mounted fat32 filesystem becomes readonly after sometime, wierd?

koodoo · 11-02-2005, 12:28 PM

Hi,

I have two fat32 partitions which I've mounted under /win1 & /win2 respectively. The lines corresponding to those in /etc/fstab are :

Code:

 /dev/hda1        /win1            vfat        defaults,umask=000         1   0
 /dev/hda5        /win2            vfat        defaults,umask=000         1   0

Now sometimes /win2 becomes readonly. I then have to unmount it and then mount it again. It works fine but then after sometime it becomes readonly again. I've had no problems with /win1. Don't know why this is happening. Any ideas ?

Thanks in anticipation.

MasterC · 11-02-2005, 09:10 PM

No idea, but you may be able to grep through your dmesg to see if there are any corresponding entries for /dev/hda5 to help you diagnose the problem.

dmesg | grep hda5
dmesg | grep win2

You might also check your system log and see if /proc has anything about it. Check /etc/mtab after it gets mounted just to see if it actually is.

Lastly, are you sure it's becoming read-only, or is it just specific files?

Cool

koodoo · 11-03-2005, 10:09 AM

Hi,

Thanks for the reply. I tried using dmesg but it's giving me problems as well. The output of dmesg is :

Code:

koodoo@knapsacker:~$ dmesg
verflow.
Neighbour table overflow.
Neighbour table overflow.
Neighbour table overflow.
Neighbour table overflow.
Neighbour table overflow.
Neighbour table overflow.
Neighbour table overflow.
NET: 64491 messages suppressed.
Neighbour table overflow.
NET: 24705 messages suppressed.
Neighbour table overflow.
NET: 39794 messages suppressed.

######### lots of this ###########

Neighbour table overflow.
Neighbour table overflow.
NET: 64491 messages suppressed.
Neighbour table overflow.
NET: 8924 messages suppressed.
Neighbour table overflow.
NETDEV WATCHDOG: eth0: transmit timed out
eth0: Tx timed out, cable problem? TSR=0x6, ISR=0x0, t=24.

I checked the system log files. There were some message files that contained the following contents :

Code:

Nov  1 04:53:29 knapsacker kernel: 03:05: rw=0, want=1050881691, limit=9783553
Nov  1 04:53:29 knapsacker kernel: attempt to access beyond end of device
Nov  1 04:53:29 knapsacker kernel: 03:05: rw=0, want=1050881692, limit=9783553
Nov  1 04:53:29 knapsacker kernel: attempt to access beyond end of device
Nov  1 04:53:29 knapsacker kernel: 03:05: rw=0, want=1050881692, limit=9783553
Nov  1 04:53:29 knapsacker kernel: attempt to access beyond end of device
Nov  1 04:53:29 knapsacker kernel: 03:05: rw=0, want=1050881693, limit=9783553
Nov  1 04:53:29 knapsacker kernel: attempt to access beyond end of device

I checked the file /var/log/dmesg. The only references to hda5/win2 that it contained were :

Code:

Partition check:
hda: hda1 hda2 < hda5 hda6 hda7 >

Ya, I'm sure it's becoming read-only, i.e. the entire filesystem and not just the individual files.
/etc/mtab shows that it's mounted when I mount it.
BTW If this might be useful, my system's been up for 9 days. Here's the output of uptime :

Code:

koodoo@knapsacker:~$ uptime
 20:44:56 up 9 days,  8:50,  1 user,  load average: 0.02, 0.08, 0.20
koodoo@knapsacker:~$

Thanks in anticipation.

koodoo · 11-13-2005, 03:28 AM

Hi,

So I restarted my P.C. (after a long time). When I run the dmesg command I get a lot of output but there are only two lines repeated over and over again.

Code:

Neighbour table overflow.
NET: <some integer> messages suppressed.

tkedwards · 11-13-2005, 07:01 PM

Neighbour table overflow: http://groups.google.com.au/group/fa...c9cf107fb29f07

Probably doesn't have anything to do with your FAT32 drive problem.

koodoo · 11-17-2005, 03:08 PM

Hi thanks Tkedwards for the help and sorry for replying so very late. I searched google as well as these forums for Neighbour Table Overflow problems. It seems that this problem is because the arp cache gets overflowed because of :
1) Some application that generates much traffic (eg. Lisa)
2) Some machine in your subnet that generates a lot of traffic and
3) Related things

Possible solutions included :
1) Increasing the size of the arp cache
2) Shutting down services like Lisa and,
3) Identifying machines generating a lot of traffic and shutting them down.

I tried all of these but none worked for me. I guess I'll stick with one of these threads to get my problem solved :

1) http://www.linuxquestions.org/questi...table+overflow
2) http://www.linuxquestions.org/questi...table+overflow
3) http://www.linuxquestions.org/questi...table+overflow
4) http://www.linuxquestions.org/questi...table+overflow
5) http://www.linuxquestions.org/questi...table+overflow
6) http://www.linuxquestions.org/questi...table+overflow

Seems like a lot of people are having problems with this.

The good part however is that amazingly my /win2 partition has stopped becoming read-only on its own (I seriously don't know why?).
I'll post the output of dmesg as soon as I am able to get it working.

Thanks again for the help.

devbro · 11-24-2005, 12:34 AM

it may have been a cluster problem. i had the same problem and read a forum saying that using /sbin/fsck.vfat will remove the cluster problem. after i did it i had no more problems.