Closed franziskuskiefer closed 2 years ago
This is with the latest version of the canary, yes? What os/config is this running on?
I have been kicking off full runs on an AWS Ubuntu instance and have not seen this issue recently, so I'm curious to reproduce.
The docker image we use for this is this one https://github.com/mozilla/canary-harvester/tree/master/docker. So it's the latest canary installed via pip
on ubuntu:latest
.
Ah, thank you. I can reproduce now. I think this is something we fixed since last release. I'm assuming that these problems will go away if we update the latest TLS Canary release available via pip
.
CR is on PTO at the moment, and I don't feel comfortable pushing a release w/o her approval. I think once she's back in a few days, it should be pretty easy.
👍 Sounds good. Let me know when you released the new version and we'll update the docker image.
TLS Canary v3.2.0 is out. Do these errors persist after pip install -U tlscanary
?
Running our harvester it often happens that at the start of a scan run the first few worker instances drop their results without processing many of the hosts assigned to them. For example:
WARNING Thread-1 tlscanary.worker_pool Worker task dropped results, yielded 8 instead of 50
These results are then missing at the end. @jallmann do you have a link to a run with this error?