Re: Odd performance problem (hitching)

From: Daniel O'Connor <doconnor_at_gsoft.com.au>
Date: Thu, 8 Sep 2005 12:56:42 +0930
On Wednesday 31 August 2005 11:25, Don Lewis wrote:
> On 31 Aug, Daniel O'Connor wrote:
> > On Tuesday 30 August 2005 17:50, Daniel O'Connor wrote:
> >> > You might try significantly decreasing MAXVNODES_MAX in
> >> > sys/kern/vfs_subr.c and rebuilding your kernel.
> >>
> >> OK..
> >> I'll try timing when the drop outs happen :)
> >
> > Hmm they appear to be every 10 seconds on the dot.
>
> Hmn, I wonder what runs every 10 seconds ...

I can't think of anything.. I tried stopping a few things (IM clients, smartd, etc)

My systat output seems a little odd too..

    3 users    Load  2.00  1.41  1.33                  Sep  8 12:55

Mem:KB    REAL            VIRTUAL                     VN PAGER  SWAP PAGER
        Tot   Share      Tot    Share    Free         in  out     in  out
Act  279336   43076   541224    79892   95848 count    2
All  427596   48604 11928120   105464         pages    4
                                                                 Interrupts
Proc:r  p  d  s  w    Csw  Trp  Sys  Int  Sof  Flt        cow    1333 total
           2115      3601  154 3741 2267  388   38  93180 wire   1001 0: clk
                                                   251840 act       2 1: atkb
15.2%Sys   5.3%Intr 18.9%User  0.0%Nice 60.6%Idl    60244 inact       3: sio1
|    |    |    |    |    |    |    |    |    |      24996 cache       4: sio0
========++>>>>>>>>>>                                70852 free        7: ppc0
                                                          daefr   128 8: rtc
Namei         Name-cache    Dir-cache                     prcfr    47 9: pcm0
    Calls     hits    %     hits    %                   4 react   155 11: nvi
     2463     2463  100                                   pdwak       12: psm
                                       36 zfod            pdpgs       13: npx
Disks   ad0   cd0 pass0                36 ofod            intrn       14: ata
KB/t   0.00  0.00  0.00                   %slo-z    61024 buf         15: ata
tps       0     0     0               167 tfree        39 dirtybuf
MB/s   0.00  0.00  0.00                             35360 desiredvnodes
% busy    0     0     0                             12471     8409s

ie all the process appear to be in disk wait, rather than sleep.

The hitching appears to synchronise with some disk activity but
it's not like the disk gets hammered every 10 seconds.

-- 
Daniel O'Connor software and network engineer
for Genesis Software - http://www.gsoft.com.au
"The nice thing about standards is that there
are so many of them to choose from."
  -- Andrew Tanenbaum
GPG Fingerprint - 5596 B766 97C0 0E94 4347 295E E593 DC20 7B3F CE8C

Received on Thu Sep 08 2005 - 01:27:09 UTC

This archive was generated by hypermail 2.4.0 : Wed May 19 2021 - 11:38:43 UTC