XP1000, V5.1 at 100% system time

From: Uwe Lienig (Uwe.Lienig@fif.mw.htw-dresden.de)
Date: Mon Feb 23 2004 - 10:58:25 EST


Hi managers,

this one drives me crazy. But at first the basics.

hardware: XP1000, EV6@500MHz, 2 GByte mem
OS: v5.1, PK 5??

Before I wrote this question I'd like to check what pk is installed and issued
dupatch as a normal user and used the menu. I had the output for the installed
patches, but thought of the patchkit version. So I used the commandline with

# > dupatch -track -type kit

But I got only the answer, that there are no patches installed. I looked into
dupatch and noticed that dupatch looks for OSFPAT*.lk in /usr/.smbd. . But there
are no OSFPAT*.lk files anymore, despite one, that is empty. Am I allowed to
delete the OSFPAT*.lk files as a normal user? Even as root I do get the answer,
that there are no patches installed. If I check the /usr/.smdb. for OSFPAT*.inv
files they are there. Even if I go via the menu I'm no getting the answer, that
there are no patches installed. I'm stumped.

But anyway, this is not the problem I have. The system has to carry out a lot of
calculations. I'wonderer why the system is that slow. The ps showed me that the
calculation process is at 25% CPU. At first I thougt there may be some other
tasks running just at that time I took the ps snapshot. But after a while I
looked again and ps told me the calculation process is at 25% CPU. I issued
iostat 1 and saw the system is running at 100% in system mode. There is a
snipped of iostat if I stop (kill -STOP) the calculation process.
     tty floppy0 dsk0 cpu
  tin tout bps tps bps tps us ni sy id
    0 4 0 0 0 0 14 0 15 71
    0 52 0 0 0 0 0 0100 0
    0 52 0 0 0 0 0 0100 0
    0 52 0 0 0 0 0 0100 0
    0 52 0 0 0 0 0 0100 0
    0 52 0 0 0 0 0 0100 0
    0 52 0 0 0 0 0 0100 0

If I cont (kill -CONT) the calculation process, it gets around 25% of the CPU.
     tty floppy0 dsk0 cpu
  tin tout bps tps bps tps us ni sy id
    0 4 0 0 0 0 14 0 15 71
    0 52 0 0 0 0 21 0 79 0
    0 51 0 0 0 0 38 0 62 0
    0 52 0 0 0 0 32 0 68 0
    0 52 0 0 0 0 24 0 76 0
    0 52 0 0 0 0 31 0 69 0
    0 52 0 0 0 0 23 0 77 0
    0 52 0 0 0 0 28 0 72 0
    0 52 0 0 0 0 24 0 76 0
    0 52 0 0 0 0 23 0 77 0
    0 52 0 0 0 0 29 0 71 0
    0 52 0 0 0 0 23 0 77 0
    0 52 0 0 0 0 29 0 71 0
    0 52 0 0 0 0 23 0 77 0
    0 52 0 0 0 0 35 0 65 0
    0 52 0 0 0 0 25 0 75 0
    0 52 0 0 0 0 25 0 75 0

No other process is actually doing any CPU intensive things. All processes are
sleeping. I tried to figure out which process is causing this high sys demand,
but to no avail (used ps -ef -o with a lot of options). I'm not able to get
information which process is demanding. I guess if I shutdown the system the
behaviour returns to normal but I'm very suspicious whats going on here so I'll
let the system run until I know the reason. And yes I checked paging and
netstat. Nothing looks wedged.

> netstat 1
    input (tu0) output input (Total) output
  packets errs packets errs colls packets errs packets errs colls
34340088 0 58267440 3 0 35190202 0 59117554 3 0
        6 0 0 0 0 6 0 0 0 0
        8 0 0 0 0 8 0 0 0 0
        9 0 0 0 0 9 0 0 0 0
        9 0 0 0 0 9 0 0 0 0
        5 0 0 0 0 5 0 0 0 0
        8 0 0 0 0 8 0 0 0 0
        5 0 0 0 0 5 0 0 0 0

> vmstat 1
Virtual Memory Statistics: (pagesize = 8192)
   procs memory pages intr cpu
   r w u act free wire fault cow zero react pin pout in sy cs us sy id
   3 157 28 228K 11K 15K 16M 3M 4M 584 2M 0 41 1K 38K 14 15 71
   3 157 28 228K 11K 15K 4 16 18 0 28 0 15 82 242K 24 76 0
   3 157 28 228K 11K 15K 0 0 0 0 0 0 12 45 233K 26 74 0
   3 157 28 228K 11K 15K 0 0 0 0 0 0 15 53 240K 24 76 0
   3 157 28 228K 11K 15K 0 0 0 0 0 0 15 42 218K 31 69 0
   3 157 28 228K 11K 15K 0 0 0 0 0 0 14 60 244K 23 77 0
   3 157 28 228K 11K 15K 0 0 0 0 0 0 17 55 236K 26 74 0
   3 157 28 228K 11K 15K 0 0 0 0 0 0 13 47 223K 29 71 0
   3 157 28 228K 11K 15K 0 0 0 0 0 0 20 44 217K 31 69 0
   3 157 28 228K 11K 15K 0 0 0 0 0 0 142 472 265K 16 84 0

I'm out of options. Any thoughts are very welcome.

TIA

-- 
Uwe Lienig  | fon: (+49 351) 462 2780 | mailto:uwe.lienig@fif.mw.htw-dresden.de
             | fax: (+49 351) 462 3476 | http://www.fif.mw.htw-dresden.de
HTW Dresden | parcels: Gutzkowstr. 22 | letters: PF 12 07 01
    -FiF-    |          01069 Dresden  |          01008 Dresden


This archive was generated by hypermail 2.1.7 : Sat Apr 12 2008 - 10:49:52 EDT