Sorry, I've been so busy to find, read and reply to this thread:
 
TheWolf, please try opening the Disk Usage Analyzer after some BA running to find where the disk space is getting consumed.  It could be an errant log(s), though probably not folding@home as I can see it was behaving.  I highly doubt it would be a swap file dynamically growing ala windows-style since it will stay contained within your swap partition.  You would have to intentionally create a swap file (non-default configuration) and even at that, it remains a fixed size.  (avoids fragmentation, if you try that in windows)
 

Xavier Zepheriousthe problem of most install is that most don't incorporate Swap partitions like the old days where we made separate partitions for swap space
:
:
the new installer defaults to all in one - so if you didn't put up enough disk space you could be in an issue (since they share the same space now like M$)

I however did a custom install and made swap partition in linux mint and made a significant drive space for the OS
general rule should be if your using all in one - 2x DDR memory
so for 16gig of mem you need 32 gigs for space min)

I thought the default Mint installs include a swap partition the size of available RAM, even up to Mint 14.  I have yet to find one that does away with a swap partition and preallocates a file in the file system for swap usage.  (which is a good practice IMO, as we can reclaim the space live if we need without shrinking/deleting swap partition, resizing partitions, and possibly rebooting)
TheWolf, it could be that a default install like was dedicating too much of 30GB HDD to a swap partition that matches your RAM capacity (12,16,24GB?) If folding exclusively, that swap partition would & should also never* be used.  *Will have to sacrifice the ability to hibernate.  To confirm how it partitioned, in the terminal please enter the following:

sudo partx /dev/sda

...and let us know what the output is.

Xavier Zepheriousnote -  the default swap on Linux is small less  than 1GB

Besides computers with <1GB installed RAM, which distros have been defaulting to that?
 
Back to TheWolf,

Xavier Zepheriousfor folding I don't think you'll have a BIGBETA or Bigadv that takes more than 4GB
so you wont hit a memory issue

If your GUI-less rig's purpose is for folding, you can skip a swap partition or file altogether since memory won't be an issue, and fully dedicate your ~30GB HDD to the OS and BA folding workspace.  I find that multi-GB swaps end up killing more performance than they're supposed to benefit because SSDs/HDDs speeds haven't caught up to our voratious memory appetites. And that's where large & fast RAM modules come in to serve the need much better than a SATA/SAS disk can.   Once we hit the need for virtual RAM with apps that saturate a 12GB or 24GB system, it's too late as the disk requests will also be so large, we will notice it stall.  (ie. how long does it take to read/write up to 48000MB?) By then, we need to consider reducing memory load, or adding more RAM. 
 
A reason Mint 14 might not be a favourite distro is it's default inheritence of unstable kernel(s)
http://forums.linuxmint.c....php?f=61&t=121383
It's really not the end of the world as you can easily rollback to a well-functioning one.
 
I'm sure you've already ruled out hardware instability since it has been going for such long periods without a hitch.
 
Though too late now, I should have steered you towards LVM2 so you could tweak with different partition setups without having to restart, let alone reinstall.  I hope you find whatever is misbehaving before it consumes your remaining ~26GB (Mint + GUI install fits in under 4GB)
 
ps. XZ, I really hope I don't sound like I'm picking on you.

daemauk

ps. XZ, I really hope I don't sound like I'm picking on you.

don't worry about it
there are plenty of people more versed that me on linux or unix (Sun work stations)
so the more info we have from experienced people the better
 
as for the swap size I did see a linux mint install having a default with a very small size (recent version and a tutorial on the install)
 
Im not an expert to be sure - just more than a casual user 
it's come a long way from when I started using linux (and win 98)
 
would you rather be working a PDP 11 and paper tape and punch cards
or  building your first PC and soldering the Chips sockets into the board (like Wozniak)
(Im from that era)
 

daemauk

To examine your swap file size & free disk space, in a terminal, can you please type in:

ls -l /host/linuxmint/disks/swap.disk && df

there shouldn't be any personal info in the output.
Cheers

From rig SR2 24GB ram 128GB SSD. I'll jump over to 1 of the other SR2 and post its info as well.
 
/host/linuxmint/disks/swap.disk
Filesystem     1K-blocks     Used Available Use% Mounted on
/dev/loop0      17884496 14169484   2806520  84% /
udev            12331964        4  12331960   1% /dev
tmpfs            4936468     1096   4935372   1% /run
none                5120        0      5120   0% /run/lock
none            12341164       76  12341088   1% /run/shm
none              102400       24    102376   1% /run/user
/dev/sda1      117211572 36139588  81071984  31% /host
 
Edit: SR2 1xE5645 + 1xE5620 16GB ram 80GB HDD.
I have this rig set to 20GB install NOW  instead of the default 18GB install.
These are all pretty fresh installs, so I'm not seeing any problems at the moment.

/host/linuxmint/disks/swap.disk
Filesystem     1K-blocks     Used Available Use% Mounted on
/dev/loop0      19900284 16075160   2814232  86% /
udev             6138868        4   6138864   1% /dev
tmpfs            2459232     1052   2458180   1% /run
none                5120        0      5120   0% /run/lock
none             6148072       76   6147996   1% /run/shm
none              102400       24    102376   1% /run/user
/dev/sda2       78075896 44724180  33351716  58% /host
 

Cpu speed from cpuinfo 3510.00Mhz 
True Frequency (without accounting Turbo) 3510 MHz 

Socket [0] - [physical cores=4, logical cores=8, max online cores ever=4] 
  CPU Multiplier 18x || Bus clock frequency (BCLK) 195.00 MHz 
  TURBO ENABLED on 4 Cores, Hyper Threading ON 
  Max Frequency without considering Turbo 3705.00 MHz (195.00 x [19]) 
  Max TURBO Multiplier (if Enabled) with 1/2/3/4 cores is  20x/20x/19x/19x 
  Real Current Frequency 3705.00 MHz (Max of below) 
        Core [core-id]  :Actual Freq (Mult.)      C0%   Halt(C1)%  C3 %   C6 %  Temp 
        Core 1 [0]:       3705.00 (19.00x)       100       0       0       0    57 
        Core 2 [1]:       3705.00 (19.00x)       100       0       0       0    61 
        Core 3 [2]:       3705.00 (19.00x)       100       0       0       0    64 
        Core 4 [3]:       3705.00 (19.00x)       100       0       0       0    67 



Socket [1] - [physical cores=6, logical cores=12, max online cores ever=6] 
  CPU Multiplier 18x || Bus clock frequency (BCLK) 195.00 MHz 
  TURBO ENABLED on 6 Cores, Hyper Threading ON 
  Max Frequency without considering Turbo 3705.00 MHz (195.00 x [19]) 
  Max TURBO Multiplier (if Enabled) with 1/2/3/4/5/6 cores is  21x/21x/20x/20x/19x/19x 
  Real Current Frequency 3705.00 MHz (Max of below) 
        Core [core-id]  :Actual Freq (Mult.)      C0%   Halt(C1)%  C3 %   C6 %  Temp 
        Core 1 [4]:       3705.00 (19.00x)       100       0       0       0    46 
        Core 2 [5]:       3705.00 (19.00x)       100       0       0       0    50 
        Core 3 [6]:       3705.00 (19.00x)       100       0       0       0    48 
        Core 4 [7]:       3705.00 (19.00x)       100       0       0       0    47 
        Core 5 [8]:       3705.00 (19.00x)       100       0       0       0    50 
        Core 6 [9]:       3705.00 (19.00x)       100       0       0       0    49 

C0 = Processor running without halting 

have you updated your system with apt-get update && apt-get upgrade?.If the answer was YES do an apt-get clean this command clears out the local repository of retrieved package files. It removes everything but the lock file from /var/cache/apt/archives/ and /var/cache/apt/archives/partial/.Probably that´s the space you have lost and you are searching ;)
Even on emergency you can put an unused pendrive and use it as a swap file after locate the mount point
 
Example : On a terminal type the following commands
cfdisk /dev/sdX(where x is the name and mountpoint of the usb pendrive) for example i will use /dev/sdd so : cfdisk /dev/sdd
change partition type to 82 (linux swap) quit and save changes.
mkswap /dev/sdd1 (formats the pendrive as a swap device)
swapon /dev/sdd1 (initializes the pendrive to be used as a swap device and you have an emergency swap )
 

Think I found the problem, well a temp fix that is. I'm still guessing here,
but looks like its holding a lot of log files in var/log.
Doing a fsck as root from within the GUI seems to clean those up quite a bit.
This fsck didn't seem to work from a recovery boot pure root.
Also not sure why those are growing over time for the one rig. Lost a 8103
at 87% done last night because of the low drive space problem.

I don't know enough about Linux to provide much help Wolf, but if I can compare something on one of my rigs, let me know.  My gut reaction to a log file that could fill-up a lot of disk space would be the same as Daemauk's.  I forget, is this a native Linux install, Wubi or VBox?  If I'm reading correctly above, you have ~2.8GB free on your disk?

Wolf, doing a little searching I see that this log file eating disk space problem (while for some people is not an issue) sometimes becomes an issue under certain conditions.  I'm gonna guess that many Folders are not triggering a condition since I don't recall people talking about it happening to them.  In your case, it would appear that you have something that is triggering a lot of log data.  Most advice that I've read is to find out what is causing the problem (which may or may not be easy for us non-Linux experts). 
 
I also see lots or people talking about using the "Logrotate" program that (if I'm reading correctly) is included with Linux.  Wonder if that program would help at least manage the logs while you diagnose the issue?
 
http://library.linode.com...ls/utilities/logrotate
 
http://www.techrepublic.c...=content;siu-container
 

They were compressed files, I don't recall there names.
I'm not at that rig at the moment "its in the shop" so I'll have a look later to see if I can recall what they where.
I backed my OC down some since there is probably some stability issue causing this continued log write.
I haven't seen any instability, its been folding well until the allowed space gets low. Folding is all the rigs does.

View common log file /var/log/messages using any one of the following command:# tail -f /var/log/messages
# less /var/log/messages
# more -f /var/log/messages
# most /var/log/messages (you need most package install with apt-get install most)

Common Linux log files name and usage

• /var/log/message: General message and system related stuff
• /var/log/auth.log: Authenication logs
• /var/log/kern.log: Kernel logs
• /var/log/cron.log: Crond logs (cron job)
• /var/log/maillog: Mail server logs
• /var/log/qmail/ : Qmail log directory (more files inside this directory)
• /var/log/httpd/: Apache access and error logs directory
• /var/log/lighttpd: Lighttpd access and error logs directory
• /var/log/boot.log : System boot log
• /var/log/mysqld.log: MySQL database server log file
• /var/log/secure: Authentication log
• /var/log/utmp or /var/log/wtmp : Login records filles

In short /var/log is the location where you should find all Linux logs file. However some applications such as httpd have a directory within /var/log/ for their own log files. You can rotate log file using logrotate (man logrotate for manual) software and monitor logs files using logwatch (apt-get install logwatch and man logwatch) software.

daemauk

Those compressed logs are just for archival purposes for humans to read.  The good thing about that is that you can simply delete them if not needed any longer to recover some space-better than fsck would.  So when you get your rig back, do extract and read them to see if you can spot where the cries for help are coming from.  You may need to open them with administrative privileges.

After reading back over this I believe this whole log build up and disk loss is caused by this error I just posted.
I been just deleting those logs since our last talks on this subject,
but would like to find a way to fix or stop this steady error reading.
I'm sure it is also hurting my folding TPF with all the activity going on in the background with the same error steady running.
 

texinga

From everything I've read (from people much more knowledgeable about Linux than me), they all (first) recommend discovering and solving what is causing the error condition that is filling the log.  I know that may sound kinda basic, but is likely the best way to deal with this situation.

Only thing I find strange about this is its only happening on the SR2 "All Three" not just one and the error is the same on each.
Error is:
picehp 0000:00:1c:0pcie04>Link Training error occurs
picehp 0000:00:1c:0pcie04>Failed to check link status
 
I'd say something to do with PCIE since that's part of the error.
Do you have Linux on your SR2 Tex? If so what flavor?
Could you do a test for me with yours to see if its getting the same error?
Pretty simple test, boot into Linux use CTRL+ALT+F1 to get to that terminal,
if you have the same error you'll see it right away or not, you can just CTRL+ALT+F8 back to the GUI at this point.
 
I'm trying to see if its the flavor of Linux that's running that might be the root cause, perhaps trying to read hardware it doesn't know how to then producing the error, because I kind of find it hard to believe that all three SR2's could have some sort of failed hardware exactly the same and be causing this error.
 

I never seen this using Vbox so I doubt you would get the error that way.
It may be Mint 14 that's the problem, but I didn't want to do a clean install
of something else at this point to see if that is the case.
All are running fine and folding well, other than I have to keep a eye on the
HDD space and clean those logs about once a week to regain the HDD space.

Maybe with any luck someone else that is running a SR2 and Linux will see this
and do the test and report back if they see the error or not and what flavor of Linux there running.
 
My next option is to try to catch one close to finishing its work unit and give the -oneunit flag.
Then wipe the HDD and install a different flavor of Linux hoping I fine one that doesn't have this same error.

texinga

If you do try another "flavor", give Ubuntu 12.04 Server version with the Musky Script a try.

Thanks Tex I did a search with the error code but didn't find anything.
As for recompile the kernel I have to clue how to do this and not wanting to learn either. lol
Not wanting to go to a full terminal client either, that would be fine if I was running a VBox in W7.
I really like Mint, but not this problem. I'll look around maybe try the Ubuntu 12.04 GUI version.

Let's see - install kernel sources, make menuconfig,.... that's all I got.  Haven't done it in 10 years and the info has dropped out of the memory banks.