@jwildeboer I have no problem with people mirroring the entire LVFS, but there's no need to re-download the same file every 1h! :)

Does anyone recognise "Nexus/3.83.1-03 (PRO; Linux; 6.5.9-300.fc39.x86_64; amd64; 17.0.16) 0" as anything sensible? Machines with that user agent have been "pinging" the LVFS by downloading firmware on the hour, every hour -- for the last couple of days.

It appears to be an instance of https://github.com/sonatype/nexus-public but I'm very confused why it's abusing the LVFS like this.

RE: https://infosec.exchange/@joshbressers/115565633384087459

Recording this with Josh was a huge amount of fun -- and it was really nice to look back on 10 years+ of #fwupd and the #LVFS. Feedback welcome!

@bodil it's great to hear, thanks!

@RTheren I think German has the best selection of words.

A new #fwupd release just dropped. A little late, but with lots of good stuff. A few nice features, the usual smattering of bugfixes and quite a bit of new hardware supported too. Enjoy!

https://github.com/fwupd/fwupd/releases/tag/2.0.17

Reminder about the badwords project that has all the kinds of words you wouldn't want to see in an application review -- contributions and updates very welcome! https://github.com/hughsie/badwords/blob/master/badwords.csv

Congratulations to the shortlisted nominees in the 6th annual OpenUK Awards 2026, @andypiper, @bboreham, Daniel Gale, @davidtwco, Dermoscopea, @flox, Godfrey Inyama, Jan Faracik, @ctz, Linux Vendor Firmware Service @hughsie, lowRISC CIC, Dr Margaret Hartnett, @manyfold, Ministry of Justice UK Splink Team, OpenActive, Opemipo Disu, Dr. Ravinder Singh, Richard Purdie, Rustls, Sonia Cooper,...

In case anyone wants any more stats, there's a live feed here: https://fwupd.org/lvfs/metrics -- although be careful as lies, damned lies, statistics... e.g. the various Lenovo teams are actually 5 (!) of those "vendors".

That's a lot of zeros! 135 million firmware updates supplied from the LVFS and deployed using #fwupd.

@jibec how about "all translations submitted to weblate before 20th Nov will make the release" -- I'm not sure I even know the RHEL release dates :)

@jibec there's no deadline really; we do one release a month so there's no rush. Ideally we'd get it done within the next release cycle as that's the version going into RHEL 9 and 10.

Looking pretty good so far! Go to https://hosted.weblate.org/projects/fwupd/fwupd/ if you want to help. Thanks!

I've been polishing up the source strings too, but if anything is hard to translate please let me know! Many thanks to all you wonderful people so far.

I've just pushed the button and moved the #fwupd translations from Transifex to @weblate

We're planning a new release in a few days and it would be *wonderful* to get the translations looking beautiful before that.

Thank you all my clever 2nd, 3rd and 4th language friends! All the new functionality is available here: https://hosted.weblate.org/projects/fwupd/fwupd/

@purpleidea we could remap that in fwupd; file a bug and we can discuss there.

@hp I know somebody tried it, but I have no idea if it works or not. I think in my heart-of-hearts I know that the i386 dbx is probably a waste of time -- but I don't want to give up on it if there's one very important use case that I've missed.

If anyone does have any 32 bit UEFI SecureBoot-capable system please let me know (and I'll send you a file to test!) but if there's nobody actually using this hardware it's pointless me writing release notes and uploading a file nobody is going to download.

The 20241101 ia32 firmware was downloaded just 64 times in 2 years, and only once (!) by fwupdmgr -- all the others were just mirror syncs or AI scrapers.

A few hours ago Microsoft dropped a new dbx that resolves CVE-2025-47827. If you're running #fwupd >= 1.9.28 the fix is as simple as "fwupdmgr refresh && fwupdmgr update" although it might take an hour or so for all the CDN endpoints to catch up.

I've uploaded both the x64 and ia32 versions -- but only the former is in the stable remote. The "i386" version is 100% untested as I don't actually own any 32 bit UEFI hardware -- and there doesn't appear to be any in the Red Hat QA system either.

@zhenech I guess phased works -- what would the "number" be called? the phased_factor or phased_modulo or something like that?