Watchable Security topic/tag + Dashboard Security Update Alert Enhancements

סיכום מבוסס בינה מלאכותית

The discussion centers on improving security alert mechanisms within Discourse. agemo proposes creating a dedicated “Security” watchable tag and implementing high-alert visual indicators, such as “breaking ticker” banners or red box advisories in the admin dashboard, to ensure critical CVEs and zero-day exploits receive immediate attention. agemo argues that current notifications are too subtle, leading to “attention fatigue,” and emphasizes the need to minimize the attack vector by forcing admins to update quickly when exceptional security issues arise.

david suggests watching the #release-notes tag, noting that recent releases frequently contain security fixes. In response, RGJ supports the idea of a specific security-notice tag, highlighting that base image updates also constitute security-relevant releases.

david addresses the challenge of distinguishing critical security issues amidst the high volume of software vulnerabilities. He explains that the Discourse team aims to bundle critical base image updates with core releases to ensure they are covered by standard notification channels like the admin dashboard and automatic emails. David advises admins to treat every update as critical and apply patches immediately, as the speed of vulnerability exploitation leaves no room for delay. The consensus highlights a tension between filtering noise and ensuring urgent security actions are taken promptly.

I thought I posted this suggestion weeks ago when things were getting hotter security wise, but I think I did not.

Suggestion - a Security specific high alert watchable topic/tag would be useful - what do others think?

I would also suggest to bolster this on META:

  • special coloured banner alert (lnews “breaking ticker” banner) reserved for urgent CVE / Zero Day security issues.

ADMIN Dashboard DISTRIBUTED Alerts

I understand there is the red sad face but easy to ignore, sometimes, a more in your face alert that has an exceptional status (is not seen often/rare) may be warranted considering the escalation of security concerns.

  • Red banner alert in admin dash or RED BOX the update advisory

  • Include basic info for e.g. Patching Pending 12 hours (set)

  • Links to advisories / dedicate topic on meta

I recommend watching release-notes. These days, every release includes some form of security fix in Discourse or in upstream dependencies.

Thanks for that pointer. I have watched that tag. Thanks. I do update regularly and understand security fixes are always included.

I would suggest that this is still too ordinary in terms of the challenge at hand. I remember now it was was the PHP exploit news 30 days that kicked this all off, and it hasn’t stopped.

The way I see it is - you want critical alert to get to the admin before the hack does.

Maximising user base being appraised, ready and waiting status, so the update is grabbed asap.

Reducing the time vector period of attack as quickly as possible across discourse deployments.

In summary: Force Critical update watching via the dashboard more effectively

For the exceptional moments, that are becoming all too common.

I’d call these RED BOX updates.

  • exceptional CVE’s
  • critically urgent

The admin won’t see often but will grab attention beyond the smiley face status points or alert icon. Attention fatigue is a real thing. Novelty must remain novel to have its alert force remain potent.

These days we tend to see quite a lot of base image updates as well, and they do constitute a ‘release’.

I think a security-notice watchable tag would be a pretty good and flexible way to deal with all these combined.

The trouble is that there are so many security issues in all software products these days, that it’s hard to distinguish them.

This week’s NGINX and Ruby vulnerabilities seem bad, and have received a lot of press attention. But there are many other recent vulnerabilities in Discourse and other packages which have presented equal or higher risk for self-hosted Discourse communities.

When we update something in the base image which is known to be critical for Discourse (e.g. the libheic updates a few weeks ago), we aim to cut a core version around the same time, so it’s all covered through the same notification system. Most people don’t come to Meta, so releases.discourse.org, the admin dashboard, and the automatic notification emails are the main way we distribute that information.

My advice would be that everyone should be applying every update as soon as possible after they receive a notification. Every one should be considered critical. The speed of vulnerability discovery and potential exploitation is so fast these days - waiting is not an option.