I woke up to my site being down this morning. I called the tech guys and they told me my system partition was full, here is what they said.<br /><br />There were symlinks for /usr and /var that were pointing to themselves, which I deleted, but your / directory if still showing to be 100% full. You need to log into your machine via ssh and this data that has been generated and filled this directory. For us to perform any further work, it will incur administrative charges.<br /><br />I understand that these things are going to happen and I will have to pay but I haven't the slightest idea of what a symlink is and why they are pointing to themselves. They are currently working on it but I am wondering if there is something in my threads that may be causing that, I would like to prevent it in the future.

They found my problem, it turns out I am saving my backups to my /home which doesn't sound like a good idea <img src="http://www.ubbdev.com/forum/images/graemlins/blush.gif" alt="" /> I can't find an option to just overwrite nightly, this seems like it would be ideal. So I am forced to change the location of it, just not sure where to. Any ideas of a good place off of /home for this? Here is what the current setup is.

well you could.. do it then move it either that or get a second drive for backups..

I could do what, overwrite them nightly you mean?

I always run a second drive, and mount it as /backup and then dump all the backups to that. You need to have a fairly large amount of space to be able to hold the backups as things like your log files and such grow as well.

So there is no way to just have a nightly backup that overwrites the previous nights? Which is cool I will just have to RTFM <img src="http://www.ubbdev.com/forum/images/graemlins/wink.gif" alt="" /> Also, I take it I will have to command line in there to create and mount a partition is that correct? Thanks for the help guys <img src="http://www.ubbdev.com/forum/images/graemlins/smile.gif" alt="" />

well seeing as each backup comes in the way of backup-couch-tomatoe.cc-2-21-2004.tar.gz or such it wouldn't overwrite it because of the date part. What you would have to do is remove the old one once a ne one is made. OR a better solution is to get a second drive.

I think what you want is maybe the option -u after you make the file name change so that it only updates the files in the archive that have changed?<br /><br />NAME<br />tar - The GNU version of the tar archiving utility <br />SYNOPSIS<br />tar [ - ] A --catenate --concatenate | c --create | d --diff --compare | r --append | t --list | u --update | x -extract --get [ --atime-preserve ] [ -b, --block-size N ] [ -B, --read-full-blocks ] [ -C, --directory DIR ] [ --checkpoint ]<br />[ -f, --file [HOSTNAME:]F ] [ --force-local ]<br />[ -F, --info-script F --new-volume-script F ] [ -G, --incremental ] [ -g, --listed-incremental F ] [ -h, --dereference ] [ -i, --ignore-zeros ] [ -j, -I, --bzip ] [ --ignore-failed-read ] [ -k, --keep-old-files ] [ -K, --starting-file F ] [ -l, --one-file-system ] [ -L, --tape-length N ] [ -m, --modification-time ] [ -M, --multi-volume ] [ -N, --after-date DATE, --newer DATE ] [ -o, --old-archive, --portability ] [ -O, --to-stdout ] [ -p, --same-permissions, --preserve-permissions ] [ -P, --absolute-paths ] [ --preserve ]<br />[ -R, --record-number ] [ --remove-files ] [ -s, --same-order, --preserve-order ] [ --same-owner ] [ -S, --sparse ] [ -T, --files-from=F ] [ --null ]<br />[ --totals ]<br />[ -v, --verbose ] [ -V, --label NAME ] [ --version ]<br />[ -w, --interactive, --confirmation ] [ -W, --verify ]<br />[ --exclude FILE ] [ -X, --exclude-from FILE ] [ -Z, --compress, --uncompress ] [ -z, --gzip, --ungzip ]<br />[ --use-compress-program PROG ] [ --block-compress ] [ -[0-7][lmh] ]<br /><br />filename1 [ filename2, ... filenameN ] <br />directory1 [ directory2, ...directoryN ] <br /> <br />DESCRIPTION<br />This manual page documents the GNU version of tar , an archiving program designed to store and extract files from an archive file known as a tarfile. A tarfile may be made on a tape drive, however, it is also common to write a tarfile to a normal file. The first argument to tar must be one of the options: Acdrtux, followed by any optional functions. The final arguments to tar are the names of the files or directories which should be archived. The use of a directory name always implies that the subdirectories below should be included in the archive. <br /><br />FUNCTION LETTERS<br />One of the following options must be used: <br />-A, --catenate, --concatenate <br />append tar files to an archive <br />-c, --create <br />create a new archive<br /><br />-d, --diff, --compare <br />find differences between archive and file system <br />--delete <br />delete from the archive (not for use on mag tapes!)<br /><br />-r, --append <br />append files to the end of an archive<br /><br />-t, --list <br />list the contents of an archive<br /><br />-u, --update <br />only append files that are newer than copy in archive<br /><br />-x, --extract, --get <br />extract files from an archive<br /><br /> <br />OTHER OPTIONS<br />--atime-preserve <br />don't change access times on dumped files<br /><br />-b, --block-size N <br />block size of Nx512 bytes (default N=20)<br /><br />-B, --read-full-blocks <br />reblock as we read (for reading 4.2BSD pipes)<br /><br />-C, --directory DIR <br />change to directory DIR<br /><br />--checkpoint <br />print directory names while reading the archive<br /><br />-f, --file [HOSTNAME:]F <br />use archive file or device F (default /dev/rmt0)<br /><br />--force-local <br />archive file is local even if has a colon<br /><br />-F, --info-script F --new-volume-script F <br />run script at end of each tape (implies -M) <br />-G, --incremental <br />create/list/extract old GNU-format incremental backup<br /><br />-g, --listed-incremental F <br />create/list/extract new GNU-format incremental backup <br />-h, --dereference <br />don't dump symlinks; dump the files they point to<br /><br />-i, --ignore-zeros <br />ignore blocks of zeros in archive (normally mean EOF)<br /><br />-j, -I, --bzip <br />filter the archive through bzip2. Note: -I is deprecated and may get a different meaning in the near future. <br />--ignore-failed-read <br />don't exit with non-zero status on unreadable files<br /><br />-k, --keep-old-files <br />keep existing files; don't overwrite them from archive<br /><br />-K, --starting-file F <br />begin at file F in the archive<br /><br />-l, --one-file-system <br />stay in local file system when creating an archive<br /><br />-L, --tape-length N <br />change tapes after writing N*1024 bytes<br /><br />-m, --modification-time <br />don't extract file modified time<br /><br />-M, --multi-volume <br />create/list/extract multi-volume archive<br /><br />-N, --after-date DATE, --newer DATE <br />only store files newer than DATE<br /><br />-o, --old-archive, --portability <br />write a V7 format archive, rather than ANSI format<br /><br />-O, --to-stdout <br />extract files to standard output<br /><br />-p, --same-permissions, --preserve-permissions <br />extract all protection information <br />-P, --absolute-paths <br />don't strip leading `/'s from file names<br /><br />--preserve <br />like -p -s<br /><br />-R, --record-number <br />show record number within archive with each message<br /><br />--remove-files <br />remove files after adding them to the archive<br /><br />-s, --same-order, --preserve-order <br />list of names to extract is sorted to match archive<br /><br />--same-owner <br />create extracted files with the same ownership <br /><br />-S, --sparse <br />handle sparse files efficiently<br /><br />-T, --files-from=F <br />get names to extract or create from file F<br /><br />--null <br />-T reads null-terminated names, disable -C<br /><br />--totals <br />print total bytes written with --create<br /><br />-v, --verbose <br />verbosely list files processed<br /><br />-V, --label NAME <br />create archive with volume name NAME<br /><br />--version <br />print tar program version number<br /><br />-w, --interactive, --confirmation <br />ask for confirmation for every action<br /><br />-W, --verify <br />attempt to verify the archive after writing it<br /><br />--exclude FILE <br />exclude file FILE<br /><br />-X, --exclude-from FILE <br />exclude files listed in FILE<br /><br />-Z, --compress, --uncompress <br />filter the archive through compress<br /><br />-z, --gzip, --ungzip <br />filter the archive through gzip<br /><br />--use-compress-program PROG <br />filter the archive through PROG (which must accept -d)

Is this a cpanel server?<br /><br />Create a /backups directory off the server root and things should be fine. If you backup to the home directory - you are backing up the backup every night. And using the normal Cpanel backups (daily) you'll have 3 copies - monthly, weekly and daily. You just don't want this anywhere inside your /home directory.<br /><br />If you use the "Incrimental Backup" then rather than tarring and gzipping all the accounts in the home directory nightly, it'll just make one backup in use rsynch to update nightly ONLY what has changed.<br /><br />I have one server with a HUGE account on it - and the nightly tar and gzip really send server load through the roof as they have lots of photo and video files. So I use the incrimental backup function. That sounds like what you are looking for. Just create a new directory off the web root called /backup or something. <img src="http://www.ubbdev.com/forum/images/graemlins/smile.gif" alt="" />

Got it Josh, once again you come through and saved the day <img src="http://www.ubbdev.com/forum/images/graemlins/yay.gif" alt="" /> Thanks to all for the help here. I am getting pretty used to command lining my way around in there, funny thing is that once you log in as root you still have to cd .. up a level. I also installed Pico Josh, I wasn't liking that VI much <img src="http://www.ubbdev.com/forum/images/graemlins/tongue.gif" alt="" />

Yeah, I prefer pico - never really used vi much, so it's not what I'm comfortable in.