eBPF tracer Crashes for Kernel versions older than 5.5 about kubeshark HOT 7 CLOSED

I confirm this bug. Thanks for reporting this! We are working on a patch release. ETA - a few days.

from kubeshark.

Fixed with: https://github.com/kubeshark/kubeshark/releases/tag/v52.3.0

from kubeshark.

@mateuszdrab quick question while we're investigating this. Did you add any CLI/helm property to enable the new eBPF mechanism? (it's supposed to be disabled by default)

from kubeshark.

@mateuszdrab quick question while we're investigating this. Did you add any CLI/helm property to enable the new eBPF mechanism? (it's supposed to be disabled by default)

Nope, I've not touched anything.

Flux just upgraded the release and since then it wouldn't start.

Current values (chart kubeshark-52.2.30):

license: redacted
tap:
  ingress:
    annotations:
      blackbox-monitoring/module: http_2xx_ca
    className: nginx
    enabled: true
    host: redacted
    tls:
    - hosts:
      - redacted
      secretName: redacted
  namespaces:
  - default

from kubeshark.

Can you try v52.2.39? If the behavior continues, can you look at the logs and see what fails to start? Also, more information about the K8s distribution and cloud provider would help. Thanks!

from kubeshark.

I confirm this bug. Thanks for reporting this! We are working on a patch release. ETA - a few days.

Thanks again for validating it

from kubeshark.

Fixed with: https://github.com/kubeshark/kubeshark/releases/tag/v52.3.0

Confirm the new release fixes the issue.

Thank you ;)

from kubeshark.