1 |
Alan McKinnon <alan.mckinnon <at> gmail.com> writes: |
2 |
|
3 |
|
4 |
> > One of the workstations (amd64 2gig ram) has a load that never drops |
5 |
> > below 1.0, as seen by top. Looking at a ps nothing stands out. I did |
6 |
> > notice that 'X' is at the top of the list, even when the machine is |
7 |
> > quiescent (nobody doing anything). Suspiciaous. Clearly I have a run |
8 |
> > away or hidden process using resources. Although all my system run |
9 |
> > kde 3.5.8 only one shows this problem. |
10 |
|
11 |
|
12 |
> vmstat is your friend here. It's all in the man page, so use it and |
13 |
> narrow down the process that's blocking. Maybe you have a threading |
14 |
> race condition or similar. |
15 |
|
16 |
# vmstat |
17 |
procs -----------memory---------- ---swap-- -----io---- -system-- ----cpu---- |
18 |
r b swpd free buff cache si so bi bo in cs us sy id wa |
19 |
0 0 0 847368 224736 403404 0 0 26 12 172 251 1 0 98 1 |
20 |
|
21 |
vmstat -s |
22 |
|
23 |
2057808 total memory |
24 |
1212156 used memory |
25 |
611628 active memory |
26 |
341672 inactive memory |
27 |
845652 free memory |
28 |
224784 buffer memory |
29 |
404524 swap cache |
30 |
6273340 total swap |
31 |
0 used swap |
32 |
6273340 free swap |
33 |
20189 non-nice user cpu ticks |
34 |
110 nice user cpu ticks |
35 |
3748 system cpu ticks |
36 |
2444468 idle cpu ticks |
37 |
28905 IO-wait cpu ticks |
38 |
588 IRQ cpu ticks |
39 |
80 softirq cpu ticks |
40 |
0 stolen cpu ticks |
41 |
659529 pages paged in |
42 |
289340 pages paged out |
43 |
0 pages swapped in |
44 |
0 pages swapped out |
45 |
4307893 interrupts |
46 |
6269353 CPU context switches |
47 |
1202832933 boot time |
48 |
7300 forks |
49 |
|
50 |
|
51 |
|
52 |
> Also look into a hardware difference between this machine and the |
53 |
> others, and differences in the kernel config and loaded modules. |
54 |
|
55 |
Nothing here, all is similar to other system. And historically, |
56 |
this system has not had this problem. I'm not certain when it started: |
57 |
|
58 |
$ w |
59 |
14:39:25 up 3:23, 1 user, load average: 1.00, 1.00, 1.00 |
60 |
|
61 |
|
62 |
It looks more like corruption in the application binary to me. |
63 |
When have you ever seen a system at all three timing interval |
64 |
locked at 1.00 when a system is quiescent? |
65 |
|
66 |
What package is top part of? |
67 |
|
68 |
|
69 |
> If all this reveals nothing, then maybe you do have a suspicious |
70 |
> problem. In which case, post back real quick |
71 |
|
72 |
|
73 |
I do not suspect a 'hack' is involved, because if I pull the ethernet |
74 |
cable, it does not effect the load (still at 1.00). |
75 |
|
76 |
|
77 |
|
78 |
Jame |
79 |
|
80 |
|
81 |
|
82 |
|
83 |
-- |
84 |
gentoo-user@l.g.o mailing list |