multiple logins

edgjerp · 02-23-2005, 02:59 PM

My computer has started to increase the number of local logins, making the system slow. typically I have 5+ copies of myself logged in at any given time (I think it is 2 + number of console sessions). like this:

Code:

$ users
gjerp gjerp gjerp gjerp gjerp gjerp

this was not the case until recently, before I only got one login per tty.

Code:

$ w
 21:50:12 up  1:43,  6 users,  load average: 0,53, 0,64, 0,60
USER     TTY        LOGIN@   IDLE   JCPU   PCPU WHAT
gjerp    pts/0     20:11    1:39m  0.00s  0.32s kdeinit: kwrited
gjerp    pts/1     20:22   19:59   0.20s  0.07s /bin/bash
gjerp    pts/2     21:33    6:03   1.44s  1.38s top
gjerp    pts/3     21:48    0.00s  0.04s  0.00s w
gjerp    pts/6     20:19   16:17   0.15s  0.11s ssh

anyone know how to fix this?

using mdk 10.1 OE

Matir · 02-23-2005, 03:25 PM

That's one login per tty.... what's the problem? It looks fairly normal to me.

edgjerp · 02-23-2005, 03:33 PM

the top part, with six logins is new, and annoying. before, whenever I typet "users", I got one of each logged in user, no matter how many consoles I had open. this problem appears to significantly slow down my system.

Matir · 02-23-2005, 03:36 PM

First off, users is supposed to display one entry PER terminal that is logged in. Showing 6 users is the _correct_ result. Second, what are the specs on your system that is being slowed by this? What all are you running?

edgjerp · 02-23-2005, 03:46 PM

the tricky part is that top shows minimal cpu and memory usage, the system just responds slowly. It does not care whether half the consoles are root, still only my regular user is registered by "users"

system specs are: intel celeron 2,4 G and 512 ram

bunnadik · 02-24-2005, 01:23 AM

What does 'dmesg ', 'cat /var/log/messages' and 'free' show? Anything unusual?

Do you run different terminals as root and gjerp? On my system (10.1OE) aterm does not register addidional
users whereas xterm does. That depends on whether they write to /var/run/utmp or not.

- Peder

edgjerp · 02-24-2005, 02:49 AM

the main problem seems to be some memory hog I cannot seein 'top'

Code:

top - 09:38:46 up 5 min,  5 users,  load average: 1.45, 0.98, 0.45
Tasks:  83 total,   1 running,  82 sleeping,   0 stopped,   0 zombie
Cpu(s): 47.7% us, 20.2% sy,  0.0% ni,  0.0% id, 28.5% wa,  3.0% hi,  0.7% si
Mem:    483912k total,   480664k used,     3248k free,      404k buffers
Swap:  1124508k total,      164k used,  1124344k free,   322232k cached

  PID USER      PR  NI  VIRT  RES  SHR S %CPU %MEM    TIME+  COMMAND
 3563 root      16   0 50296  13m  38m S  8.3  2.8   0:23.41 X
   35 root      15   0     0    0    0 S  1.3  0.0   0:00.67 kswapd0
 4970 gjerp     15   0 57996  23m  19m S  1.0  4.9   0:08.50 valknut
 4011 gjerp     17   0 27936  19m  15m S  0.3  4.1   0:01.65 net_applet
 4718 gjerp     15   0 27640  15m  25m S  0.3  3.3   0:02.19 kdeinit
 4732 gjerp     15   0 27168  14m  24m S  0.3  3.1   0:02.19 kdeinit
    1 root      16   0  1408  496 1256 S  0.0  0.1   0:01.57 init
    2 root      34  19     0    0    0 S  0.0  0.0   0:00.00 ksoftirqd/0
    3 root       5 -10     0    0    0 S  0.0  0.0   0:00.00 events/0
    4 root      14 -10     0    0    0 S  0.0  0.0   0:00.00 khelper
    5 root       5 -10     0    0    0 S  0.0  0.0   0:00.07 kblockd/0
   31 root      15   0     0    0    0 S  0.0  0.0   0:00.00 kapmd
   33 root      15   0     0    0    0 S  0.0  0.0   0:00.00 pdflush
   34 root      15   0     0    0    0 S  0.0  0.0   0:00.01 pdflush
   36 root      14 -10     0    0    0 S  0.0  0.0   0:00.00 aio/0
  149 root      16   0     0    0    0 S  0.0  0.0   0:00.00 kseriod
  277 root      15   0     0    0    0 S  0.0  0.0   0:00.03 kjournald
  427 root       6 -10  1388  360 1236 S  0.0  0.1   0:00.04 udevd
  820 root      15   0     0    0    0 S  0.0  0.0   0:00.00 khubd
 1370 root      15   0     0    0    0 S  0.0  0.0   0:00.00 kjournald

as you can see, I have about 3 M of memory left after 5 min runtime. valknut seems to be related to the problem, but it is not the only one, since I could use it all the time before.

bunnadik · 02-24-2005, 03:21 AM

Quote:

the main problem seems to be some memory hog I cannot seein 'top'

Yes and no.
You have 483912k total and 3248k free but out of that 483912k, 322232k is cache.
Unix allocates the memory not used for programs as filesystem cache to speed up file access.
When more memory is needed for programs the cache size is reduced automagically.

Your load seems a bit high though "load average: 1.45, 0.98, 0.45"
My dual 400MHz which acts as webserver/squidproxy/spamfilter has a load of "0.67, 0.70, 0.85"
Does 'ps -ax' show any zombie (Z) or dead (D) processes?

- Peder

edgjerp · 02-24-2005, 10:10 AM

I think there is something wrong with my second HDD, during reboot after a "slowdown" (happens suddenly, free memory jumps from 270M down to 3-4M), I get massive amounts of error messages, all stating the same over and over:

Code:

Feb 24 10:07:00 localhost kernel: hdb: read_intr: status=0x59 { DriveReady SeekComplete DataRequest Error } 
Feb 24 10:07:00 localhost kernel: hdb: read_intr: error=0x40 { UncorrectableError }, LBAsect=40902155, sector=40902155 
Feb 24 10:07:00 localhost kernel: end_request: I/O error, dev hdb, sector 40902155 
Feb 24 10:07:00 localhost kernel: Buffer I/O error on device hdb1, logical block 40902092

is there any way I can tell the system that "this block is broken, stay away"?