Not receiving BGP updates over bgpstreamkafka

Hi,

I’m running Artemis v1.5.0 in kubernetes, deployed via the provided helm chart. I have setup openbmp-collector and openbmp-kafka running on an external VM (using openbmp-kafka as the kafka FQDN and I have created an entry in the artemis-monitor hosts file to map openbmp-kafka to the ip of the VM).

I am publishing data to the collector using mrt2bmp and I can subscribe to the kafka topics from within the monitor container (using the kafka cli) but no route updates are ingested into Artemis. With appropriate use of print statements 😃 it looks like the kafka provider hangs on rec = stream.get_next_record() in bgpstreamkafka.py: https://github.com/FORTH-ICS-INSPIRE/artemis/blob/88626b7e2f1e1835b0f9e93f33adb928d2c9215d/monitor/core/taps/bgpstreamkafka.py#L95

Artemis Monitor Logs:

messaging - 2020-07-20 16:08:36,198 - INFO @ _receive_callback: Initiating monitor...
messaging - 2020-07-20 16:08:36,198 - INFO @ _receive_callback: Starting building monitor prefix tree...
messaging - 2020-07-20 16:08:36,199 - INFO @ _receive_callback: 3 prefixes integrated in monitor prefix tree in total
messaging - 2020-07-20 16:08:36,199 - INFO @ _receive_callback: Finished building monitor prefix tree.
messaging - 2020-07-20 16:08:36,199 - INFO @ _receive_callback: Calculating monitored prefixes for monitor to supervise...
messaging - 2020-07-20 16:08:36,199 - INFO @ _receive_callback: Calculated monitored prefixes for monitor to supervise.
messaging - 2020-07-20 16:08:36,199 - INFO @ _receive_callback: Initiating configured monitoring instances....
messaging - 2020-07-20 16:08:36,206 - INFO @ _receive_callback: All configured monitoring instances initiated.
messaging - 2020-07-20 16:08:36,206 - INFO @ _receive_callback: Monitor initiated, configured and running.
<string> - 2020-07-20 16:08:36,211 - INFO @ <module>: started
<frozen importlib - 2020-07-20 16:08:36,471 - INFO @ _call_with_frames_removed: Loaded configuration from /etc/artemis/logging.yaml
<frozen importlib - 2020-07-20 16:08:36,475 - INFO @ _call_with_frames_removed: Loaded configuration from /etc/artemis/logging.yaml
exabgp_client - 2020-07-20 16:08:36,481 - INFO @ <module>: Loaded configuration from /etc/artemis/logging.yaml
exabgp_client - 2020-07-20 16:08:36,484 - INFO @ <module>: Starting ExaBGP on 10.254.5.34:5000 for /root/monitor_prefixes.json (auto-conf: False)
bgpstreamkafka - 2020-07-20 16:08:36,489 - INFO @ <module>: Loaded configuration from /etc/artemis/logging.yaml
2020-07-20 16:08:37,491 INFO success: monitor entered RUNNING state, process has stayed up for > than 1 seconds (startsecs)

Artemis Bgpstreamkafka config:

prefixes:
  ...
monitors:
  ...
  bgpstreamkafka:
    host: 10.254.5.33
    port: 9092
    topic: openbmp.bmp_raw
asns:
  ...
rules:
  ...

[vkotronis]

@slowr @MarkMckessock we leave this issue open until it is also solved on bgpstream’s side, correct?

[vkotronis]

Closing due to inactivity. Please re-open in case this issue is still relevant.