[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]

[cobalt-users] SOLVED ... RAQ3i - out of space, but really shouldn't be



I SOLVED IT :)

For anyone interested: my batch file must have gotten corrupted someplace, I found that my /dev/hdbX names had a ? after them, ie: /dev/hdb1?

I had a bunch of other file matching the mask of hdb1X so I took a chance and did a mv on the other file, ie: mv /dev/hdb10 /dev/hdbX10

So all my good files were hdbX10,11, etc., Then I did an rm hdb1* then mv'd all my files back. I found that my batch file did this on all 4 partitions, so I have to remove 4 bad files in this mannner, but ALL of my #'s are back to normal, df looks good once again.

WHAT A RELIEF!!!!!!!!!!!!!!!!!!!!

Thanks for thinking about me :)
Jale

At 09:34 AM 11/2/2002 -0500, you wrote:
Good morning, what a way to start a day.

My Raq3i is telling me / is full - but I think something I did is causing the message. I have a 2nd drive in the machine and one per week I have been using dd to copy drive to drive. Here are the commands I've been using:

dd if=/dev/hda1 of=/dev/hdb1
dd if=/dev/hda2 of=/dev/hdb2
dd if=/dev/hda3 of=/dev/hdb3
dd if=/dev/hda4 of=/dev/hdb4

So instead of typing them in one at a time, I created a small text file with the 4 commands in it, ftp'd this file to my RAQ and ran it from root.

After the 1st line completed, the last 3 lines all failed saying the system was out of disk space, so I just figured it didn't like the way I made my batch file and I typed in the lines one at a time. That seemed to be going just fine but then I got in an ADMIN email saying / was out of space.

I did a df and sure enough, it shows it is out of space. I deleted about 18meg of files and it still says out of space. Active monitor was flashing when I went into the GUI, DISK was red, but when I went into DISK, both show as green, with / saying it has 179 meg free, but df still says 0.

Now what I found is that I have a file created today called /dev/hdb1 owned by root that is 258meg ... but all the other files are very tiny, ie: /dev/hdb2. The other item of interest is the permission on the good entries start with a b (lower case b) and the bad file is - (hyphen). I tried to just delete the file and get File Not Found.

SO - I think I may have found the answer to why it got created, it probably has an invisible character of some kind in the name, probably a typo of some kind in my batch file ... and I don't think I can just rm hdb* as I have hdb10 and other files, and I don't know where the invisible character might be, like right after the h.

***** Is there a way to see if a file name has an invisible character in it? ****** If I can delete this file, the rest of this message becomes irrelevant I would believe.

----------
I did mount on my hdb1 (backup drive) and df on that looked fine with available: 253042 66% in use, but I'd expect that.

Now I went to unmount my backup drive - maybe I don't remember my unmount command, I thought it was either umount mnt or umount /mnt - so I tried both, and what I get back is:

error writing /etc/mtab.tmp: No space left on device

So I did an ls -l on that file as get:
-rw-rw-r--   1 root     root            0 Nov  2 09:13 /etc/mtab.tmp

----------------------------
This is from the GUI:
Total Space Free Space % Used
OS and Programs         920.0 MB        179.7 MB        80.5
Virtual Sites            11643.1 MB     9122.6 MB       21.6
home                    10917.1 MB      8875.5 MB       18.7
-----------------------------

So, what do I do? Why does the GUI think I have space and df does not? Any clues on how to delete this 258meg file that doesn't seem to want to go away?

Do I put my backup drive in as primary (use my backup)???

You all know how much assistance is greatly appreciated as I am really confused.

Thanks!!
Jale
_____________________________________
cobalt-users mailing list
cobalt-users@xxxxxxxxxxxxxxx
To subscribe/unsubscribe, or to SEARCH THE ARCHIVES, go to:
http://list.cobalt.com/mailman/listinfo/cobalt-users