Yep, that did the trick. Now, at least I’m back i my desktop. 
That’s right.
The line in /etc/fstab now reads
/dev/disk/by-id/ata-Crucial_CT256MX100SSD1_14260C72F4A5-part4 / ext4 errors=continue,relatime,discard 0 1
Hm, the fsck found no error apparently.
Looks good as well.
I forgot one thing.
The / file system is mounted by the initrd already (you can’t read /etc/fstab before / is mounted obviously 
), but the initrd will still contain your old fstab.
Run “sudo /sbin/mkinitrd” to re-create the initrd with the new file.
Still strange that it is ro, as apparently there doesn’t seem to be an error that would cause this.
If the above doesn’t help, try to add “rw” to the mount options in fstab as well (don’t forget to run mkinitrd). It really should be mounted rw then…
And please post your kernel command line:
cat /proc/cmdline
There are some kernel options that would force the root filesystem to be mounted read-only, like “ro” or “rootflags”
Aah, that may explain it! I totally forgot about that part. Ran sudo /sbin/mkinitrd. But it didn’t help. A few things changed in the output I saw at boot, but I saw first an error message that firewall couldn’t be initiated, then one about postfix (there may have been others, I don’t know, it goes too fast to read it), and I ended up in the cli.
“mount” confirmed that / was mounted as ro AGAIN!
So I added "rw2 to /etc/fstab.
/dev/disk/by-id/ata-Crucial_CT256MX100SSD1_14260C72F4A5-part4 / ext4 rw,errors=continue,relatime,discard 0 1
There were some possible issues in the output of /sbin/mkinitrd. I’ve posted it here just in case. http://paste.opensuse.org/72668321
And / is still mounted ro.
cat /proc/cmdline
BOOT_IMAGE=/boot/vmlinuz-3.17.1-2.g5c4d099-desktop root=UUID=6969a6ea-7b80-443a-a708-2cb2a9d4846b resume=/dev/disk/by-id/ata-WDC_WD10JPVT-75A1YT0_WX61A9201917-part10 splash=silent quiet showopts acpi_osi= rd.break
“rd.break” was a temporary option from a session where I had started openSUSE with that option.
The only error I see is about “fcoe-uefi”. No idea what exactly that is, but I wouldn’t think this causes your / partition to be mounted read-only if missing.
cat /proc/cmdline BOOT_IMAGE=/boot/vmlinuz-3.17.1-2.g5c4d099-desktop root=UUID=6969a6ea-7b80-443a-a708-2cb2a9d4846b resume=/dev/disk/by-id/ata-WDC_WD10JPVT-75A1YT0_WX61A9201917-part10 splash=silent quiet showopts acpi_osi= rd.break
That “acpi_osi” parameter looks suspicious. The actual option is missing.
Try to remove it. Problems with acpi can cause all sorts of strange problems.
Didn’t make a difference.
The parameter was a deliberate addition. I have had problems with the GUI in the past that were either caused by faulty gfx drivers or Kernel mode setting specifically (I don’t recall). This was the solution recommended on the web that worked. But as I’m not having any problems right now, I may as well omit it permanently.
I wonder if we shouldn’t be looking for the problem somewhere within systemd? Any suggestions?
Thanks again.
Ok, but shouldn’t there be some value after the ‘=’?
Well, doesn’t matter now anyway.
I wonder if we shouldn’t be looking for the problem somewhere within systemd? Any suggestions?
Hm, the kernel does the mounting. So if there’s nothing in the kernel log (dmesg), I don’t think we’ll find something in systemd’s journal.
But well, try this:
sudo journalctl -b |grep sda
This should show all messages regarding /dev/sdaX…
We already tried “systemctl status /”, what about “sudo systemctl status /dev/sda4”?
I see that you are using kernel 3.17.1. Maybe it’s some kernel issue? Try an older one please, you should find them in “Advanced Options” in the boot menu.
Other than that I have no further idea at the moment, sorry.
Well, I just had another idea:
maybe there’s something wrong with the filesystem’s own parameters?
Please post the output of:
sudo /sbin/tune2fs -l /dev/sda4
For some reason, adding no value behind the “=” was exactly the point. Don’t ask me why. 
I don’t think it’s a kernel issue. I’ve had the problem with both kernels. Kernel 3.17 ist a remnant from 13.1 where I had to use the official kernel repository to deal with graphics driver issues. Among other things, it kept my machine from overheating and helped with powersaving.
sudo journalctl -b |grep sda
root's password:
Nov 11 08:32:18 ichthys kernel: sd 0:0:0:0: [sda] 500118192 512-byte logical blocks: (256 GB/238 GiB)
Nov 11 08:32:18 ichthys kernel: sd 0:0:0:0: [sda] 4096-byte physical blocks
Nov 11 08:32:18 ichthys kernel: sd 0:0:0:0: [sda] Write Protect is off
Nov 11 08:32:18 ichthys kernel: sd 0:0:0:0: [sda] Mode Sense: 00 3a 00 00
Nov 11 08:32:18 ichthys kernel: sd 0:0:0:0: [sda] Write cache: enabled, read cache: enabled, doesn't support DPO or FUA
Nov 11 08:32:18 ichthys kernel: sda: sda1 sda2 sda3 sda4 sda5 sda6 sda7 sda8
Nov 11 08:32:18 ichthys kernel: sd 0:0:0:0: [sda] Attached SCSI disk
Nov 11 08:32:21 ichthys kernel: EXT4-fs (sda4): mounted filesystem with ordered data mode. Opts: (null)
Nov 11 08:32:22 ichthys kernel: Adding 2186236k swap on /dev/sda5. Priority:-1 extents:1 across:2186236k SSFS
Nov 11 08:32:22 ichthys systemd-fsck[849]: /dev/sda3: clean, 272383/3227648 files, 7784287/12881920 blocks
Nov 11 08:32:22 ichthys kernel: FAT-fs (sda7): Volume was not properly unmounted. Some data may be corrupt. Please run fsck.
Nov 11 08:32:22 ichthys kernel: EXT4-fs (sda3): mounted filesystem with ordered data mode. Opts: discard
Nov 11 08:33:43 ichthys.site kernel: EXT4-fs (sda4): re-mounted. Opts: (null)
sudo systemctl status /dev/sda4
dev-sda4.device - Crucial_CT256MX1
Follow: unit currently follows state of sys-devices-pci0000:00-0000:00:1f.2-ata1-host0-target0:0:0-0:0:0:0-block-sda-sda4.device
Loaded: loaded
Active: active (plugged) since Di 2014-11-11 08:32:22 CST; 16min ago
Device: /sys/devices/pci0000:00/0000:00:1f.2/ata1/host0/target0:0:0/0:0:0:0/block/sda/sda4
sudo /sbin/tune2fs -l /dev/sda4
tune2fs 1.42.12 (29-Aug-2014)
Filesystem volume name: Linux
Last mounted on: /
Filesystem UUID: 6969a6ea-7b80-443a-a708-2cb2a9d4846b
Filesystem magic number: 0xEF53
Filesystem revision #: 1 (dynamic)
Filesystem features: has_journal ext_attr resize_inode dir_index filetype needs_recovery extent flex_bg sparse_super large_file huge_file uninit_bg dir_nlink extra_isize
Filesystem flags: signed_directory_hash
Default mount options: user_xattr acl
Filesystem state: clean
Errors behavior: Continue
Filesystem OS type: Linux
Inode count: 1782368
Block count: 7116800
Reserved block count: 355840
Free blocks: 1398735
Free inodes: 1315618
First block: 0
Block size: 4096
Fragment size: 4096
Reserved GDT blocks: 1022
Blocks per group: 32768
Fragments per group: 32768
Inodes per group: 8176
Inode blocks per group: 511
Flex block group size: 16
Filesystem created: Wed Jul 23 21:56:54 2014
Last mount time: Tue Nov 11 08:33:43 2014
Last write time: Tue Nov 11 08:32:18 2014
Mount count: 7
Maximum mount count: -1
Last checked: Mon Nov 10 07:37:29 2014
Check interval: 0 (<none>)
Lifetime writes: 373 GB
Reserved blocks uid: 0 (user root)
Reserved blocks gid: 0 (group root)
First inode: 11
Inode size: 256
Required extra isize: 28
Desired extra isize: 28
Journal inode: 8
First orphan inode: 1343231
Default directory hash: half_md4
Directory Hash Seed: 0dd1587d-e572-4a5f-b1a8-bf2a172b3e83
Journal backup: inode blocks
I appreciate all you help up to this point. I may just have to bite the bullet and reinstall.
Ok, I’m maybe grasping at straws now, but:
Maybe that problem with sda7 causes sda4 to be re-mounted read-only erroneously?
Please run “fsck -a /dev/sda7” to fix it.
sudo /sbin/tune2fs -l /dev/sda4 tune2fs 1.42.12 (29-Aug-2014) Filesystem volume name: Linux Last mounted on: / Filesystem UUID: 6969a6ea-7b80-443a-a708-2cb2a9d4846b Filesystem magic number: 0xEF53 Filesystem revision #: 1 (dynamic) Filesystem features: has_journal ext_attr resize_inode dir_index filetype needs_recovery extent flex_bg sparse_super large_file huge_file uninit_bg dir_nlink extra_isize Filesystem flags: signed_directory_hash Default mount options: user_xattr acl Filesystem state: clean Errors behavior: Continue Filesystem OS type: Linux Inode count: 1782368 Block count: 7116800 Reserved block count: 355840 Free blocks: 1398735 Free inodes: 1315618 First block: 0 Block size: 4096 Fragment size: 4096 Reserved GDT blocks: 1022 Blocks per group: 32768 Fragments per group: 32768 Inodes per group: 8176 Inode blocks per group: 511 Flex block group size: 16 Filesystem created: Wed Jul 23 21:56:54 2014 Last mount time: Tue Nov 11 08:33:43 2014 Last write time: Tue Nov 11 08:32:18 2014 Mount count: 7 Maximum mount count: -1 Last checked: Mon Nov 10 07:37:29 2014 Check interval: 0 (<none>) Lifetime writes: 373 GB Reserved blocks uid: 0 (user root) Reserved blocks gid: 0 (group root) First inode: 11 Inode size: 256 Required extra isize: 28 Desired extra isize: 28 Journal inode: 8 First orphan inode: 1343231 Default directory hash: half_md4 Directory Hash Seed: 0dd1587d-e572-4a5f-b1a8-bf2a172b3e83 Journal backup: inode blocks
Well, that confirms at least that the “errors=continue” mount option is respected, and that the filesystem is clean, i.e. has no errors.
So you could actually remove that option again…
It seems there was a little problem with sda7 that fsck fixed. Nothing has changed about / being mounted ro though…
https://www.dropbox.com/s/rptijuwsvfi7p8m/2014-11-12%2009.30.03.jpg?dl=0
The parse error in line 8 of /etc/fstab is not related to either sda4 or sda7 (and I’m not sure what that error is looking at line 8 is, in fact).
Maybe fsck chokes on that '\ ’ there?
I don’t know why fsck even parses the fstab, maybe to find out which filesystem is on the partition you want to check?
Anyway, just as a test, I would remove or comment out all unnecessary entries (i.e. everything except /, /home and swap) to see if it maybe works then.
Just to clarify, you’re talking about unnecessary lines in /etc/fstab? On other words, I should only have those partitions mounted that are absolutely essential?
The \ was my only guess, too, but I’ve never noticed problems with that partition being mounted. (I absolutely need to clean up that other drive with those old installations some time. )
When I just tested that in dolphin, I got an error message saying
Beim Zugriff auf „244,2 GiB Festplatte“ ist ein Fehler aufgetreten, die Meldung lautet: The requested operation has failed: Error mounting system-managed device /dev/sdb8: Command-line `mount "/home/alvanx/Altes\\"' exited with non-zero exit status 32: mount: /etc/fstab: parse error: ignore entry at line 8. mount: unknown filesystem type 'Home'
I will rename that mount point (i.e. remove the space), but I’m not hopeful this is going to change anything.
Yes.
The \ was my only guess, too, but I’ve never noticed problems with that partition being mounted.
Yes. fsck does seem to have a problem with it, but that shouldn’t matter.
OTOH, fsck is run on boot too, if it returns an error this might cause a partition to stay read-only I suppose.
Anyway, I’m not saying that the '' is the cause of your problems, I just want to eliminate possible reasons.
When I just tested that in dolphin, I got an error message saying
Beim Zugriff auf „244,2 GiB Festplatte“ ist ein Fehler aufgetreten, die Meldung lautet: The requested operation has failed: Error mounting system-managed device /dev/sdb8: Command-line `mount "/home/alvanx/Altes\\"' exited with non-zero exit status 32: mount: /etc/fstab: parse error: ignore entry at line 8. mount: unknown filesystem type 'Home'
So mount or udisks2 doesn’t like it either.
Btw, “man fstab” says this:
If the name of the mount point contains spaces these can be escaped as `\040’.
Well, you should have been saying that because it was spot on! Unbelievable! Removing a backslash from /etc/fstab that had never been a problem fixed this issue!
Thank you for your help over four pages of trial and error.
P.S.: I’d like to mark the thread resolved, but it seems I can’t edit the title?
So probably fsck’s behaviour changed regarding that. Maybe it silently ignored that error upto 13.1 or accepted the '\ ’ as space?
But then, as mentioned, the fstab manpage specifically states that you should use ‘\040’ instead of a space, in 12.2 already btw.
So one could say upto now it only worked out of pure luck…
P.S.: I’d like to mark the thread resolved, but it seems I can’t edit the title?
You can only change the title of a reply. Add “(SOLVED)” or similar there.
Thanks for the tips – I ran into a similar issue when steam filled up my /tmp with a 10+ GB stdout file. Bash tab expansion wasn’t even working. It wasn’t the same as the /etc/fstab error, but the hints about tmp got me looking in the right place.