Re: Painfully slow exec() on 2.0.35/36?

Miquel van Smoorenburg (miquels@cistron.nl)
18 Sep 1998 18:27:52 +0200


In article <Pine.LNX.4.02.9809180425560.643-100000@moisil.cs.columbia.edu>,
Ion Badulescu <ionut@moisil.cs.columbia.edu> wrote:
>This is what happens: the server is getting a pretty fat feed, so its load
>average hovers usually around 2-3. Even so, it is pretty snappy. However,

Hmm, I also have a few servers with a full feed and load is around 0.5.
Looks like you don't have enough memory..

>a news *reader* connection is established only after a very annoying
>delay, during which innd fork()'s and then exec()'s nnrpd.

Yup. That's why with inn-2.1 you can run nnrpd as a seperate daemon
which _really helps_. Also run ``controlchan'' so that inn doesn't have
to fork for control messages (and it helps against HipCrime attacks).

>strace output shows that it's the exec() syscall itself that's taking an
>unusual amount of time -- between 10 and 50 seconds! While this is

I can confirm this, and the less memory you have free the worse it gets.
On a machine with 128MB it is much more noticable than on a 256MB one.

Mike.

-- 
  "Did I ever tell you about the illusion of free will?"
    -- Sherrif Lucas Buck, ultimate BOFH.

- To unsubscribe from this list: send the line "unsubscribe linux-kernel" in the body of a message to majordomo@vger.rutgers.edu Please read the FAQ at http://www.tux.org/lkml/