<@apiaseck:matrix.org>
16:30:13
!startmeeting fedora_coreos_meeting
<@meetbot:fedora.im>
16:30:15
Meeting started at 2024-10-16 16:30:13 UTC
<@meetbot:fedora.im>
16:30:15
The Meeting name is 'fedora_coreos_meeting'
<@apiaseck:matrix.org>
16:30:28
!topic roll call
<@jbtrystram:matrix.org>
16:30:42
!hi
<@zodbot:fedora.im>
16:30:44
Jean-Baptiste Trystram (jbtrystram) - he / him / his
<@dustymabe:matrix.org>
16:31:03
!hi
<@hricky:fedora.im>
16:31:04
!hi
<@zodbot:fedora.im>
16:31:05
Hristo Marinov (hricky) - he / him / his
<@zodbot:fedora.im>
16:31:07
Dusty Mabe (dustymabe) - he / him / his
<@aaradhak:matrix.org>
16:32:02
!hi aaradhak
<@zodbot:fedora.im>
16:32:04
Aashish Radhakrishnan (aaradhak)
<@apiaseck:matrix.org>
16:32:22
Hi all! It's been a while since I ran the last meeting, so please be patient with me today :)
<@marmijo:fedora.im>
16:33:02
!hi
<@dustymabe:matrix.org>
16:33:03
Don't worry. You'll do great
<@zodbot:fedora.im>
16:33:03
Michael Armijo (marmijo)
<@apiaseck:matrix.org>
16:33:26
We have a few folks here already, lets start with...
<@apiaseck:matrix.org>
16:33:28
!topic Action items from last meeting
<@apiaseck:matrix.org>
16:34:51
I'm failing to find those at this stage, do we have any?
<@mnguyen:fedora.im>
16:35:22
!hi mnguyen
<@zodbot:fedora.im>
16:35:23
Michael Nguyen (mnguyen)
<@dustymabe:matrix.org>
16:35:39
did we run the meeting last week (I couldn't attend so wasn't there)
<@jlebon:fedora.im>
16:35:44
!hello
<@zodbot:fedora.im>
16:35:45
None (jlebon)
<@apiaseck:matrix.org>
16:36:16
I think jbtrystram ran the last one?
<@marmijo:fedora.im>
16:36:31
I don't remember any action items. The meeting was pretty short last week
<@jbtrystram:matrix.org>
16:37:03
yes I ran the last one. no action items !
<@apiaseck:matrix.org>
16:37:18
Oh, perfect. Thanks JB!
<@apiaseck:matrix.org>
16:37:36
Let's move on to the next...
<@jbtrystram:matrix.org>
16:37:36
Maybe I forgot to send the notes ?
<@apiaseck:matrix.org>
16:38:12
!topic Dependency on fuse-overlayfsincontainers-commonmoved to Suggests
<@apiaseck:matrix.org>
16:38:20
!link https://github.com/coreos/fedora-coreos-tracker/issues/1749
<@apiaseck:matrix.org>
16:38:56
I see travier and dustymabe there. Would you mind introducing this one for us please?
<@dustymabe:matrix.org>
16:39:55
hmm. I guess travier isn't here.
<@ravanelli:matrix.org>
16:40:07
!hi ravanelli
<@jbtrystram:matrix.org>
16:40:11
The issue have a good summary: fuse-overlayfs has been moved from a Recommends to a Suggests in Fedora 40 and later
<@zodbot:fedora.im>
16:40:12
Renata Ravanelli (ravanelli)
<@jbtrystram:matrix.org>
16:40:56
Right now we pull that in explicitly but the podman team want to deprecate it eventually
<@jbtrystram:matrix.org>
16:41:08
In favor of in-kernel overlayfs
<@jbtrystram:matrix.org>
16:41:33
so travier is suggesting we also drop this as part as the f41 rebase
<@jbtrystram:matrix.org>
16:41:53
otherwhise we'll have to carry it until f42
<@jbtrystram:matrix.org>
16:42:09
I think that's a good enough summary to start discussion
<@dustymabe:matrix.org>
16:42:16
- does having fuse-overlayfs installed cause it to be preferred?
<@dustymabe:matrix.org>
16:42:16
<@dustymabe:matrix.org>
16:42:16
I think it would still be a good idea to know the answer to my question in the ticket:
<@dustymabe:matrix.org>
16:44:51
IOW are we exacerbating the problem, or is it only a small subset of users who would even be using the old thing anyway?
<@jbtrystram:matrix.org>
16:45:19
looks like it's going to be a dependency of `container-conmon` until f40 goes EOL : https://github.com/containers/common/pull/2203#pullrequestreview-2371863732
<@dustymabe:matrix.org>
16:46:06
jbtrystram: I don't think that matters too much. Right now we explicitly include it in our package list, so RPM deps don't affect us
<@jbtrystram:matrix.org>
16:46:34
dustymabe: some quick looking suggest it's used for rootless containers
<@jbtrystram:matrix.org>
16:47:15
I guess it boils down to your question yes, Dusty
<@dustymabe:matrix.org>
16:47:17
maybe Jonathan Lebon would know the answer - otherwise yeah, we'd need to do some googling
<@jbtrystram:matrix.org>
16:48:13
And also, are we OK (potentially) breaking containers in the next `testing` release without having a CLHM for a few month before
<@jlebon:fedora.im>
16:48:15
dustymabe: added a comment there
<@jbtrystram:matrix.org>
16:48:22
And also, are we OK (potentially) breaking (some) containers in the next `testing` release without having a CLHM for a few month before
<@dustymabe:matrix.org>
16:48:53
ok agree.
<@dustymabe:matrix.org>
16:49:30
if this is something only a subset of upgrading users are hitting I still don't really see a need to make them migrate early before `podman` itself deprecates it officially
<@dustymabe:matrix.org>
16:50:20
now, i'd prefer the podman team to do that deprecation on a Fedora major boundary and file an associated Fedora change request, but we can't require that
<@jlebon:fedora.im>
16:51:18
dustymabe: seems like a reasonable ask. worth bringing it up to them for f42.
<@dustymabe:matrix.org>
16:51:58
jbtrystram yeah - i'd prefer to have done it when `next` switched to F41 and we sent out the communication email around that
<@dustymabe:matrix.org>
16:51:58
<@dustymabe:matrix.org>
16:51:58
> are we OK (potentially) breaking (some) containers in the next testing release without having a CLHM for a few month before
<@jbtrystram:matrix.org>
16:52:56
oddly, `man container-storage.conf` does list `fuse-overlayfs` as a valid driver
<@dustymabe:matrix.org>
16:53:13
jbtrystram: well, it is a valid driver?
<@dustymabe:matrix.org>
16:53:20
it's not deprecated yet
<@jbtrystram:matrix.org>
16:53:25
sorry
<@jbtrystram:matrix.org>
16:53:28
typo
<@jbtrystram:matrix.org>
16:53:34
does *not* list
<@jbtrystram:matrix.org>
16:53:39
😅
<@dustymabe:matrix.org>
16:53:56
:) - yeah. that's probably part of the deprecation then. discouraging use
<@dustymabe:matrix.org>
16:54:26
apiaseck: I think we are in agreement on this.. want to do a proposed/agreed?
<@jbtrystram:matrix.org>
16:54:45
ah yes, maybe it's not documented but the value would still work
<@apiaseck:matrix.org>
16:57:36
!proposed We contact podman team to do deprecation on a Fedora major boundary and file an associated Fedora change request
<@apiaseck:matrix.org>
16:57:59
Is that more less reflecting the agreement in your opinion folks?
<@jbtrystram:matrix.org>
16:58:55
it should also mention that we will only drop the package when podman drop support ?
<@apiaseck:matrix.org>
16:59:57
!proposed: We contact podman team to do deprecation on a Fedora major boundary and file an associated Fedora change request. We will only drop the package when podman drop support
<@apiaseck:matrix.org>
17:00:21
!proposed We contact podman team to do deprecation on a Fedora major boundary and file an associated Fedora change request. We will only drop the package when podman drop support
<@apiaseck:matrix.org>
17:00:33
The bot doesn't like me today :)
<@apiaseck:matrix.org>
17:00:51
!proposed We contact podman team to do deprecation on a Fedora major boundary and file an associated Fedora change request. We will only drop the package when podman drops support
<@dustymabe:matrix.org>
17:01:05
it's not just you. I think proposed isn't a keyword for the bot, but it's useful for us here in the meeting
<@dustymabe:matrix.org>
17:01:10
+1 vote from my side
<@jlebon:fedora.im>
17:01:10
hmm, ok i think i have more relevant details to add
<@jlebon:fedora.im>
17:03:23
i think i might've been confused in my comment. fuse-overlayfs is used for _mounting_ the overlay only in some situations, but the backend would still be overlay. it used to be used all the time for rootless in older kernels, but hasn't been for a while and i think that's the rationale behind changing it to weak. i'm not sure if there are any plans to "deprecate" it per se since AIUI it's still _sometimes_ useful
<@apiaseck:matrix.org>
17:03:36
That's the "patience" I was referring to in the beginning of our meeting lol
<@jlebon:fedora.im>
17:03:58
so... probably should contact the podman team, but for us right now i think the proposal is more "we will not drop fuse-overlayfs for f41" ?
<@dustymabe:matrix.org>
17:04:02
in that case I don't understand why we would remove the package
<@jlebon:fedora.im>
17:04:45
we aren't actively removing it. it's just us not pulling in weak deps
<@dustymabe:matrix.org>
17:05:12
i mean if we can say that it's 100% not being used anymore then yeah, let's remove it
<@dustymabe:matrix.org>
17:05:34
i.e. there would be no downsides of removing it
<@jlebon:fedora.im>
17:05:54
the `podman system reset` comment still stands i think. anyone using it as their "mount_program" would be broken.
<@dustymabe:matrix.org>
17:06:30
would they have had to configure that explicitly? or they would just have had to have an old system upgraded over many releases?
<@jlebon:fedora.im>
17:06:56
i'm not sure
<@dustymabe:matrix.org>
17:07:19
yeah. so let's just say we won't do it for F41 and we'll work with the podman team to determine when would be appropriate to remove it
<@jbtrystram:matrix.org>
17:07:33
looking at https://github.com/containers/podman/issues/16899 if it was used at some point it will be used until the storage is wiped
<@dustymabe:matrix.org>
17:08:24
we did require people to wipe their storage in one of our major version upgrades (I think because podman v5 was not backwards compat) so most of our systems probably are already doing the right thing
<@dustymabe:matrix.org>
17:08:50
if that's the case let's just slot in the change for F42 and let it roll out next cycle
<@dustymabe:matrix.org>
17:10:20
and we can do a CLHM just in case
<@dustymabe:matrix.org>
17:11:37
!proposed We won't remove the package for F41 but will work with the podman team to determine the appropriate time to do the deprecation. This may be on a Fedora major boundary with an associated Fedora change request.
<@jlebon:fedora.im>
17:13:21
ack
<@jbtrystram:matrix.org>
17:13:48
nice ! +1
<@apiaseck:matrix.org>
17:13:48
+1 from me, any other votes?
<@marmijo:fedora.im>
17:13:52
+1
<@aaradhak:matrix.org>
17:13:55
+1
<@apiaseck:matrix.org>
17:15:15
I think we have enough votes above to hit `agree` on this?
<@ravanelli:matrix.org>
17:15:46
+1
<@apiaseck:matrix.org>
17:16:30
!agreed We won't remove the package for F41 but will work with the podman team to determine the appropriate time to do the deprecation. This may be on a Fedora major boundary with an associated Fedora change request.
<@apiaseck:matrix.org>
17:17:26
Can't see any other topics on the list, is there anything else we would like to discuss (we still have some time)
<@apiaseck:matrix.org>
17:17:32
?
<@dustymabe:matrix.org>
17:18:33
I was going to chat about what we should do about the VMWare bug, but it looks like we might be able to get a fix for that soon
<@dustymabe:matrix.org>
17:18:48
I was going to possibly propose we revert to the older version of GRUB in `next`
<@dustymabe:matrix.org>
17:18:56
but looks like we won't have to
<@ravanelli:matrix.org>
17:19:59
I already have a build with the fix to test
<@ravanelli:matrix.org>
17:20:27
will test it soon
<@jbtrystram:matrix.org>
17:20:39
Renata Ravanelli++
<@zodbot:fedora.im>
17:20:40
No Fedora Accounts users have the @ravanelli:matrix.org Matrix Account defined
<@jbtrystram:matrix.org>
17:21:00
ravanelli++
<@dustymabe:matrix.org>
17:21:01
Renata Ravanelli: that's a scratch build, right?
<@zodbot:fedora.im>
17:21:02
jbtrystram gave a cookie to ravanelli. They now have 12 cookies, 1 of which were obtained in the Fedora 40 release cycle
<@ravanelli:matrix.org>
17:21:20
Yes
<@dustymabe:matrix.org>
17:21:23
it will be really nice if we can get an official build (of course after you report your test results) :)
<@ravanelli:matrix.org>
17:22:41
Not sure how much time for the rest of the process to get everything where we need, I'm mean into fedora
<@dustymabe:matrix.org>
17:24:17
really shouldn't be too much of a stretch
<@dustymabe:matrix.org>
17:24:17
<@dustymabe:matrix.org>
17:24:17
we have a patch. all we would need is for the grub maintainers to include it and do a build and then we can fast track
<@dustymabe:matrix.org>
17:24:39
should we do open floor?
<@apiaseck:matrix.org>
17:25:00
!topic Open Floor
<@ravanelli:matrix.org>
17:25:37
Thanks all for helping with the test week!!! 💙
<@dustymabe:matrix.org>
17:25:54
gursewak++
<@zodbot:fedora.im>
17:25:55
No Fedora Accounts users have the @gurssing:matrix.org Matrix Account defined
<@dustymabe:matrix.org>
17:26:00
marmijo++
<@zodbot:fedora.im>
17:26:02
dustymabe has already given cookies to marmijo during the F40 timeframe
<@apiaseck:matrix.org>
17:26:04
gursewak++
<@zodbot:fedora.im>
17:26:05
dustymabe gave a cookie to ravanelli. They now have 13 cookies, 2 of which were obtained in the Fedora 40 release cycle
<@zodbot:fedora.im>
17:26:07
c4rt0 gave a cookie to gursewak. They now have 5 cookies, 1 of which were obtained in the Fedora 40 release cycle
<@apiaseck:matrix.org>
17:26:08
marmijo++
<@zodbot:fedora.im>
17:26:10
c4rt0 gave a cookie to marmijo. They now have 6 cookies, 5 of which were obtained in the Fedora 40 release cycle
<@zodbot:fedora.im>
17:26:26
c4rt0 gave a cookie to ravanelli. They now have 14 cookies, 3 of which were obtained in the Fedora 40 release cycle
<@marmijo:fedora.im>
17:26:34
Yes, +1. thanks to all for helping with the test week!
<@apiaseck:matrix.org>
17:29:41
If there's nothing else to add, thank you very much @all for attending!
<@apiaseck:matrix.org>
17:30:25
!endmeeting