I believe it's already fixed in amber-phone (we ship gnome-bluetooth 3.34).

We already have GNOME Chess 3.34 in amber-phone.

We already have Taquin 3.34 in amber-phone. Newer versions seem to have problems with the version of Vala used there though.

Opened https://source.puri.sm/Librem5/phoc/-/issues/168

Opened: https://source.puri.sm/Librem5/phosh/-/issues/396

I'm not sure if it was correct to initially set priority of this task to Freedom Issue, because it's possible the issue went unnoticed by team members.

Hmm... sounds a tad expensive to me? Are the other servers located at Hetzner still? Then having storage within their net will be way faster and they also offer pure storage boxes:

Hi @Moop - thanks for reporting this. A new live image has been released. May I ask that you test again?

FWIW Linode provides 250 Gigs for $5 a month. We could get a Terrabyte with them of web addressable storage for USD 20 per month ($240 per year).

So from a quick check it looks like /dev/md3 is the disk that holds our repos;
/dev/md3 1.8T 822G 919G 48% /srv

Something like this maybe:

This IIRC does not work with the Librem laptops since these do not measure the energy properly, judge estimate based on discharge rate. And even if, it is really rough (even Thinkpads do a pretty bad job at it).

I don't have a power meter but I will use powertop. :-)

@jeremiah.foster powetop shows the amount of power drawn from the battery as a rough indicator

Do you have an AC power meter? That's what I used... let the system sit there idle, battery fully charged, look at power consumption. Then just change _one_ thing at a time, let it settle and become idle again, measure again.

So how do we do this test in a way that produces useful data? Do we run it at two different settings on two laptops concurrently? Or should I measure battery level over time at specific intervals with each setting on the same device?

@jeremiah.foster could you check if min_power makes a difference? It seems to save 0.5W here things are fluctuating quite a bit so a longer test would be good.

That's a matter for the sys team. We would need to know how much more space would be needed, how much is available at the hosting site right now and if we require more, how much extra cost this would involve. If this is a matter of just reconfiguring something then sys team can handle this, if it requires something that increases cost please let me know the amount and I will try to get the budget.

Oh good!
Or also not so good, because then I wonder what people do with their laptops not getting more battery life... hmm...

@mak how much disk space would that be ? @jeremiah.foster @nicole can you get us the necessary disk space?

Those packages are available, but only for the latest builds of the development version of PureOS and only for the ones we rebuilt ourselves: https://repo.pureos.net/pureos-debug/dists/
Anything else would require mirroring the debug packages from Debian, and we do not have the disk space for that, currently.

Following Guido's link, I ran

something like https://wiki.archlinux.org/index.php/Power_management#SATA_Active_Link_Power_Management

i wouldn't call it frozen, it's just that the settings and app menu grab all input (i pretty much got used to it and just hit ESC in these situations). This is fixable but not in phosh alone, phoc needs to help here. I'd be great to have a bug on phosh as well since this also affects convergence.

phosh itself is not involved here, if at all it's the compositor. @david.hamner could you file a phoc bug because having bugs in upstream software will likely get lost.