1. Forum
  2. FidoNet
  3. Z2REGION25

  • Polling and sending all echo/files crash

    From Vincent Coen@2:250/1 to All on Thu Feb 10 16:26:38 2022
    Hello All!

    At the moment I am sending all Files and Echos that come in as Crash ASAP.

    Great idea but I can get many incoming echo packets per day and I mean 'many'.

    So far it is not a problem but I can see in the not to distance future that the
    system has not completed all crash polls before new echos have arrived and my system is totally tied up doing that .


    Now for many of you downlinks, you are not in front of your computer all the time - you are at work or have better things to do than read echo's then there is after say midnight when we are all in bed (at least in Europe).

    So I am considering doing this in the event it gets even busier :

    To mark all systems currently on Crash to Normal (some are only on Hold because
    system cannot get through even after trying 30 times) then

    Say once per hour and if needed to change to 30 minutes changing these Normal flow files to Crash having temporarily putting the mailer on Stop/Hold then releasing the system to rerun.

    At the moment the system is sending from one echo in a packet out crash and sometimes may be up to three echos likewise. This takes 3 - 5 seconds once there is a connection then the system rests for 15-20 seconds before the sending to the next system in the queue. In between these are incoming calls that are dropping of more traffic so it is getting a never ending stream for very small traffic where the processing of such is more than the sending out.

    So doing it this way say in blocks of 60 or 30 minutes it will reduce the load allowing outbound traffic to be larger.

    IF you have any thoughts on please respond other wise I will assume every one is happy with this change - mind you I still have to work out the best way of implementing it with my mailer (mbse) :)



    Vincent

    --- Mageia Linux v8 X64/Mbse v1.0.7.24/GoldED+/LNX 1.1.5-b20180707
    * Origin: Air Applewood, The Linux Gateway to the UK & Eire (2:250/1)
  • From Tristan Greaves@2:250/11 to Vincent Coen on Thu Feb 10 17:39:03 2022
    Re: Polling and sending all echo/files crash
    By: Vincent Coen to All on Thu Feb 10 2022 04:26 pm

    So I am considering doing this in the event it gets even busier :

    To mark all systems currently on Crash to Normal (some are only on Hold because
    system cannot get through even after trying 30 times) then

    Say once per hour and if needed to change to 30 minutes changing these Normal
    flow files to Crash having temporarily putting the mailer on Stop/Hold then
    releasing the system to rerun.

    Fine from my side. Equally happy to revert to polling, say, once per hour if this would help in any way as an alternative.

    Thanks for your work on this!

    Tristan.
    --- SBBSecho 3.14-Linux
    * Origin: Extricate BBS - bbs.extricate.org (2:250/11)