Re: 6.0 VFS/VM over-runs. Was: Apparent strange disk behaviour in 6.0

From: Julian Elischer (julian_at_elischer.org)
Date: 07/29/05

  • Next message: Kris Kennaway: "Re: here we go again with the weird kernel build errors..."
    Date: Fri, 29 Jul 2005 12:40:31 -0700
    To: Marc Olzheim <marcolz@stack.nl>
    
    

    Marc Olzheim wrote:

    >On Thu, Jul 28, 2005 at 02:13:44PM -0700, Julian Elischer wrote:
    >
    >
    >>mount a large partition that has some room.. hopefully several times
    >>the space of all RAM.
    >>.
    >>
    >>do:
    >>
    >>dd if=/dev/zero of=/${mountpoint}/bigfile bs=128k count=1000000 & iostat 1
    >>
    >>
    >>notice that dd dies with "not enough swapspace" after some number of
    >>seconds.
    >>(there is no swap attached). also notice that until the crash, IO speed
    >>was max for the
    >>disk.
    >>
    >>then add swap and do it again.
    >>
    >>notice that at teh point where dd previously crashed, the swap disk
    >>springs into life and disk IO on your main disk goes into the toilet.
    >>notice that dd gets swapped out. (!!!!)
    >>
    >>when the disk recovers enough, dd pages back in and creates more IO but
    >>in teh mean time there is no IO
    >>going on to the disk.. (watch the LEDs).
    >>
    >>
    >
    >Just to be clear, this happens on 5-STABLE, 6-STABLE and 7-CURRENT.
    >
    >

    it doesn't appear to hit 5.4. at least not for my small ATA based
    workstation.

    >Marc
    >
    >
    _______________________________________________
    freebsd-current@freebsd.org mailing list
    http://lists.freebsd.org/mailman/listinfo/freebsd-current
    To unsubscribe, send any mail to "freebsd-current-unsubscribe@freebsd.org"


  • Next message: Kris Kennaway: "Re: here we go again with the weird kernel build errors..."

    Relevant Pages

    • Re: swap size for 2gb 2 processors?
      ... out a partition magic run on a 120 gb disk, ... The rule of thumb is 2 * RAM size, in your case that would be 4G. ... price for a swap partition has remained reasonably constant. ...
      (comp.os.linux.misc)
    • Re: [ck] swap prefetching merge plans
      ... After I close some memory demanding aplication that caused huge portion of the memory to be swapped out to disk, the time needed to swith back to firefox or eclipse is really really shorter. ... And software engeneerers and programmers are really working hard to eat all your RAM even if you think that "now you have really much". ... I have so much swap because I am compiling I my packages on tmpfs. ...
      (Linux-Kernel)
    • Re: headsup: swap_pager.c
      ... disk could not keep up with page-in demand. ... "add" another disk while the box is spasticly trying to swap ... around a 200M plus working-set into 32M physical RAM. ... bash me more...I still dont understand why you are trying to fix ...
      (freebsd-arch)
    • Re: VM system - memory seems to be running out, when its not
      ... No extra RAM allocation would occur until the pages are used. ... amount of RAM as a swap space. ... The VM space can have a dedicated area on disk. ...
      (comp.unix.solaris)
    • Re: why the difference in swap value for vmstat output
      ... > RAM and 56 G of disk swap ... The second is free RAM. ... < This line left intentionally blank to confuse you. ...
      (comp.unix.solaris)