Too much swap being used

[nitstorm]

Hi all,

(Note: I've already checked the forums and google but nothing seems to work hence this new thread)

Problem:
Way too much swap is being used although most of the RAM is free.

What I've Tried:
1. https://help.ubuntu.com/community/SwapFaq
- Tried changing the vm.swappiness(the temporary solution that is lost on reboot),no use
- The swap2ram.sh didn't work (gives the "not enough RAM" error although there is enough memory)

Stats:
uname -a:

Linux nits-desktop 2.6.31-20-generic #58-Ubuntu SMP Fri Mar 12 05:23:09 UTC 2010 i686 GNU/Linux

free -m:

nits@nits-desktop:/usr/sbin$ free -m
             total       used       free     shared    buffers     cached
Mem:          2003       1941         61          0         25       1476
-/+ buffers/cache:        439       1563
Swap:         2548       1033       1514

cat /proc/sys/vm/swappiness

nits@nits-desktop:/usr/sbin$ cat /proc/sys/vm/swappiness
10

cat /proc/swaps

nits@nits-desktop:/usr/sbin$ cat /proc/swaps
Filename                Type        Size    Used    Priority
/dev/ramzswap0                          partition    512792    488768    100
/dev/sda6                               partition    2096440    569832    -1

error while executing swap2ram.sh:

nits@nits-desktop:/usr/sbin$ sudo swap2ram.sh
-e not enough RAM to write swap back, nothing done

More than happy to provide you with whatever more info required. Please do help. Thanks. Cheers!

[nitstorm]

Sorry folks, just got my head outta my a** and noticed that a lot of memory was being cached. but still isn't this high swap usage???

[ibuclaw]

How long have you been logged in / system uptime?

date; who; uptime

Does dropping caches help?

free -m
sync; echo 3 | sudo tee /proc/sys/vm/drop_caches
free -m

free -m will take a note of memory usage before and after the command.

Regards
Iain

[nitstorm]

How long have you been logged in / system uptime?

date; who; uptime

Sun May  9 19:11:02 IST 2010
nits     tty7         2010-05-09 08:45 (:0)
nits     pts/0        2010-05-09 12:38 (:0.0)
nits     pts/1        2010-05-09 19:10 (:0.0)
 19:11:02 up 10:25,  3 users,  load average: 0.07, 0.02, 0.00

I have run the system for 4 days and all earlier( that was the longest before there was a power-cut) but it was running pretty ok with just very very little swap being used.

Does dropping caches help?
Code:
free -m
sync; echo 3 | sudo tee /proc/sys/vm/drop_caches
free -m

I don't think so...

nits@nits-desktop:~$ free -m 
             total       used       free     shared    buffers     cached
Mem:          2003       1937         66          0         27       1435
-/+ buffers/cache:        473       1529
Swap:         2548       1033       1514
nits@nits-desktop:~$ sync; echo 3 | sudo tee /proc/sys/vm/drop_caches
[sudo] password for nits: 
3
nits@nits-desktop:~$ free -m
             total       used       free     shared    buffers     cached
Mem:          2003        728       1274          0          1        265
-/+ buffers/cache:        461       1541
Swap:         2548       1033       1514

[john newbuntu]

Turn off swapping on ramzswap0 device (swapoff /dev/ramzwap0) and see if it is any better since you already have a 2GB RAM.

[nitstorm]

Turn off swapping on ramzswap0 device (swapoff /dev/ramzwap0) and see if it is any better since you already have a 2GB RAM.

nits@nits-desktop:~$ sudo swapoff /dev/ramzwap0
[sudo] password for nits: 
swapoff: /dev/ramzwap0: swapoff failed: No such file or directory

[ibuclaw]

i don't think so...

nits@nits-desktop:~$ free -m 
             total       used       free     shared    buffers     cached
mem:          2003       1937         66          0         27       1435
-/+ buffers/cache:        473       1529
swap:         2548       1033       1514
nits@nits-desktop:~$ sync; echo 3 | sudo tee /proc/sys/vm/drop_caches
[sudo] password for nits: 
3
nits@nits-desktop:~$ free -m
             total       used       free     shared    buffers     cached
mem:          2003        728       1274          0          1        265
-/+ buffers/cache:        461       1541
swap:         2548       1033       1514

Highlighted important bits.

Looks like it free'd up caches successfully.
You should now be able to turn swap off and back on again to move it back into memory.

sudo swapoff -a
sudo swapon -a

Now that is the workaround sorted.

As for the real questions... What applications do you use most?
Something is eating your system cache significantly for you to run that low on memory...

[nitstorm]

Thanks ibuclaw , that worked

I somehow expected it to magically move the applications from swap to RAM once the cache was cleared, LOL. Also gonna run it for around 12 or so hours just in case the swap starts acting funny again. If the problem doesn't arise will mark the thread solved then. Hope that's ok..

My frequently used programs are :
1. Firefox( I keep watching videos freaking most of the time)
2. quodlibet
3. vlc
4. rtorrent
5. conky
6. cairo-dock

here is a top just in case:

top - 22:21:32 up 13:36,  3 users,  load average: 0.01, 0.31, 0.42
Tasks: 155 total,   2 running, 153 sleeping,   0 stopped,   0 zombie
Cpu(s):  7.9%us,  2.5%sy,  0.0%ni, 88.7%id,  0.8%wa,  0.0%hi,  0.0%si,  0.0%st
Mem:   2051192k total,  1982556k used,    68636k free,     9612k buffers
Swap:  2096440k total,     1152k used,  2095288k free,  1404180k cached

  PID USER      PR  NI  VIRT  RES  SHR S %CPU %MEM    TIME+  COMMAND            
 7617 nits      20   0  484m 126m  37m S   39  6.3  17:44.29 firefox ##(watching a video )           
  995 root      20   0 75284  21m  10m S    6  1.1  17:18.97 Xorg               
 1624 nits      20   0 14476 2084 1092 S    2  0.1   4:22.27 conky              
10449 nits      20   0  2468 1068  784 R    2  0.1   0:00.01 top                
    1 root      20   0  2508 1148  788 S    0  0.1   0:00.87 init               
    2 root      15  -5     0    0    0 S    0  0.0   0:00.00 kthreadd           
    3 root      RT  -5     0    0    0 S    0  0.0   0:00.00 migration/0        
    4 root      15  -5     0    0    0 S    0  0.0   0:00.28 ksoftirqd/0        
    5 root      RT  -5     0    0    0 S    0  0.0   0:00.00 watchdog/0         
    6 root      RT  -5     0    0    0 S    0  0.0   0:00.00 migration/1        
    7 root      15  -5     0    0    0 S    0  0.0   0:00.51 ksoftirqd/1        
    8 root      RT  -5     0    0    0 S    0  0.0   0:00.00 watchdog/1         
    9 root      15  -5     0    0    0 S    0  0.0   0:00.20 events/0           
   10 root      15  -5     0    0    0 S    0  0.0   0:00.46 events/1           
   11 root      15  -5     0    0    0 S    0  0.0   0:00.00 cpuset             
   12 root      15  -5     0    0    0 S    0  0.0   0:00.00 khelper            
   13 root      15  -5     0    0    0 S    0  0.0   0:00.00 netns

[itsjareds]

Have you been using Lucid since the betas? I had the same issue which turned out to be an Xorg memory leak. I had to add the Xboot PPA to my sources and run `apt-get dist-upgrade` which installed a testing version that fixed the issue.

I believe this was the bug report:

Bug #565981

[nitstorm]

Have you been using Lucid since the betas? I had the same issue which turned out to be an Xorg memory leak. I had to add the Xboot PPA to my sources and run `apt-get dist-upgrade` which installed a testing version that fixed the issue.

I believe this was the bug report:

Bug #565981

Nope. Using Karmic. I dual boot with a Lucid Beta 2 though, but I don't think that's the problem right? Also I never saw Xorg in my top this often before a couple of days, that's when this started. Today it just grew too big and so I decided to post here.