15:00:40 <pwhalen> #startmeeting Fedora IoT Working Group Meeting
15:00:40 <zodbot> Meeting started Wed Jul 12 15:00:40 2023 UTC.
15:00:40 <zodbot> This meeting is logged and archived in a public location.
15:00:40 <zodbot> The chair is pwhalen. Information about MeetBot at https://fedoraproject.org/wiki/Zodbot#Meeting_Functions.
15:00:40 <zodbot> Useful Commands: #action #agreed #halp #info #idea #link #topic.
15:00:40 <zodbot> The meeting name has been set to 'fedora_iot_working_group_meeting'
15:00:40 <pwhalen> #chair pwhalen pbrobinson coremodule idiez saypaul djachimo
15:00:40 <zodbot> Current chairs: coremodule djachimo idiez pbrobinson pwhalen saypaul
15:00:40 <pwhalen> #topic roll call
15:00:52 <pwhalen> Good morning, who's here today?
15:01:31 <coremodule> hey pwhalen
15:01:32 * coremodule is here
15:01:37 <pwhalen> howdy coremodule
15:02:08 * miabbott is here
15:03:38 <pwhalen> Looks like we have a quorum!
15:03:45 <pwhalen> #topic 1) ==== Working Group ====
15:04:34 <pwhalen> As I do some debugging of FDO and simplified installer on Fedora, making some notes to add to our docs
15:05:37 <pwhalen> Anything else for admin?
15:05:45 * coremodule has nothing
15:05:54 <pwhalen> #topic 2) ==== Fedora 38 - stable ====
15:05:54 <pwhalen> #info OpenQA testing - https://openqa.fedoraproject.org/tests/overview?distri=fedora&version=38&build=Fedora-IoT-38-20230711.0&groupid=1&groupid=5
15:06:44 <pwhalen> The usual failures, likely Ok after rerunning
15:06:54 <pwhalen> base_reboot_unmount is new?
15:08:21 <pwhalen> Worked ok on the 7th, so if rerunning doesn't help it might need a closer look
15:08:25 <coremodule> I spent some time with the clevis test and got to the same point I got to the time before: I don't have any way to virtualize an aarch64 system with a TPM to make the test.. And I don't have a hardware one. I did test on x86 just to see, but it worked fine.
15:09:02 <coremodule> pwhalen, can you think of a way to test clevis on aarch64 w/o hardware?
15:09:40 <pwhalen> It should work with an emulated tpm? As the x86_64 test does
15:10:42 <coremodule> for x86, I used virtualization to add a virtual TPM, but I don't have a system that can emulate an aarch64 system, at least not without taking several days to get anything done...
15:10:56 <pwhalen> iirc, the emulated TPM didn't work for aarch64 when I looked
15:10:59 <pwhalen> RPi4?
15:11:10 <coremodule> yeah, I'm using the rpi400
15:11:18 <pwhalen> Or hardware in beaker
15:11:42 <pwhalen> the rpi400 was slow with the emulated tpm?
15:12:04 <pwhalen> never tried, but definitely some beefy hw in beaker to try
15:12:27 <coremodule> no, I didn't know I could emulate a tpm with hardware, I thought it was only available in virtualization...
15:13:32 <pwhalen> coremodule: sorry, not understanding- I am talking about virt as well
15:14:29 <coremodule> oh, I think I see what you're saying. use the rpi4 to host a virtual aarch64 iot image and use the emulated tpm there. I can try that. Or like you said, try some of the beefy stuff in beaker. I never use beaker, so it doesn't cross my mind.
15:14:31 <miabbott> there's `swtpm` builds for aarch64...should work to install that and pass it to a virtual machine?
15:15:03 <pwhalen> I can assist if needed
15:15:40 <pwhalen> miabbott: right, I dont think we had that, or there was a bug there previously
15:15:47 <coremodule> Okay, let me try that way. I didn't even consider it, so let me see... I'll write a workstation image now and report in channel later today.
15:16:13 <pwhalen> thanks!
15:16:35 <pwhalen> Anything else for F38?
15:17:10 <coremodule> that's all from me
15:17:15 <pwhalen> #topic 3) ==== Fedora 39 - Rawhide ====
15:17:15 <pwhalen> #info OpenQA testing - https://openqa.fedoraproject.org/tests/overview?distri=fedora&version=39&build=Fedora-IoT-39-20230711.0&groupid=1&groupid=5
15:19:39 <pwhalen> looks similar, the rebase test has been failing for a long time,
15:21:25 <pwhalen> Does anyone recall why the rebase test is failing?
15:22:03 <pwhalen> https://openqa.fedoraproject.org/tests/2012538#step/rpmostree_rebase/19
15:22:17 <coremodule> I don't...
15:23:02 <pwhalen> ok, I'll try that one as I look at other stuff
15:23:31 <miabbott> there was an action from june 7 for coremodule to look into that failure, i believe
15:23:43 <miabbott> https://meetbot.fedoraproject.org/fedora-meeting/2023-06-07/fedora_iot_working_group_meeting.2023-06-07-15.00.html
15:24:17 <coremodule> ah! well. I definitely didn't do that! I'll add it to my list here and see what I can find.
15:24:55 <pwhalen> thanks coremodule, I feel like I should know as well, so its bothering me and I'll try too
15:25:23 <pwhalen> For f39 there are some changes that affect us
15:25:31 <coremodule> cool, I'll be in and out of the iot channel today, we can talk there if we find anything
15:25:34 <pwhalen> #info Fedora 39 Change - EnableFwupdRefreshByDefault
15:25:34 <pwhalen> #link https://fedoraproject.org/wiki/Changes/EnableFwupdRefreshByDefault
15:25:34 <pwhalen> #link https://pagure.io/fedora-iot/issue/52
15:26:11 <pwhalen> pbrobinson updated the iot issue, with no concerns on his end. Does anyone have anything to add?
15:26:52 <pwhalen> We have it installed by default, so its just enabling it. Sounds good to me.
15:28:03 <miabbott> i agree that it is useful...given that we are an iot platform where devices have limited resources, i wonder if the size of the update metadata from LVFS would ever be a problem
15:29:25 <pwhalen> Right, not sure either. Or how often it polls to an update
15:30:56 <pwhalen> #info List of changes for Fedora 39 - https://fedoraproject.org/wiki/Releases/39/ChangeSet
15:31:11 <pwhalen> Anything else  for F39?
15:31:51 <pwhalen> #topic 4) ==== Open Floor ====
15:32:02 <coremodule> pwhalen, will you be at flock this year?
15:32:22 <pwhalen> I might! Just trying to work out the details
15:32:26 <pwhalen> Yourself?
15:32:50 <coremodule> Cool! Yes, I will be. We should get together at some point.
15:33:14 <pwhalen> coremodule: definitely
15:34:04 <pwhalen> Anything else for open floor?
15:34:52 <pwhalen> #endmeeting