High CPU system usage
Ugo Bellavance
ugob at CAMO-ROUTE.COM
Wed Aug 24 16:38:10 IST 2005
[ The following text is in the "ISO-8859-1" character set. ]
[ Your display is set for the "US-ASCII" character set. ]
[ Some characters may be displayed incorrectly. ]
Glenn Steen wrote:
> On 24/08/05, Ugo Bellavance <ugob at camo-route.com> wrote:
>
>>Glenn Steen wrote:
>>
>>>On 23/08/05, Ugo Bellavance <ugob at camo-route.com> wrote:
>>>(snip)
<snip>
>
> Not really, although this should do... I was thinking of
>
> 00:01:01 DEV tps rd_sec/s wr_sec/s
> 00:11:01 dev3-0 3.24 1.93 112.65
Here it is:
09:50:01 dev8-0 72.46 32.03 2121.82
09:50:01 dev8-1 0.00 0.00 0.00
09:50:01 dev8-2 0.00 0.00 0.00
09:50:01 dev8-3 5.20 0.00 1040.16
09:50:01 dev8-4 0.00 0.00 0.00
09:50:01 dev8-5 55.32 14.43 716.79
09:50:01 dev8-6 11.94 17.60 364.88
10:00:00 dev8-0 99.58 51.44 2297.69
10:00:00 dev8-1 0.00 0.00 0.00
10:00:00 dev8-2 0.02 0.00 2.75
10:00:00 dev8-3 4.76 0.03 925.54
10:00:00 dev8-4 0.00 0.00 0.00
10:00:00 dev8-5 80.35 19.32 985.79
10:00:00 dev8-6 14.46 32.10 383.60
10:10:01 dev8-0 57.68 22.15 2480.38
10:10:01 dev8-1 0.00 0.00 0.00
10:10:01 dev8-2 0.09 0.00 18.95
10:10:01 dev8-3 7.36 0.04 1516.32
10:10:01 dev8-4 0.00 0.00 0.00
10:10:01 dev8-5 38.12 9.07 577.70
10:10:01 dev8-6 12.11 13.04 367.41
10:20:01 dev8-0 40.39 14.40 1889.31
10:20:01 dev8-1 0.00 0.00 0.00
10:20:01 dev8-2 0.00 0.00 0.00
10:20:01 dev8-3 5.85 0.00 1171.87
10:20:01 dev8-4 0.00 0.00 0.00
10:20:01 dev8-5 24.63 6.30 383.94
10:20:01 dev8-6 9.90 8.10 333.50
10:30:01 dev8-0 36.28 13.49 1597.77
10:30:01 dev8-1 0.00 0.00 0.00
10:30:01 dev8-2 0.00 0.00 0.00
10:30:01 dev8-3 4.71 0.00 920.84
10:30:01 dev8-4 0.00 0.00 0.00
10:30:01 dev8-5 21.87 6.10 348.67
10:30:01 dev8-6 9.71 7.39 328.26
10:40:00 dev8-0 46.00 16.32 2276.11
10:40:00 dev8-1 0.00 0.00 0.00
10:40:00 dev8-2 0.00 0.01 0.00
10:40:00 dev8-3 6.62 0.00 1316.00
10:40:00 dev8-4 0.00 0.00 0.00
10:40:00 dev8-5 26.94 7.21 419.03
10:40:00 dev8-6 12.44 9.09 541.09
10:49:59 dev8-0 124.77 52.40 2598.49
10:49:59 dev8-1 0.00 0.00 0.00
10:49:59 dev8-2 0.07 0.11 13.04
10:49:59 dev8-3 3.81 0.00 733.85
10:49:59 dev8-4 0.00 0.00 0.00
10:49:59 dev8-5 99.05 35.33 1098.75
10:49:59 dev8-6 21.85 16.96 752.84
11:00:01 dev8-0 232.24 97.07 2778.52
# mount
/dev/sda5 on / type ext3 (rw)
none on /proc type proc (rw)
none on /dev/pts type devpts (rw,gid=5,mode=620)
usbdevfs on /proc/bus/usb type usbdevfs (rw)
/dev/sda1 on /boot type ext3 (rw)
none on /dev/shm type tmpfs (rw)
/dev/sda3 on /tmp type ext3 (rw)
/dev/sda6 on /var type ext3 (rw)
none on /var/spool/MailScanner/incoming type tmpfs (rw)
> ....
> which perhaps works better in 1.6 anyway, so...
> You seem to have a "write peak" at ~ 1.36 MiB/s, which shouldn't be
> partcularily close to your theoretical max, ergo not likely to be
> close to your real max (whatever that is:-).
I get this from hdparm, while the server has a load of 2.5 (not idle)
/dev/sda:
Timing buffer-cache reads: 3348 MB in 2.00 seconds = 1674.00 MB/sec
Timing buffered disk reads: 168 MB in 3.01 seconds = 55.81 MB/sec
>
>
>>I can't really think of a way to determine this 'measurable max'.
>>HDparm? Analysis of a sar report while testing hard drive?
>
>
> More or less so, yes. Also try using iostat.
> Perhaps add in Bonnie there too
> (http://www.garloff.de/kurt/linux/bonnie/)... and perhaps eye through
> http://www.netlib.org/performance/papers/bonnie/ ... disk and fs
> performance measuring isn't easy;-).
I am affraid of installing bonnie, as it screws Mondorescue, which I use
for backups :(.
>
> <thinking_out_load>
> But... I don't *think* you're I/O-starved. That top someone sugegsted
> should probably show what's up (or at least "who":).
Will do in the next burst.
>
> From what you show in the previous messages I think we can drop swap
> from the picture... there hardly seem to be any taking place.
That is what I think too...
> What fs do you use? Perhaps an issue with that...
ext3, might be worthwhile trying without journalling. We have an UPS
and the server is controlled by Apcupsd.
>
> Or perhaps some really crummy NIC driver and a spell of network
> congestion driving the CPU nuts? It has been known to happen (mostly
> back in the dark ages, but still:-).
Stock driver in RHEL, e1000 driver, Dell poweredge server, intel pro
1000 NIC.
>
> Oh well, 'nuff rambling
> </thinking_out_load>
Thanks a lot,
--
Ugo
-> Please don't send a copy of your reply by e-mail. I read the list.
-> Please avoid top-posting, long signatures and HTML, and cut the
irrelevant parts in your replies.
------------------------ MailScanner list ------------------------
To unsubscribe, email jiscmail at jiscmail.ac.uk with the words:
'leave mailscanner' in the body of the email.
Before posting, read the Wiki (http://wiki.mailscanner.info/) and
the archives (http://www.jiscmail.ac.uk/lists/mailscanner.html).
Support MailScanner development - buy the book off the website!
More information about the MailScanner
mailing list