-
I'm using Nuclei as an embedded lib. This code snippet will give you an idea how I'm using it:
The goal, of course, being that we automatically check In my integration tests, I have a server listening only on I'm reading down into the Nuclei code to see where/how this might be happening, and theorized that the HostErrorsCache might be configured incorrectly, but that does not seem to be the problem as it is configured with 30 For completeness, I'm using the hosterrorscache from Nuclei: Perhaps you can point me the right direction? |
Beta Was this translation helpful? Give feedback.
Replies: 3 comments
-
I have a clue. I put a
The first makes sense; So I studied deeper into Nuclei code, wondering what state the Dialer keeps. So I changed the line from So there is some odd way dialer state is kept when the first attempt against the same host is a |
Beta Was this translation helpful? Give feedback.
-
Some notes... I put some
|
Beta Was this translation helpful? Give feedback.
-
At this point I'm wondering if the I've added some print statements, like:
And the output occasionally shows:
... which implies the cache has been populated with an entry that lacks an Answer ( Which of course results in returning FWIW, my testing is inside a lambda, inside a localstack container, inside docker-compose. I've opened a proposal PR: projectdiscovery/fastdialer#51 |
Beta Was this translation helpful? Give feedback.
At this point I'm wondering if the
hm *hybrid.HybridMap
infastdialer/dialer.go
.Dialer is thread (errr "go routine") safe, or if the cache needs more protection against bad entries.I've added some print statements, like:
And the output occasionally shows:
... which implies the cache has been populated with…