Error connecting to CertStream - Handshake status 521 None
Hello there,
Since today, Tuesday December 5 04:27:42 UTC 2023, certstream-python can not connect to certstream:
# pip install certstream
Collecting certstream
Downloading certstream-1.12.tar.gz (9.6 kB)
Preparing metadata (setup.py) ... done
Requirement already satisfied: termcolor in /usr/lib/python3/dist-packages (from certstream) (1.1.0)
Requirement already satisfied: websocket-client>=0.58.0 in /usr/local/lib/python3.11/dist-packages (from certstream) (0.59.0)
Requirement already satisfied: six in /usr/lib/python3/dist-packages (from websocket-client>=0.58.0->certstream) (1.16.0)
Building wheels for collected packages: certstream
Building wheel for certstream (setup.py) ... done
Created wheel for certstream: filename=certstream-1.12-py3-none-any.whl size=4728 sha256=9d6e4c0ec98f7beb282199f9511f3bbf58d9d79fb2f2ac49d0fc71130483301d
Stored in directory: /sec/root/.cache/pip/wheels/88/02/89/d6a96184779630f955590660a4d2ad038ff39c9c4e7a77d9c1
Successfully built certstream
# certstream
[ERROR:certstream] 2023-12-05 11:04:32,394 - Error connecting to CertStream - Handshake status 521 None - Sleeping for a few seconds and trying again...
And a visit to https://certstream.calidog.io/ returns a CloudFlare 521 error:
Any insight about the recovery of this awesome service ?
Cheers.
Noticed here as well, relying heavily on this service. Hope Calidog manages to get the server back up!
Do you know some alternative instances of certstream-server not being the official one "certstream.calidog.io" ?
Another solution could be to setup ourself an instance of a certstream server, either the official one; or the promising alternative https://github.com/d-Rickyy-b/certstream-server-go
EDIT :
As I am using certstream in a github action workflow, I added a custom instanciation of a local alternative server with this snippet:
cd /tmp/
wget -nv "https://github.com/d-Rickyy-b/certstream-server-go/releases/download/v1.4.0/certstream-server-go_1.4.0_linux_amd64" && chmod u+x ./certstream-server-go_1.4.0_linux_amd64
cat > /tmp/config.yaml <<EOL
webserver:
listen_addr: "127.0.0.1"
listen_port: 8080
full_url: "/full-stream"
lite_url: "/"
domains_only_url: "/domains-only"
cert_path: ""
cert_key_path: ""
prometheus:
enabled: false
listen_addr: "0.0.0.0"
listen_port: 8080
metrics_url: "/metrics"
expose_system_metrics: false
real_ip: false
whitelist:
- "127.0.0.1/8"
EOL
nohup ./certstream-server-go_1.4.0_linux_amd64 > nohup.out 2> nohup.err < /dev/null &
certstream --url ws://127.0.0.1:8080
See https://github.com/appleboy/ssh-action/issues/40#issuecomment-602325598 for the nohup magic.
Good stuff, thanks for pointing us in the right direction!
Really appreciate this thread! Saved me a bunch of time after needing to migrate as well.
Seems like the certstream service has been repaired.
Service down again, cf. https://github.com/CaliDog/certstream-python/issues/62
Server went down from 3rd Feb 2024. Seems subscription was for only 1 year exactly.
Many thanks for this alternative. Go server works fine. You can change ip bind as you need. below will keep it for local use only:
webserver:
listen_addr: "127.0.0.1"
listen_port: 8080
ofc, monitoring over prometheus is optional. Try it out locally and should solve your problem.