Hello!
After a couple of huge tarball extracts (`make extract' in jdk14 and jdk15)
I noticed, things are a little slower. During the extracts, the mouse was
moving with visible jerks. Indeed, the system seems VERY busy:
   11 users    Load  1.18  1.52  1.40                  Jul 20 00:14
Mem:KB    REAL            VIRTUAL                     VN PAGER  SWAP PAGER
        Tot   Share      Tot    Share    Free         in  out     in  out
Act 1060360  105932  1341624   145252  151016 count
All 1750432  115908  8911872   174532         pages
                                                          zfod   Interrupts
Proc:r  p  d  s  w    Csw  Trp  Sys  Int  Sof  Flt        cow    1277 total
     4     1146      1819  345 443k 1640  672      266572 wire      4 irq1: atkb
                                                  1204008 act    1004 irq0: clk
84.7%Sys   0.0%Intr 15.3%User  0.0%Nice  0.0%Idl   241536 inact       irq6: fdc0
|    |    |    |    |    |    |    |    |    |      62232 cache   128 irq8: rtc
==========================================>>>>>>    88784 free
irq9: acpi
                                                          daefr   113 irq12: psm
Namei         Name-cache    Dir-cache                     prcfr       irq15: ata
    Calls     hits    %     hits    %                     react       irq16: ahc
     3030     3030  100                                   pdwak    26 irq17: pcm
                                                          pdpgs       irq18: fwo
Disks   da0   cd0   cd1 pass0 pass1 pass2                 intrn       irq19: ohc
KB/t   4.00  0.00  0.00  0.00  0.00  0.00          218912 buf         irq27: skc
tps       2     0     0     0     0     0              21 dirty     2 irq29: cis
MB/s   0.01  0.00  0.00  0.00  0.00  0.00          100000 desiredvnodes
% busy    0     0     0     0     0     0           92043 numvnodes
The machine is idle and is not doing anything in user-space according to both
top and vmstat's "pigs" display.
Yet it is noticably slower. Trying to compile something pushes the load above
2. What is it doing?
This is a single-CPU Opteron running:
	FreeBSD 5.4-STABLE #0: Fri Jun 10 09:11:30 EDT 2005 amd64
The box has 2Gb of RAM, but NO SWAP.
Any ideas? Thanks!
	-mi
Mikhail T. wrote this message on Wed, Jul 20, 2005 at 00:19 -0400:> After a couple of huge tarball extracts (`make extract' in jdk14 and jdk15) > I noticed, things are a little slower. During the extracts, the mouse was > moving with visible jerks. Indeed, the system seems VERY busy: > > 11 users Load 1.18 1.52 1.40 Jul 20 00:14 > > Mem:KB REAL VIRTUAL VN PAGER SWAP PAGER > Tot Share Tot Share Free in out in out > Act 1060360 105932 1341624 145252 151016 count > All 1750432 115908 8911872 174532 pages > zfod Interrupts > Proc:r p d s w Csw Trp Sys Int Sof Flt cow 1277 total > 4 1146 1819 345 443k 1640 672 266572 wire 4 irq1: atkb^^^^ well, I'd say 443k syscalls/time interval isn't doing nothing... [...]> The machine is idle and is not doing anything in user-space according to both > top and vmstat's "pigs" display.the problem is that your machine is sooooo fast that all of the processes that are running are exiting before they can be observed by pigs or top (or even accumulate enough cpu time to be worth showing)...> Yet it is noticably slower. Trying to compile something pushes the load above > 2. What is it doing? > > This is a single-CPU Opteron running: > > FreeBSD 5.4-STABLE #0: Fri Jun 10 09:11:30 EDT 2005 amd64 > > The box has 2Gb of RAM, but NO SWAP.run ps lax a few times, and notice which process is fork bombing your box by seeing which process has the most changing children... (i.e. the ppid, 3rd column, of the process that isn't in the next run)... sort -n +1 -2 + diff will help find which ones... ps lax | sort -n +1 -2 > tmpa; sleep 2; ps lax | sort -n +1 -2 > tmpb; diff tmpa tmpb look at the ppid (3rd column) of any new or missing processes, and you probably have your culprit... -- John-Mark Gurney Voice: +1 415 225 5579 "All that I will do, has been done, All that I have, has not."
> Subject: OS suddenly VERY busy > > Hello! > > After a couple of huge tarball extracts (`make extract' in> jdk14 and jdk15) I noticed, things are a little slower. > During the extracts, the mouse was moving with visiblejerks.> Indeed, the system seems VERY busy: > > The machine is idle and is not doing anything inuser-space> according to both top and vmstat's "pigs" display. > > Yet it is noticably slower. Trying to compile something > pushes the load above 2. What is it doing? > > This is a single-CPU Opteron running: > > FreeBSD 5.4-STABLE #0: Fri Jun 10 09:11:30 EDT 2005amd64> > The box has 2Gb of RAM, but NO SWAP. > > Any ideas? Thanks! >Irq 15 is your IDE disk. The slowest part of a server will always be the disk, this is your bottleneck. I tried working with jdk14: (*and still cannot install it, will you post your install sequence please?*) Some of the files where large. I noticed nothing really performance wise when working with with some of the 50M plus tar files. I do have dual 242's and a 64bit 3Ware card (in a 64Bit slot) that has harware XOR's in it so my cpu's never have to do direct disk work per se. One of the install processes recommends you have 1.4Gb of free work space. This may also be an issue. Why no swap partition? Was this intentional? Also how much RAM? I have 1Gb. What does df -h say? I have mine like this. Raid 1 Array with SATA drives. I'm worried I made /var too small... Filesystem Size Used Avail Capacity Mounted on /dev/twed0s1a 989M 93M 817M 10% / devfs 1.0K 1.0K 0B 100% /dev /dev/twed0s1d 3.9G 458K 3.6G 0% /tmp /dev/twed0s1f 178G 18G 146G 11% /usr /dev/twed0s1e 39G 882M 35G 2% /var
?????? 20 ?????? 2005 01:34, John-Mark Gurney ?? ????????:> > This is a single-CPU Opteron running: > > > > ??????FreeBSD 5.4-STABLE #0: Fri Jun 10 09:11:30 EDT 2005 amd64 > > > > The box has 2Gb of RAM, but NO SWAP.> run ps lax a few times, and notice which process is fork bombing your > box by seeing which process has the most changing children...If found the culprit, and it was not fork bombing. It was vlc (multimedia/vlc-devel), that was not supposed to do anything. On the list of interrupts I posted originally, observe the 40 to 50 irqs/sec on pcm. Once I told vlc to exit, the system returned back to normal... Does not seem right :-( -mi