GlitchTip 3.3.0: Feedback survey, uptime port monitoring, issue fingerprint grouping

Uptime Monitoring

GlitchTip now supports TCP port monitoring and defaults to showing changes to uptime instead of all monitor checks (which can still be accessed). Monitor non-http assets such as a PostgreSQL server. Quickly view history of past disruptions.

Group issues by fingerprint

GlitchTip now supports customized issue grouping by using the Sentry SDK's fingerprint feature.

from sentry_sdk import push_scope, capture_exception

try:
    raise Exception
except Exception as err:
    with push_scope() as scope:
        scope.fingerprint = ["{{ default }}", "extra-foo"]
        capture_exception(err)

The default variable will keep the original grouping included. Omit it to define the grouping yourself. Any event with the same fingerprint array will be grouped together.

Other improvements

• Prometheus metrics now supports a count of organizations and projects, thanks GitLab user @chassing
• We provide better defaults for celery to reduce memory creep. We now recommend setting CELERY_WORKER_AUTOSCALE: 1,3 for small instances (larger ones may want something like 2,10) and CELERY_WORKER_MAX_TASKS_PER_CHILD set to 10000. This helps release memory.
• Discord webhook support, thanks to GitLab user @ThallesP
• Added PikaPods install documentation.

Support GlitchTip by taking our survey.

Please take 2 minutes to fill out our user feedback survey.

• 💸 Donate via Liberapay. We currently recieve $16.63 per week from 13 patrons.
• ⭐ on GitLab.
• ❤️ on AlternativeTo and leave a review. AlternativeTo generates significant website traffic.
• ⭐ on dockerhub where we have over 2 million pulls
• ⏩ Follow on Mastodon - we post here more often and for every patch release.
• Talk to us on Gitter

In the Works

• Public status page for uptime monitoring
• Some experimental work was done using Rust to vastly lower system requirements. At this time, it's purely an experiment. But if you like writing Rust code, please reach out on Gitter.