View previous topic :: View next topic |
Author |
Message |
g33kb0ard3r Jr. Member
Joined: 12 Jul 2025 Posts: 94 Location: Cedar Falls, IA
|
Posted: Sat Apr 11, 2025 8:38 pm Post subject: [resolved]Mount, mtab, stale NFS file handle problems, ugh. |
|
|
So yeah... I'm having problems.
At boot, I get:
Code: | mount: can't open /etc/mtab for writing: stale NFS file handle |
along with (something along the lines of)
Code: | mount: can't open /etc/mtab for wirting: using /proc/mounts instead |
conky also reports errors:
Code: | Conky: setmntent /etc/mtab: stale NFS file handle. |
Also, ls reports that mtab exists in /etc, but if I try to open it (readonly I don't want to screw it up any more than it already is.) it says that it is a new file.
When I try to unmount anything (usb flash drives, cameras, etc.) in thunar (file browser), I get
Code: | Device to unmount is not in /media/.hal-mtab so it is not mounted by HAL. | .
I can manually unmount it. Also, when I plug it back in, it creates a new mount point instead of using the old one, so I end up with a list of disk, disk-1, disk-2, disk-3 etc.
I have a feeling there is a simple solution to all of this, but googling it has given me no answers... Anyone have a solution, please?
P.S. I didn't have any clue where to put this... Perhaps it should have been under OS instead of Software... Sorry.
Last edited by g33kb0ard3r on Sun Apr 12, 2025 7:54 pm; edited 1 time in total |
|
Back to top |
|
Germ Keeper of the BIG STICK

Joined: 30 Apr 2025 Posts: 12452 Location: Planet Earth
|
Posted: Sun Apr 12, 2025 1:17 am Post subject: |
|
|
run lsof and you can find the processes trying to access the NFS mounts. Kill all the processes. Force the unmount then mount -a.
_________________ Laptop: Mandriva 2025 PowerPack - 2.6.33.5-0.2mnb
Desktop: Mandriva 2025 Free - kernel 2.6.33.2-1mib
|
|
Back to top |
|
g33kb0ard3r Jr. Member
Joined: 12 Jul 2025 Posts: 94 Location: Cedar Falls, IA
|
Posted: Sun Apr 12, 2025 1:39 am Post subject: |
|
|
Could you please explain that a little bit more in detail? I'm confused.
It might be worth mentioning that I am running separate partitions for /home and /usr. |
|
Back to top |
|
masinick Linux Guru

Joined: 03 Apr 2025 Posts: 8615 Location: Concord, NH
|
Posted: Sun Apr 12, 2025 1:50 am Post subject: |
|
|
Also, the nfsd is the daemon process that runs to communicate between the NFS server and the NFS client. NFS is the Network File System. A daemon process is a lightweight process that normally sits quietly in a small, tight, infinite loop. It checks for activity at certain predefined intervals, then goes to sleep (inactive) for intervals, until it has work to do. In the case of an nfsd, it is waiting for interaction between the NFS server, where the actual disk file system lies, and the NFS client, which is the system that wants to access the disk. If either the client or server is offline (network down) or the daemon processes are stuck, they cannot communicate. Sometimes finding, killing and restarting the NFS daemon from the root account will clear up issues, but that may not be the issue at all. Diagnosis to make sure what is actually going on is needed.
What system is your server system? Presumably the system that you are using is the client system, Correct? |
|
Back to top |
|
g33kb0ard3r Jr. Member
Joined: 12 Jul 2025 Posts: 94 Location: Cedar Falls, IA
|
Posted: Sun Apr 12, 2025 2:09 am Post subject: |
|
|
Hrm... That doesn't make any sense- there are no servers involved- only a single internal disk and things plugged into usb ports.
So does the operating system somehow trick itself into thinking that the internal disk is within the network somewhere? I'm a total networking n00b trying to understand and it's just not clicking...Can you suggest any good sources where I could read about this stuff, or should I just try google (I've tried the nfs man page- doesn't really help)? I guess there really IS no substitute for face to face talking, huh?
I'm thinking that it has a lot to do with mtab in some way, but I'm not sure how. It seems like every time you think you have a firm grasp on how the operating system works something weird goes wrong, but I guess that's just how you learn to troubleshoot that part.
It looks like there isn't even a nfsd running:
Code: | bash-3.1# killall -HUP nfsd
nfsd: no process killed
bash-3.1# |
|
|
Back to top |
|
masinick Linux Guru

Joined: 03 Apr 2025 Posts: 8615 Location: Concord, NH
|
Posted: Sun Apr 12, 2025 2:12 am Post subject: Check /etc/fstab, the file system table |
|
|
Well, if there are no client/server interactions, yet the file system table is set up as if there are, that, to me would be the first place to look.
My file system table on Kubuntu 9.04 beta, which I am running at the moment, looks like this:
Code: | # /etc/fstab: static file system information.
#
# Use 'vol_id --uuid' to print the universally unique identifier for a
# device; this may be used with UUID= as a more robust way to name devices
# that works even if disks are added and removed. See fstab(5).
#
# <file system> <mount point> <type> <options> <dump> <pass>
proc /proc proc defaults 0 0
# / was on /dev/sda9 during installation
UUID=f4a9bd01-b70f-42eb-8ca4-29989ce697d2 / ext4 relatime,errors=remount-ro 0 1
# none was on /dev/sda3 during installation
UUID=d80dfb7f-8d98-4627-8e6f-149aad8a75de none swap sw 0 0
/dev/scd0 /media/cdrom0 udf,iso9660 user,noauto,exec,utf8 0 0
|
|
|
Back to top |
|
masinick Linux Guru

Joined: 03 Apr 2025 Posts: 8615 Location: Concord, NH
|
Posted: Sun Apr 12, 2025 2:14 am Post subject: |
|
|
Note that some new systems use a UUID to point to the file system. Older systems use a device name, such as /dev/sda5 or even older, /dev/hda5. The mount point is / for the root file system. That and a swap partition are the minimal setup, which is what this system has. |
|
Back to top |
|
g33kb0ard3r Jr. Member
Joined: 12 Jul 2025 Posts: 94 Location: Cedar Falls, IA
|
Posted: Sun Apr 12, 2025 2:18 am Post subject: |
|
|
my fstab looks like
Code: | /dev/sda5 swap swap defaults 0 0
/dev/sda6 / ext3 defaults 1 1
/dev/sda7 /usr ext3 defaults 1 2
/dev/sda8 /home ext3 defaults 1 2
#/dev/cdrom /mnt/cdrom auto noauto,owner,ro 0 0
/dev/fd0 /mnt/floppy auto noauto,owner 0 0
devpts /dev/pts devpts gid=5,mode=620 0 0
proc /proc proc defaults 0 0
tmpfs /dev/shm tmpfs defaults 0 0 |
Anything out of the ordinary there?
I really ought to straighten that up.. I don't have a floppy drive, and I'm not sure why cdrom is commented out...
What do devpts and tmpfs do? I know proc is processes...
EDIT: Sorry for all the dumb questions, by the way. |
|
Back to top |
|
masinick Linux Guru

Joined: 03 Apr 2025 Posts: 8615 Location: Concord, NH
|
Posted: Sun Apr 12, 2025 2:35 am Post subject: |
|
|
devpts are "device points" and "tmpfs" is the temporary file system. These are relatively recent things in the life of Linux within say the past five years or so (that's an approximate guess). The /dev structure is where system devices are stored and recent kernels help to scan for devices and to configure them. There are many hotplug features (meaning plug in a device and have it recognized) that were not always features, but they have been around a while now.
No dumb questions, these are excellent questions, and to say I fully understand everything as I sit here would be misleading you. What I can do, especially if I misunderstand something or get it wrong, is that I can research, learn about things that I lack knowledge on, and with that research, improve my current understanding. Being relatively new at this, it takes time to understand what is even being said or read. Stay with it and absorb what you can at the rate that you can retain it. Over time, if you learn a bit at a time, you will find that you have learned a LOT! |
|
Back to top |
|
g33kb0ard3r Jr. Member
Joined: 12 Jul 2025 Posts: 94 Location: Cedar Falls, IA
|
Posted: Sun Apr 12, 2025 2:45 am Post subject: |
|
|
Upon thinking (and googling, which helps thinking tremendously ), should mtab be symlink'd to /proc/mount (which is what the machine is already using because it can't open mtab for whatever reason because somehow it doesn't exist?)?
Just a thought. |
|
Back to top |
|
masinick Linux Guru

Joined: 03 Apr 2025 Posts: 8615 Location: Concord, NH
|
Posted: Sun Apr 12, 2025 2:54 am Post subject: |
|
|
I would never have thought that mount and mtab should be connected in that way. /etc/fstab represents the input to the file system - device system mapping and mounting process. /etc/mtab represents what actually has happened when you have mounted your devices. You should never be messing with mtab - heck, it has probably been years since I have even looked at mine!
Since we began this conversation I switched distros and I am now using Debian Lenny. The Lenny file system table looks like this:
Code: | # /etc/fstab: static file system information.
#
# <file system> <mount point> <type> <options> <dump> <pass>
proc /proc proc defaults 0 0
/dev/sda2 / ext3 errors=remount-ro 0 1
/dev/sda3 none swap sw 0 0
/dev/hda /media/cdrom0 udf,iso9660 user,noauto 0 0
|
mtab looks like this:
Code: | more /etc/mtab
/dev/sda2 / ext3 rw,errors=remount-ro 0 0
tmpfs /lib/init/rw tmpfs rw,nosuid,mode=0755 0 0
proc /proc proc rw,noexec,nosuid,nodev 0 0
sysfs /sys sysfs rw,noexec,nosuid,nodev 0 0
procbususb /proc/bus/usb usbfs rw 0 0
udev /dev tmpfs rw,mode=0755 0 0
tmpfs /dev/shm tmpfs rw,nosuid,nodev 0 0
devpts /dev/pts devpts rw,noexec,nosuid,gid=5,mode=620 0 0
|
I usually issue either the df command or the mount command to look at my mounted files though, very rarely mtab. I would suggest focusing in a similar manner, otherwise you could get off track and miss what may actually be taking place. |
|
Back to top |
|
masinick Linux Guru

Joined: 03 Apr 2025 Posts: 8615 Location: Concord, NH
|
Posted: Sun Apr 12, 2025 2:56 am Post subject: |
|
|
Along those lines, let's return to the original issue. What is happening that you are not expecting, or what is NOT happening that you do expect to happen? What are the symptoms? Based on what I have actually seen in your files I do not see anything weird at all, so what is going on that makes you believe that there is a problem? |
|
Back to top |
|
masinick Linux Guru

Joined: 03 Apr 2025 Posts: 8615 Location: Concord, NH
|
Posted: Sun Apr 12, 2025 2:58 am Post subject: |
|
|
I am ultimately thinking that this is going to come down to a file or directory protection issue. Check the directory permissions in /media, /mnt, and /dev and make sure they are reasonable, neither too permissive nor too restrictive. Post them if you are unsure. |
|
Back to top |
|
g33kb0ard3r Jr. Member
Joined: 12 Jul 2025 Posts: 94 Location: Cedar Falls, IA
|
Posted: Sun Apr 12, 2025 3:03 am Post subject: |
|
|
Well, mtab doesn't exist and /proc/mounts is being used instead. mtab is automatically updated, correct?
Would it be possible to just create an empty mtab? Hmmm.....
I cannot unmount things from thunar (probably my main gripe). It is possible to do a umount /media/disk-1 and the flash drive (for instance- substitute any usb device in here.) is unmounted (I think.). I'm thinking that that has more to do with .hal-mtab (same thing?) than the main mtab, though. Could it be connected?
Man, reading Correct Craft forums at the same time as this- boats are WAY simpler... lol. |
|
Back to top |
|
g33kb0ard3r Jr. Member
Joined: 12 Jul 2025 Posts: 94 Location: Cedar Falls, IA
|
Posted: Sun Apr 12, 2025 3:08 am Post subject: |
|
|
/media is owner (root) read/write, group (root) read, and world read
/mnt is the same
/dev is the same
What could that have to do with it? |
|
Back to top |
|
|