Seconded.

Note that we've changed the default umask to be slightly more strict since this issue was posted, yet the change has not affected ypkg and eopkg negatively as far as I can tell.

I cannot reproduce:

Add note on the 30s compile-time default.

Edit commit to have longer lines

Noop to satisfy arc diff.

• Re-add default timeout of 30s per request
• Fix type in conf.d/ file

Add secure boot cruft and bump the package properly.

If you decide to land this, be aware that it _deletes_ the secure boot certificate (/usr/lib/systemd/boot/solus-mok.cer) , so you will likely want to re-add that?

Add DefaultDeviceTimeoutSec=30s to defaults

@ReillyBrogan : Do we want to switch this to 128?

@anoraktrend: Ping? See above.

Do you want us to disable your account then?

Were you the user with the very old GF GT220 generation card using the nouveau driver?

Thanks for the report.

Update build to newest nightly as of this writing.

AIUI, the way phabricator works makes it a _Bad Idea™_ to outright delete an account, as deleting an object in the store has all sorts of bad side effects.

Add missing pspec and abi files

Closing as wontfix, as it's not complete.

I think we'll want to chmod -Rc 0770 /root during baselayout creation (to be consistent with /home/foo permissions), since it looks like the wrapped mkdir command _is_ indeed using a default umask of 0755 like I idly speculated a day or two ago.

GJ @Harvey

I'll add this to the 4.4 task list and begin investigating ASAP.

Re-assigning this to Jacek as this will be one of his responsibilities when he joins.

GPU: Radeon RX Vega 64
Desktop: KDE X11 + Wayland

You're it!

You're it!

I use Gemini every once in a while, so I can take amfora.

In T10555#201776, @16CharactersWork wrote:

Well at the moment it is not so bad where it becomes unusable, but weirdly every once in a while it starts doing this.

Yeah, that's a SATA-600 SSD.

In D14197#243628, @silke wrote:

This does not fix the issue with ncdu for me. It looks to me like zig uses instructions from the build environment, and it probably shouldn't.
I can reproduce the issue as follows:

1. Build zig on a machine with AVX512 (like the buildserver or my desktop).
2. Transfer and install package on a machine without AVX512.
3. Run zig and get a crash with 'Illegal instruction'.

Assigning this to Joey for him to fill out the preferred result format and good video test cases

If we are planning to begin offering (optional) Wayland sessions, this is high priority IMO.

Is this still an issue for you?

Fixed in R3571:c32cd293ee7873125cb76ab9ddf803712c7acac0 (not in R1966 yet).

OK, workaround confirmed working.

@Harvey pointed me towards https://unix.stackexchange.com/questions/361214/disable-gdm-suspend-on-lock-screen