Closed frost19k closed 2 years ago
This seems to be something really specific from linkfinder tool, it would be probably better to report into LinkFinder repository, anyway I will try to reproduce this and keep you updated.
Thanks!
So, I was making a mistake with my Docker command. There's another parameter --memory-swap
which is set to the same value as --memory
if not explicitly defined. Essentially, I was allotting Docker 48GB of memory + swap - which is more than I have.
When I instead do --memory 24g --memory-swap 32g
the system doesn't hang - instead the function fails & reconftw.sh
moves on to the next thing. But, that's dependant on the output of this line so...
it would be probably better to report into LinkFinder repository
I was wondering, since reconFTW uses a fork of LinkFinder
, should I open the issue there or on the original?
Oh, interesting, thanks for debugging the issue.
Actually reconftw uses the default LinkFinder repository, not a fork, you can check on the installer script that it clones the GerbenJavado's LinkFinder repository.
you can check on the installer script that it clones the GerbenJavado's LinkFinder repository
I'm looking at the script right now... line 86 repos["LinkFinder"]="dark-warlord14/LinkFinder"
(same in dev
)
Ooooops :/ my bad, let me check why are we using this fork instead the original one
Issue seems to be specific to this one domain. The script runs just fine on other targets.
Still have no idea what's causing it though, but I'm gonna go ahead & close this.
Describe the bug Line
1251
which callsLinkFinder
on entries injs/js_livelinks.txt
takes up all system memory, including swap, crashing the system.To Reproduce
Expected behavior Function
jschecks
should complete successfully without crashing the system.Screenshots
Desktop (please complete the following information):
Additional context
As can be seen in the screenshot, 32 gigs of RAM & 16 gigs of swap are (almost) fully consumed. I killed the process before it got to 100% because then the system would become unresponsive - requiring a hard reboot.
I've tried limiting
interlace
to 2 threads - this just slows down the inevitable crash.Limiting the Docker container's threads & memory has no effect.
Seems like
run-mailcap
is the culprit - will look into this further.In the meantime, I'd really appreciate advice on further troubleshooting. I feel like I'm outta ideas.