OSEC

Neohapsis is currently accepting applications for employment. For more information, please visit our website www.neohapsis.com or email hr@neohapsis.com
 
From: Brian Somers (brianAwfulhak.org)
Date: Tue May 01 2001 - 22:04:20 CDT

  • Messages sorted by: [ date ] [ thread ] [ subject ] [ author ]

    > Hi folks;
    >
    > Brian Somers was kind enough to give me a suggestion on my last posting
    > (thanks :)) I tried adding "set openmade active 6" per Brian, which
    > appeared to work, at least once. After that, it worked only once more
    > over approximately 50 attempts.
    >
    > I realize that the "problem" is on the server side, not with the client.
    > But Brian's software seems so flexible, that even I should be able to get
    > it to talk with any (unreasonable) ppp server, if I just knew the magic
    > lines.
    >
    > Note:
    >
    > 1) Windows client, which can always connect, takes at least one minutes to
    > negotiate protocols AFTER username/password validation and IP address
    > assignemnt. Given this, I cranked "openmode active" to 45 which still did
    > not help. The one time it worked, it was set at 6.
    >
    > 2) Conection was usually dropped after three "ReqSent." Based on this, I
    > cranked lcpretry, ccpretry, and ipcpretry to "5 10 10 " each. This did not
    > help.
    >
    > Attached is my ppp.conf, log of one succesfull connection, and typical log
    > of unsuccesfull attempts.
    >
    > Any suggestions (yes, I am looking into cable modem, but in the mean
    > time...), links or further references would be apprciated.
    >
    > TIA,
    >
    > Rob

    Hi,

    With the successful connection, the peer sends it's first configure
    request as soon as it sees yours, so I think the openmode delay is
    the right thing to do here.

    Perhaps changing your LCP retry timeout to something large would help
    - say ``set lcpretry 30''. If the server is having problems (maybe
    due to being overloaded), it's not going to help if we end up
    flooding it with unprocessed LCP requests (although it should throw
    away second and subsequent repeats because of the duplicate sequence
    number).

    -- 
    Brian <brianAwfulhak.org>                        <brian[uk.]FreeBSD.org>
          <http://www.Awfulhak.org>                   <brian[uk.]OpenBSD.org>
    Don't _EVER_ lose your sense of humour !