Also, why does it hang after booting to desktop, leaving me with a blank screen for like 30 seconds or so? Where would i look to find out?
please someone explain this to me or tell me how to find what is doing it...
Yes, it happens to me, too. After the March 28 update that affected mate, the desktop stays frozen with a steady blue, blank screen for about 30 seconds or more before the custom desktop appears.
Is this happening to others as well ? Any ideas on how to fix it ? Thank you very much !
Here is an article that might be of help?
I do not run mate, so I'm unable to test.
https://www.addictivetips.com/ubuntu-linux-tips/fix-a-frozen-mate-linux-desktop/
(https://www.addictivetips.com/ubuntu-linux-tips/fix-a-frozen-mate-linux-desktop/)
Thanks for the article. The "problem" is that MATE 1.27 hangs at start while the article deals with situations when MATE, already started, is frozen. And this is an article about MATE 1.20, somehow outdated.
Have you seen this post here?
https://forum.artixlinux.org/index.php/topic,5244.0.html (https://forum.artixlinux.org/index.php/topic,5244.0.html)
There was a problem with 1.26 - but then it seemed it would be replaced with 1.27 which it has been now, which has a bigger problem. But what's the point of trying to debug 1.26 when it's been replaced now? And what is the point of trying to debug 1.27 when it isn't even in the Arch repos and is a prerelease testing snapshot which might have got in the repos by mistake? Such errors have happened before on occasion. If you wanted to do meaningful dev work on Mate then you'd need to start installing -git versions of relevant Mate packages from the AUR, that 1.27 snapshot is older than the latest 1.26 stable release and this problem could be fixed by now upstream, who knows. Personally I'm just waiting to see what's going to happen and not updating for now.
Only suggestions I have would be trying to temporarily move your mate settings and cache directories or whatever, in case there was something incompatible with the new version, see if any inbuilt Mate desktop logging could be enabled, disable or uninstall any Mate related stuff. The Mate notification daemon has been responsible for historic 30 second delays, although I can't find any current reports of anything like this - probably because no-one is using the pre-release version of Mate! ;D
I'm not sure that the problem is related to MATE. I've been using MATE 1.27 for a while without problems (with tint2 and openbox), but I have freeze issues with kernels later than 5.4. Have you tried the 5.4 or 4.29 kernels (in the AUR)? The 4.29 will still boot (despite a mkinitcpio error MS) if you include the uncompressed firmware in the initramfs.
Thanks for the reply, but I have no idea how to change the kernel used. I don't want to start something complicated and dangerous. Now anyway the PC works, there's just this anomaly at start-up....I can go on like this, but it's actually annoying, and it's a pity because the boot is really lightning fast, in 15 seconds I'm up and running.... But does this anomaly affect so few people? I imagined a much larger response to the post....
Thanks for the reply, but I have no idea how to change the kernel used. I don't want to start something complicated and dangerous. Now anyway the PC works, there's just this anomaly at start-up....I can go on like this, but it's actually annoying, and it's a pity because the boot is really lightning fast, in 15 seconds I'm up and running.... But does this anomaly affect so few people? I imagined a much larger response to the post....
I was also hit by this MATE update (i.e. 30s blank screen). I just downgraded MATE (1.27.0-1 => 1.26.0-1 ) through command:
sudo pacman -S community/mate community/mate-extra
Don't forget to uncomment the line in /etc/pacman.conf:
#IgnorePkg =
IgnoreGroup = mate mate-extra
Now 30s blank screen is gone. It's temporary solution of course.
Fixed with marco-1.27.0-1.5. The solution (https://github.com/mate-desktop/mate-session-manager/issues/147) had been provided years ago by one of our own. It seems the marco "bug" was reintroduced with upstream 1.27.0.
Thanks as always for your hard work!