tcp_keepalive - should resetting tcp_keepalive_time and _probes to 0 disable?

From: Whit Blauvelt (whit@transpect.com)
Date: Sat Feb 26 2000 - 21:03:13 EST

  • Next message: James Simmons: "Re: Linux console driver maintainer?"

    To answer my earlier question - no. Crashed after a few hours.

    Does anyone please have any advice on how to configure this system so that
    the tcp_keepalive bug is not awakened? It looks like the 2.3 kernel series
    is essentially the same here - or is it? Is there reason to expect better
    luck with 2.3.*? (Does the ipchains compatibility module work fully?) Or
    how far back would I have to go to find a version of the kernel which still
    had the option to compile without this? (Probably before ipchains, right?)

    Thanks again,
    Whit

    ==== previous Q:

    /proc/sys/net/ipv4 contains tcp_keepalive_time and tcp_keepalive_probes.
    I've just reset these (from 7200 and 9 respectively) to 0. Should I expect
    this to avoid invoking the oops in the tcp_keepalive code in the
    2.2.15pre10 kernel?

    Please cc me on the answer. I'm not subscribed (but I'll also check the
    archive via tux).

    Thanks,
    Whit

    -
    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/



    This archive was generated by hypermail 2b29 : Sat Feb 26 2000 - 21:18:47 EST