Well. I have one quite hard googled issue I cannot find the answer to. The issue I am experiencing from freshinstalling the new openSuse 11.4 (from a previous openSuse 11.3) is:
During installation I recognized that I have been using one of my a bit slower disks for openSuse 11.3 (yes very foolish) but that was the case. However in this installation I was to move the mountpoint from my / and /home from the slow disk to my faster one. The faster disk is sda, the slower disk that previously was / and /home was sdc
But I am not able to mount sdc on/in my new installation. It is mainly complaining about it cannot interpret the filesystem on that disk (that was/is ext4). In Expert partitioner I can see all disks, also the two others installed in the computer (which are mountable).
The other strange thing which might be a clue to someone. Is that All those three “storage disks” are 500 gigabytes. All disks are showing “size” correct size Both the other disks that where not used as / and /home in my previous installation are showing. But the sdc1 partition are showing 445 gigabytes regarding size. It seems that it has “forgotten” a bit of the partition? Remember - this partition was used as a / (started from beginning of the disc). I have got the feeling this has something to do with my problem. But I don’t really know how to solve it, and would of course be really happy to get some help here.
Print out from Partition manager - as you can see it says “nothing” on “File System”
.
Device:
• Device: /dev/sdc1
• Size: 445.76 GB
• Encrypted: No
• Device Path: pci-0000:00:1f.2-scsi-1:0:1:0-part1
• Device ID 1: ata-WDC_WD5001AALS-00L3B2_WD-WCASY5896927-part1
Device ID 2: scsi-SATA_WDC_WD5001AALS-_WD-WCASY5896927-part1
Device ID 3: wwn-0x50014ee202f40289-part1
• FS Id: 0x83 Linux native
File System:
• File System:
• Mount Point:
• Label:
I would run GParted and use it to “CHECK” the partitions that will not mount. GParted does not seem to be in openSUSE 11.4 just yet, but a bootable image can be found here:
I normally boot from the iso boot disk, find the partition and then elect to check it to fix any issues with it. If successful, it should be mountable in openSUSE 11.4.
I will try that tip for tomorrow. I have however done som (quite unfair) tests to burn the image that you are linking to, currently without success. Brasero is keeping “checking image checksum”-stage. Never seen this before - does anyone this issue might relate to? This is quite a side step in relation to the current topic but i don’t really know what i migt do to make this error to occur. (right-click at the downloaded image (done twice to somewhat ensure no checksum error - then click on brasero and then, just click “burn”).
Well this is of course saturday evening at sweden (where I am from) and I might just have to much beer in my body to even see the actual issue here rotfl!. (I do apoligise to my bad english at this present time.)
But if someone might have experienced this issue before on burning ISOs in openSuse I would be glad to hear your story. I am (FYI) a switcher from Ubuntu, since ubuntu-user since like 2-3 years and I felt the OpenSuse 11.3 Gnome-version felt solid, don’t really like the canonical “move-away-from-gnome-standard-point-of-view”). I would imagine I have hit the most “early-adoptor-errors”. Sorry for this newbie issue, late saturday evening.
This is not uncommon that the device names (sda, sdb, etc) you see during a Linux setup are not the same after you reboot and later vary from one start to another. This is because of the ‘asynchronous way’ udev loads modules and creates dev special files.
From ArchLinux Wiki:
udev loads kernel modules by utilizing coding parallelism to provide a potential performance advantage versus loading these modules serially. The modules are therefore loaded asynchronously. The inherent disadvantage of this method is that udev does not always load modules in the same order on each boot. If the machine has multiple block devices, this may manifest itself in the form of device nodes changing designations randomly. For example, if the machine has two hard drives, /dev/sda may randomly become /dev/sdb. See below for more info on this.
This can be very confusing. Never ever trust device names. Mount by labels or UUIDS (an fstab option during setup). Avoid mixing SATA and IDE hard disks.
Thanks for that point! I am aware of the “issue” that it is several ways to identfy ethiter a partition table or a physical hardrive in linux, and that might be my main issue here aswell. The only way forward if to find the actuall tool to interpret what information is availible on my sbc drive. As I said before the sbc drive was earlier used as a boot drive for my openSuse 11.3 and i do think that the boot sectors etc, are making the current partition program to go crazy to not overlook the current situation to well actually… tha actual situation really is that /sdc really contain TWO partitions (root partition, /, and the /home partition) and i do think the current partition application is being confused about the boot sectors used by the root partition etc. But I don’t know how to get around this particular issue) The only thing I am really interessted in to recieve is that old /home partition.
Rembember, I do Very much appriciate the help I am getting from this forum in solving the issue. Thanks a lot for the effort puttin’ down to help me in this.
edit:
When reading the previous post again I do understand I realy need to get on hold of the UUID’s of my previous partitions, right? But how do I do that? (Given the point “Mount by labels or UUIDS”)
Best Regards
Loofy
Try the halinfo script (latest version is here: Displaying partitions infos from hal daemon) with the syntax halinfo -uV or halinfo -hV to query either the hal daemons or udev, but both outputs should match. That will display everything, including device names, UUIDs and disk IDs and hopefully give you a big picture about your partitions.
It means “Extended partition”, a container for logical partitions. This partition doesn’t contain data and is not mountable. What happened here? If only the partition ID has been changed (but why? how?), you might give this partition the ID 0x83 (this is an hexadecimal value) and see if you can mount it … but something is obviously not right here. And what about the unallocated space in this partition? Was your /home partition there maybe? Someone mentionned a disk repare program on a knoppix live CD a week ago or so. He was able to get his partition table black. The system was messed up but it might be enough to get back some documents if that’s what you’re trying to do.
OK. I read the question (afterwards as usual lol!). It looks like the installation tried to resize/reorganize things on sdc and failed. Maybe you should check that disk for physical defects before further speculations. It the disk has bad sectors (which occures sooner or later), writing on it produces logical errors and partitioning/repartitioning might end up in a mess.
Actually during the installation of openSuse 11.4 I left the sbc drive “alone”. So I am not sure but. In the past I have read about “MBR records” etc that GRUB is creating etc in order to somewhat localize the kernels. And that this sometimes can be an issue when for instance installing a dualboot configuration with windows etc.
In theory, I might/should be able to boot up using sbc drive (in openSuse 11.3) If I only could make the bootloader to interpret what actually is on the disk? I have never been working a lot with bootmanager and GRUB since this has never been an issue to me in the past. Is there someway I could use the “bootloader” application in YAST to make it interpret sbc as another system to boot into somehow? There is several options when I choose to add a new entry in the menu.
Regarding physical errors to this disk - at least the SMART test within the Partition Manager states the disk to be healthy. Is there another, better way to check the physical status of sbc?
> In theory, I might/should be able to boot up using sbc drive (in
> openSuse 11.3) If I only could make the bootloader to interpret what
> actually is on the disk?
You can’t do anything with sdc until it is partitioned properly. You could
try “gpart” to guess what should be there, if it is not that extended
partition.
–
Cheers / Saludos,
Carlos E. R.
(from 11.2 x86_64 “Emerald” at Telcontar)
Yes. Don’t tink in term of disks (sda, sdb, etc) but in term of partitions (sdaX, sdbX, etc). In Linux - as in any other OS - you never use the disk directly. Even if the disk is not partitionated - as sdb and sdd in your example - it still has one partition, which is called sdb1 and sdd1 in your case.