• When synchronet web server is being accessed, log file switches back a

    From Keyop Troy@VERT to GitLab issue in main/sbbs on Thu Jan 8 02:54:14 2026
    open https://gitlab.synchro.net/main/sbbs/-/issues/1045

    I've noticed in my logs that when a request is made to web.magnumbbs.net (my webv4 instance), sbbs switches back and forth from http-web.magnumbbs.net-YYYY-MM-DD and http-YYYY-MM-DD.

    Here is a snapshot of my logs:

    https://pastebin.com/ZrC8Nd5y

    Here are the files:

    -rw-rw----. 1 sbbs sbbs 0 Jan 8 00:00 http-2026-01-08.log
    -rw-rw----. 1 sbbs sbbs 698204 Jan 8 10:51 http-web.magnumbbs.net-2026-01-08.log

    I'm not sure why sbbs keeps on going back and forth between logfiles only to write nothing out to the http-YYYY-MM-DD file.

    Surely this is just wasting resources?

    ---
    þ Synchronet þ Vertrauen þ Home of Synchronet þ [vert/cvs/bbs].synchro.net
  • From Rob Swindell@VERT to GitLab note in main/sbbs on Thu Jan 8 03:02:35 2026
    https://gitlab.synchro.net/main/sbbs/-/issues/1045#note_8095

    I'm not sure why it's going back and forth (Deuce would know) between hostname'd log file and without. I think if you disable VHOST support in the web server, that eliminates the hostname from the http log filenames.

    But if you don't want/use the http log files, turn off the web server "HTTP_LOGGING" option.

    ---
    þ Synchronet þ Vertrauen þ Home of Synchronet þ [vert/cvs/bbs].synchro.net
  • From Keyop Troy@VERT to GitLab note in main/sbbs on Thu Jan 8 03:03:11 2026
    https://gitlab.synchro.net/main/sbbs/-/issues/1045#note_8096

    When googlebot crawls my web pages, about 1 request per second, my sbbs process jumps up to about 50% cpu utilisation.

    3088619 sbbs 20 0 2902248 990.6m 10528 S 53.5 8.4 97:15.22 sbbs

    53.5% in this case.

    ---
    þ Synchronet þ Vertrauen þ Home of Synchronet þ [vert/cvs/bbs].synchro.net
  • From Keyop Troy@VERT to GitLab note in main/sbbs on Thu Jan 8 03:04:10 2026
    https://gitlab.synchro.net/main/sbbs/-/issues/1045#note_8097

    Will look into these - ty.

    Should we assign the ticket to Deuce so he can take a look?

    ---
    þ Synchronet þ Vertrauen þ Home of Synchronet þ [vert/cvs/bbs].synchro.net
  • From Keyop Troy@VERT to GitLab note in main/sbbs on Thu Jan 8 03:06:43 2026
    https://gitlab.synchro.net/main/sbbs/-/issues/1045#note_8098

    Re VHOSTS, I have an issue with someone "domain fronting" my web server, so I need VHOSTS to stop this issue.

    ---
    þ Synchronet þ Vertrauen þ Home of Synchronet þ [vert/cvs/bbs].synchro.net
  • From Deuc¨@VERT to GitLab note in main/sbbs on Sat Jan 10 14:30:06 2026
    https://gitlab.synchro.net/main/sbbs/-/issues/1045#note_8120

    It switches to http-YYYY-MM-DD when there's no host information... often when garbage is sent, but also when weird requests are made, here's mine from that date:

    ```
    cat /synchronet/sbbs/data/logs/http-2026-01-08.log
    45.194.92.11 - - [08/Jan/2026:00:36:02 -0500] "CONNECT www.roblox.com:443 HTTP/1.1" 501 - "-" "-"
    162.142.125.216 - - [08/Jan/2026:00:53:46 -0500] "PRI * HTTP/2.0" 501 - "-" "-" 178.22.24.64 - - [08/Jan/2026:01:09:12 -0500] "" 400 - "-" "-"
    64.23.158.207 - - [08/Jan/2026:02:23:03 -0500] "PROPFIND / HTTP/1.1" 501 - "-" "-"
    64.62.197.152 - - [08/Jan/2026:03:13:06 -0500] "" 400 - "-" "-"
    79.124.40.174 - - [08/Jan/2026:03:25:17 -0500] "¨" 400 - "-" "-"
    46.161.50.108 - - [08/Jan/2026:03:27:16 -0500] "¨" 400 - "-" "-"
    46.161.50.108 - - [08/Jan/2026:03:27:17 -0500] "¨" 400 - "-" "-" 216.180.246.226 - - [08/Jan/2026:09:01:48 -0500] "" 400 - "-" "-" 216.180.246.226 - - [08/Jan/2026:09:06:39 -0500] "" 400 - "-" "-" 206.168.34.217 - - [08/Jan/2026:11:22:16 -0500] " " 400 - "-" "-" 206.168.34.217 - - [08/Jan/2026:11:22:21 -0500] "PRI * HTTP/2.0" 501 - "-" "-" 199.45.155.89 - - [08/Jan/2026:11:38:19 -0500] "PRI * HTTP/2.0" 501 - "-" "-" 138.68.243.197 - - [08/Jan/2026:12:09:49 -0500] "" 400 - "-" "-"
    138.68.243.197 - - [08/Jan/2026:12:09:49 -0500] "" 400 - "-" "-"
    20.169.107.47 - - [08/Jan/2026:12:57:49 -0500] "MGLNDD_68.55.12.216_80" 501 - "-" "-"
    93.174.93.12 - - [08/Jan/2026:15:41:32 -0500] "o" 400 - "-" "-"
    87.251.75.163 - - [08/Jan/2026:16:27:47 -0500] "" 400 - "-" "-"
    167.94.138.46 - - [08/Jan/2026:17:46:35 -0500] "" 400 - "-" "-"
    206.168.34.120 - - [08/Jan/2026:18:49:29 -0500] "" 400 - "-" "-"
    206.168.34.120 - - [08/Jan/2026:18:49:47 -0500] "PRI * HTTP/2.0" 501 - "-" "-" 206.168.34.120 - - [08/Jan/2026:18:50:36 -0500] "" 400 - "-" "-"
    198.235.24.112 - - [08/Jan/2026:20:08:11 -0500] "" 400 - "-" "-"
    198.235.24.112 - - [08/Jan/2026:20:08:11 -0500] "" 400 - "-" "-"
    3.143.33.63 - - [08/Jan/2026:20:19:51 -0500] "" 400 - "-" "-"
    3.143.33.63 - - [08/Jan/2026:20:20:04 -0500] "" 400 - "-" "-"
    3.143.33.63 - - [08/Jan/2026:20:20:09 -0500] "SSH-2.0-Go" 501 - "-" "-" 101.198.0.150 - - [08/Jan/2026:21:27:35 -0500] "0" 400 - "-" "-" 185.224.128.136 - - [08/Jan/2026:21:51:24 -0500] "" 400 - "-" "-" 134.199.175.93 - - [08/Jan/2026:22:41:29 -0500] "" 400 - "-" "-"
    134.199.175.93 - - [08/Jan/2026:22:41:30 -0500] "" 400 - "-" "-"
    ```

    ---
    þ Synchronet þ Vertrauen þ Home of Synchronet þ [vert/cvs/bbs].synchro.net
  • From Deuc¿@VERT to GitLab note in main/sbbs on Sat Jan 10 14:35:11 2026
    https://gitlab.synchro.net/main/sbbs/-/issues/1045#note_8121

    As far as wasting resources is concerned, it was much worse when it left every file open... it could use a MRU cache or something, but that would make the happy path slower.

    The actual cost is fairly low, `fclose()`, `strcpy()`, `fopen()`, `lprintf(LOG_INFO, ...)`. the `lprintf()` is likely the most costly operation there. I wouldn't object to it being moved to LOG_DEBUG though.

    ---
    þ Synchronet þ Vertrauen þ Home of Synchronet þ [vert/cvs/bbs].synchro.net
  • From Deuc¿@VERT to GitLab note in main/sbbs on Sat Jan 10 14:35:25 2026
    https://gitlab.synchro.net/main/sbbs/-/issues/1045#note_8121

    As far as wasting resources is concerned, it was much worse when it left every file open... it could use a MRU cache or something, but that would make the happy path slower.

    The actual cost is fairly low, `fclose()`, `strcpy()`, `fopen()`, `lprintf(LOG_INFO, ...)`. the `lprintf()` is likely the most costly operation there. I wouldn't object to it being moved to `LOG_DEBUG` though.

    ---
    þ Synchronet þ Vertrauen þ Home of Synchronet þ [vert/cvs/bbs].synchro.net