I do regular automated updates. For anything requiring human intervention like the xz thing I trust Lemmy and YouTube to keep me updated. No dedicated news source because if I were to freak out about every new vulnerability found I wouldn’t be able to sleep at night.
Why does the xz thing require human intervention?
If you had it on a computer that is accessible via SSH from the internet you should proceed under the assumption that it was compromised. Which means you should reinstall from a safe medium and change your keys and passwords.
Mailing list provided by my distro. https://lists.debian.org/debian-security-announce/
Didn’t know this existed. Just subscribed. Thanks
you just made me look for my distros security list, I never even thought of that!
Fediverse and RSS mostly.
My distribution (archlinux) notifies of critical vulnerabilities that require user action. There’s a news mailing list.
After that I rely on social network (Mastodon mostly) or lemmy for news, as vulnerabilities often get some conversation. Apart from that, software i’m really interested in I also follow through RSS so I get news when they update for their vulnerabilities -that is when the vulnerabilities are not self inflicted as the xz case-.
Arch Linux (like some other distros) also has a security tracker: https://security.archlinux.org/
I didn’t really consider that there are feeds for such things, especially for my distro(s). Embarrassing, but it means you helped making me safer!
I’m now subscribed to the Debian security list, seeing as all my servers run Debian. I just had unattended upgrades with Mail logs before.
Seeing my colleagues, I fear that the answer from them is “That’s the neat part, you don’t!”
Same here. Our servers are so out of date that we might not have a version of xz with any commits from Jia Tan at all.
I don’t think up-to-date Debian stable even got it before it was discovered. No prod servers should be affected
I rely on notifications from
glsa-check
or my distro’s package manager. I was notified about a problem withxz-utils
on Thursday evening, but didn’t see anyone post about it until Friday morning.glsa-check
is a command-line tool included with the gentoolkit package in Gentoo Linux. Its primary function is to scan your system for installed packages that are vulnerable according to Gentoo Linux Security Advisories (GLSAs). GLSAs are official notifications from the Gentoo security team about security vulnerabilities that affect packages in the Gentoo repository.I tend to find out about vulnerabilities before it hits the news outlets from the rss feed at https://seclists.org/oss-sec/
Other than that, I’ve got a bunch of other security feeds I follow and also have automated updates with just about everything.
the worst ones end up on https://slashdot.org/ e.g.:
https://m.slashdot.org/story/426644
I read it like twice per day. However, my software updates should fix most automatically without me even knowing what was going on.
I just use
unattended-upgrades
and forget about itSame for the RPM ecosystem: yum-cron and walk away. Been that way for almost 25 years.
Having been involved with OS Security in the middle of my career, I also still watch feeds like I used to; just, different ones, now.
i subscribed for fedora mailist a few days ago and their talk awas helpful for me to notice that i was one of the affected, just subscribe to your distro blog/mail/etc
Found out about the xz one on Lemmy. Years ago I was briefly subscribed to Bugtraq but that was too much. Now I’m subscribed to a few OS specific security announcement mailing lists.
You can watch rss feeds to follow all CVEs like Microsoft’s https://api.msrc.microsoft.com/update-guide/rss
NIST used to have an rss feed for CVEs but deprecated it recently. They still have other ways you can follow it though https://nvd.nist.gov/vuln/data-feeds
Or if you just want to follow CVEs for certain applications you can host/subscribe to something like https://www.opencve.io/welcome which allows you to filter CVEs from NIST’s National Vulnerability Database (NVD)
I rely on Lemmy and in
pacman -Syyu
everydayThen, what does a package maintainer rely on?Edit: I’m so dumb. It’s obvious they’d check original developer’s repo or issue tracker. I’m sorry
I don’t know… I guess in mailing lists and pages like RSS feed from main enterprises like SuSE, Red Hat and Canonical
You can track this kind of stuff on Mastodon also, join into a security instance (like https://infosec.exchange/explore) or start following them from another instance.
Mainly Phoronix and Lemmy.