It should be fixed and committed real soon, I just have to check if the range is correctly formed in the url dispatch which is trivial (something like 1-lol also triggers a 500. The fix yesterday was just to relaunch the image. I have no testing environment yet and restarting the world is not exactly a good workflow.
I'm trying to install SchemeBBS locally, but the port to MIT Scheme 10.1 is not going to happen yet. Everything seems to be broken, just look at this:
MIT/GNU Scheme running under GNU/Linux
Type `^C' (control-C) followed by `H' to obtain information about interrupts.
Copyright (C) 2019 Massachusetts Institute of Technology
This is free software; see the source for copying conditions. There is NO warranty; not even for
MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.
Image saved on Saturday August 10, 2019 at 6:28:48 PM
Release 10.1.10 || Microcode 15.3 || Runtime 15.7 || SF 4.41 || LIAR/x86-64 4.118
1 ]=> (vector 1 2 3)
;Value:
;Unbound variable: nmv-header?
;To continue, call RESTART with an option number:
; (RESTART 2) => Define nmv-header? to a given value.
; (RESTART 1) => Return to read-eval-print level 1.I spotted it while playing with irregex. 10.1 is far from stable.
Patches are no longer secret: https://gitlab.com/naughtybits/schemebbs
Things like https://textboard.org/prog/39/30---,,,---,,,---30 or https://textboard.org/prog/39/1-lol will no longer trigger a 500.
Thank you again for discovering that awful bug. I believe your stress tests will hit the nginx cache, so they should be safe.