpfc scsi_tgt 7920 3 libsrp,scsi_transport_srp,scsi_transport_fc scsi_transport_sas 15168 3 aic94xx,mvsas,libsas qla1280 111560 0 megaraid_sas 24776 0 megaraid 35480 0 pdc_adma 5228 0 sata_vsc 4136 0 sata_via 6684 0 sata_uli 3064 0 sata_sx4 8144 0 sata_svw 3724 0 sata_sis 4000 0 sata_sil 6876 0 sata_sil24 10440 0 sata_qstor 5264 0 sata_promise 7832 0 sata_nv 16608 0 sata_mv 17564 1 sata_inic162x 7052 0 pata_via 6416 0 pata_triflex 2716 0 pata_sl82c105 3172 0 pata_sis 7104 1 sata_sis pata_sil680 4492 0 pata_serverworks 4608 0 pata_sch 2496 0 pata_pdc202xx_old 3536 0 pata_pdc2027x 5484 0 pata_pcmcia 7740 0 pata_opti 2552 0 pata_optidma 3876 0 pata_oldpiix 2812 0 pata_ns87415 2980 0 pata_ns87410 2672 0 pata_ninja32 2880 0 pata_netcell 2068 0 pata_mpiix 2952 0 pata_marvell 2400 0 pata_jmicron 2136 0 pata_it821x 7520 0 pata_it8213 3252 0 pata_hpt3x3 3176 0 pata_hpt3x2n 4308 0 pata_hpt37x 8632 0 pata_hpt366 4468 0 pata_efar 3188 0 pata_cypress 2560 0 pata_cs5530 4156 0 pata_cs5520 4072 0 pata_cmd64x 4368 0 pata_cmd640 3132 0 pata_atiixp 3188 0 pata_artop 3872 0 pata_amd 7364 0 pata_ali 7208 0 pata_acpi 2740 0 scsi_wait_scan 668 0 ata_generic 3144 0 Boot-system: linear sd_mod ehci_hcd uhci_hcd usbcore dm_snapshot edd dm_mod ext3 mbcache jbd fan mptspi mptscsih mptbase scsi_transport_spi ata_piix libata scsi_mod thermal processor
You should have at least one loaded during the GParted LiveCD boot that is not loaded in the normal boot.
Many, many: 125 in the RescueCD and 21 in the system boot. In addition to that, most of the modules, loaded by the system boot, are not loaded by the RescueCD: linear sd_mod ehci_hcd uhci_hcd usbcore dm_snapshot edd dm_mod ext3 mbcache jbd fan ata_piix libata scsi_mod thermal processor
Load each of those in turn,
I did a 'modprobe -l' and loaded all those available, that haven't been loaded so far: ohci-hcd usbhid
looking for the one that makes a new block device appear
No sdc appears in '/dev'. Is there a way to force this? Thanks! -- To unsubscribe, e-mail: opensuse+unsubscribe@opensuse.org For additional commands, e-mail: opensuse+help@opensuse.org
In <7DEBCCBF-A797-45BD-B953-CCD82E03C2CE@econophone.ch>, Philip Mötteli wrote:
Am 18.06.2009 um 17:15 schrieb Boyd Stephen Smith Jr.:
If the device shows up when booting from the GParted LiveCD and does show up when the system boots normally, you are most likely missing a kernel module.
Boot into each and get a list of loaded modules. (Either lsmod or cat /proc/modules.)
RescueCD:
[125 modules]
A lot of those are disk-oriented. Anything with "ide", "ata", "sas", or "scsi" in the name should be the first ones you look for. You might also need ones with "raid" in the name.
Boot-system:
[21 modules]
This is definitely not enough. You are missing a *lot* of modules. 'lsmod | wc -l' on my laptop gives 128 modules. 'modprobe -l | wc -l' on that same laptop shows 2085 modules available.
You should have at least one loaded during the GParted LiveCD boot that is not loaded in the normal boot.
Many, many: 125 in the RescueCD and 21 in the system boot. In addition to that, most of the modules, loaded by the system boot, are not loaded by the RescueCD:
Don't worry about modules that are loaded by the real system but not by the RescueCD. You aren't missing things in the RescueCD, so what it is not loading doesn't matter.
I did a 'modprobe -l' and loaded all those available, that haven't been loaded so far:
modprobe -l does actually load any modules.
ohci-hcd usbhid
Neither one of these will make ide/pata/sata/sas/scsi devices appear.
No sdc appears in '/dev'. Is there a way to force this?
How do you propose to "force" the kernel to talk to a device it cannot detect? You can create /dev/sdc with mknod, but it won't *work* unless the kernel detects the device. Once the kernel detects the device, it will communicate with udev, and udev will perform the mknod automatically. Make sure you have all the modules you need installed. On my laptop, they are coming from 3 packages: kernel-default, kernel-default-base, and kernel- default-extra. Once you have that done, you could simply write a shell loop to load each module looking from one that makes /dev/sdc appear. Something like this (as root) might work: modprobe -l | grep -E '(ide|ata|sas|scsi|raid)' | while read ko; do kof=$(basename "$ko") && module=${kof%%.ko} && modprobe "$module" && if [ -r /dev/sdc ]; then echo "Winner! Module is: $module" break else modprobe -r "$module" || echo "Sorry, couldn't remove module $module." fi done -- Boyd Stephen Smith Jr. ,= ,-_-. =. bss@iguanasuicide.net ((_/)o o(\_)) ICQ: 514984 YM/AIM: DaTwinkDaddy `-'(. .)`-' http://iguanasuicide.net/ \_/
Hi, Am 19.06.2009 um 18:46 schrieb Boyd Stephen Smith Jr.:
In <7DEBCCBF-A797-45BD-B953-CCD82E03C2CE@econophone.ch>, Philip Mötteli wrote:
Am 18.06.2009 um 17:15 schrieb Boyd Stephen Smith Jr.:
I did a 'modprobe -l' and loaded all those available, that haven't been loaded so far:
modprobe -l does actually load any modules.
I did load them with just 'modprobe' and controlled, if '/proc/ modules' is showing them.
ohci-hcd usbhid
Neither one of these will make ide/pata/sata/sas/scsi devices appear.
These are the only ones that 'modprobe -l' lists as being available but are not shown as loaded in '/proc/modules'.
No sdc appears in '/dev'.
Make sure you have all the modules you need installed.
It has worked for a long time now. So the needed modules have surely been available. At least, until these problems started. Apparently my client did some very dirty power offs. Is it possible, that these destroyed some entire directories on the disk? (This is a SuSE 9.3 system.)
On my laptop, they are coming from 3 packages: kernel-default, kernel-default-base, and kernel- default-extra.
Here I have only the following directories (listed by 'modprobe -l'): .../kernel/drivers .../kernel/drivers/ata .../kernel/drivers/usb/core .../kernel/drivers/usb/host .../kernel/drivers/firmware .../kernel/drivers/md .../kernel/drivers/message .../kernel/drivers/hid .../kernel/drivers/input .../kernel/drivers/acpi .../kernel/drivers/scsi .../kernel/fs .../kernel/fs/jbd .../kernel/fs/ext3 If there are some directories missing, I should perhaps install a 9.3 SuSE somewhere and just copy over these directories to the broken system?
you could simply write a shell loop to load each module looking from one that makes /dev/sdc appear. Something like this (as root) might work:
modprobe -l | grep -E '(ide|ata|sas|scsi|raid)' | while read ko; do kof=$(basename "$ko") && module=${kof%%.ko} && modprobe "$module" && if [ -r /dev/sdc ]; then echo "Winner! Module is: $module" break else modprobe -r "$module" || echo "Sorry, couldn't remove module $module." fi done
Thanks for creating this script. I wouldn't be able to create a shell script quickly like that. But unfortunately, not 'grep', nor 'basename' are available in the fall back shell. So I just did it by hand. After all, there are only 21 modules and actually none of the not loaded ones belongs to the group (ide|ata|sas| scsi|raid). So there was nothing to do. Thanks -- To unsubscribe, e-mail: opensuse+unsubscribe@opensuse.org For additional commands, e-mail: opensuse+help@opensuse.org
In
Am 19.06.2009 um 18:46 schrieb Boyd Stephen Smith Jr.:
In <7DEBCCBF-A797-45BD-B953-CCD82E03C2CE@econophone.ch>, Philip Mötteli wrote:
No sdc appears in '/dev'.
Make sure you have all the modules you need installed.
It has worked for a long time now. So the needed modules have surely been available. At least, until these problems started. Apparently my client did some very dirty power offs. Is it possible, that these destroyed some entire directories on the disk? (This is a SuSE 9.3 system.)
Are you sure you have the real root file system mounted? This:
.../kernel/drivers .../kernel/drivers/ata .../kernel/drivers/usb/core .../kernel/drivers/usb/host .../kernel/drivers/firmware .../kernel/drivers/md .../kernel/drivers/message .../kernel/drivers/hid .../kernel/drivers/input .../kernel/drivers/acpi .../kernel/drivers/scsi .../kernel/fs .../kernel/fs/jbd .../kernel/fs/ext3
looks like only the modules that were put into the 'initramfs' created by 'mkinitrd'. It will only have the modules that mkinitrd "thinks" are required to boot -- it can get that wrong. You can still get access to file systems on incomplete volume groups as long as all the LEs in allocated to the LV are located on PVs that LVM can find. Use partial mode like so: vgscan -P; vgchange -Pay Basically, you'll need to add "-P" to every LVM command to work in partial mode.
If there are some directories missing, I should perhaps install a 9.3 SuSE somewhere and just copy over these directories to the broken system?
I'm not familiar with SUSE 9.3 (before my time), but I believe rpm is capable of verifying that all the files in a certain package are correctly installed. If partial mode doesn't work for you and since the GParted RescueCD will boot and find all your filesystems, you could boot from that, then chroot into your real system and verify stuff using all your expected SUSE 9.3 tools and your existing RPM database. -- Boyd Stephen Smith Jr. ,= ,-_-. =. bss@iguanasuicide.net ((_/)o o(\_)) ICQ: 514984 YM/AIM: DaTwinkDaddy `-'(. .)`-' http://iguanasuicide.net/ \_/
In <5EC9CC69-3A07-4CE5-A2C0-776B6BDF52E3@econophone.ch>, Philip Mötteli wrote:
Why doesn't it just not find this single disk?
Most likely a missing kernel module.
Perhaps, because it's a whole disk lvm?
Absolutely not.
Am 17.06.2009 um 18:04 schrieb Boyd Stephen Smith Jr.:
How was fdisk able to see /dev/sdc?
Well, there's no fdisk in the fall-back shell of the system. So I call fdisk in the RescueCD.
That would be good to mention next time. The context of a command is often as important as the command itself.
It can't be a broken hard-drive.
I am not suggesting that at all.
What is used, to scan the SCSI bus during boot up?
The kernel does this internally. There is no userland tool for this, AFAIK. -- Boyd Stephen Smith Jr. ,= ,-_-. =. bss@iguanasuicide.net ((_/)o o(\_)) ICQ: 514984 YM/AIM: DaTwinkDaddy `-'(. .)`-' http://iguanasuicide.net/ \_/
participants (3)
-
Boyd Stephen Smith Jr.
-
Per Jessen
-
Philip Mötteli