gnome-keyring bug and fix

Recently, I noticed that in a GNOME session the gnome-keyring process was constantly taking up 100% of a cpu core. Naturally, I killed it, and it was fine, so I went upon my business. But then I noticed it on my next boot, and then I noticed it on another machine with a different version of software and before I knew it, I was scouring the net for answers.

Turns out, an old bug had resurfaced in some changes to code within glib2 and had shown itself in both 43.1 and 42.6. The bug was introduced in glib2 1.74.1 (but was occurring as well on 2.72.4- likely the same changes were backported), so I naturally downgraded the libs and moved on while waiting for a patch to come.

A patch has finally come, so I’ve updated the gcs43.1 and also upgraded gcs42.6 (aarch64) repos with the new packages. As well as a few other updates on aarch64 for webkit2gtk 2.38.2 and libadwaita 1.2.0, in preparation to likely ship GNOME 43 for aach64… eventually… I’m in no rush and 42 runs pretty well without many library updates on that port. Before 44 comes out, I’m sure I’ll ship it (it’s plenty build-able on your own anyway). 42 needed those webkit updates, as I can’t imagine building that on a normal aarch64 machine (M1 to the rescue!).

I’ve also updated the liveslak with -current and the new glib2. That’s all for the rest of the week out of me. Enjoy!

One thought on “gnome-keyring bug and fix”

Leave a Reply

Your email address will not be published. Required fields are marked *

This site uses Akismet to reduce spam. Learn how your comment data is processed.