1. Forum
  2. Fidonet
  3. SYNC SYSOPS

  • MQTT: highwater appears not to be working properly

    From Nigel Reed@1:103/705 to GitLab issue in main/sbbs on Thu Apr 17 18:21:40 2025
    open https://gitlab.synchro.net/main/sbbs/-/issues/910

    ![image](https://gitlab.synchro.net/main/sbbs/uploads/105d09bc7e93fb8c8e4579a2d66266db/image.png){width=279 height=194}

    Since I obviously have terminal users, web users, mail users, I would expect highwater to be higher than 0.
    --- SBBSecho 3.24-Linux
    * Origin: Vertrauen - [vert/cvs/bbs].synchro.net (1:103/705)
  • From Rob Swindell@1:103/705 to GitLab note in main/sbbs on Thu Apr 17 22:43:19 2025
    https://gitlab.synchro.net/main/sbbs/-/issues/910#note_7064

    When any server is restarted or recycle, it's "Highwater" value is republished as 0. Are you servers being recycled frequently maybe?

    If you comment-out this line from src/sbbs3/mqtt.c, does the problem go away? `mqtt_pub_uintval(mqtt, TOPIC_SERVER, "highwater", 0);`
    --- SBBSecho 3.24-Linux
    * Origin: Vertrauen - [vert/cvs/bbs].synchro.net (1:103/705)
  • From Rob Swindell@1:103/705 to GitLab note in main/sbbs on Thu Apr 17 22:43:45 2025
    https://gitlab.synchro.net/main/sbbs/-/issues/910#note_7064

    When any server is restarted or recycled, its "highwater" value is republished as 0. Are your servers being recycled frequently maybe?

    If you comment-out this line from src/sbbs3/mqtt.c, does the problem go away? `mqtt_pub_uintval(mqtt, TOPIC_SERVER, "highwater", 0);`
    --- SBBSecho 3.24-Linux
    * Origin: Vertrauen - [vert/cvs/bbs].synchro.net (1:103/705)
  • From Nigel Reed@1:103/705 to GitLab note in main/sbbs on Fri Apr 18 01:58:24 2025
    https://gitlab.synchro.net/main/sbbs/-/issues/910#note_7065

    The BBS last recycled on 17th according to the logs.

    ![image](/uploads/5e16f0c303896b5826ab5fe58464b382/image.png){width=279 height=144}

    Right now, I have 4 web users, so the highwatermark should be at least 4, and I'm logged in the terminal so that should be at least 1.

    I'll try the suggested line tomorrow since it's getting late.
    --- SBBSecho 3.24-Linux
    * Origin: Vertrauen - [vert/cvs/bbs].synchro.net (1:103/705)
  • From Nigel Reed@1:103/705 to GitLab note in main/sbbs on Fri Apr 18 01:59:16 2025
    https://gitlab.synchro.net/main/sbbs/-/issues/910#note_7066

    FWIW, It seems to be keeping the Services highwatermark OK. I have that as 12. Also 0 for ftp but that could be right since nobody uses it much these days. More testing tomorrow.
    --- SBBSecho 3.24-Linux
    * Origin: Vertrauen - [vert/cvs/bbs].synchro.net (1:103/705)
  • From Rob Swindell@1:103/705 to GitLab note in main/sbbs on Fri Apr 18 12:02:18 2025
    https://gitlab.synchro.net/main/sbbs/-/issues/910#note_7068

    I see the issue and it does appear related to recycling. A fresh start/restart of the process and everything's fine. It's after a server recycle that that the mqtt published highwater can be "stuck" at 0.
    --- SBBSecho 3.24-Linux
    * Origin: Vertrauen - [vert/cvs/bbs].synchro.net (1:103/705)
  • From Rob Swindell@1:103/705 to GitLab issue in main/sbbs on Fri Apr 18 12:30:33 2025
    close https://gitlab.synchro.net/main/sbbs/-/issues/910
    --- SBBSecho 3.24-Linux
    * Origin: Vertrauen - [vert/cvs/bbs].synchro.net (1:103/705)