ngcortes has quit [Remote host closed the connection]
Seirdy0 has quit []
Seirdy has joined #freedesktop
laguneucl has quit [Ping timeout: 480 seconds]
yk has quit [Remote host closed the connection]
yk has joined #freedesktop
yk has quit [Remote host closed the connection]
yk has joined #freedesktop
blue__penquin has joined #freedesktop
ximion has quit []
yk has quit [Remote host closed the connection]
yk has joined #freedesktop
yk has quit [Remote host closed the connection]
yk has joined #freedesktop
danvet has joined #freedesktop
yk has quit [Remote host closed the connection]
yk has joined #freedesktop
i-garrison has quit [singleton.oftc.net synthon.oftc.net]
NoGuest17 has quit [singleton.oftc.net synthon.oftc.net]
karolkro has quit [singleton.oftc.net synthon.oftc.net]
pepp has quit [singleton.oftc.net synthon.oftc.net]
jcristau has quit [singleton.oftc.net synthon.oftc.net]
infernix has quit [singleton.oftc.net synthon.oftc.net]
bl4ckb0ne has quit [singleton.oftc.net synthon.oftc.net]
ndufresne has quit [singleton.oftc.net synthon.oftc.net]
jpnurmi has quit [singleton.oftc.net synthon.oftc.net]
Sumera[m] has quit [singleton.oftc.net synthon.oftc.net]
gnfzdz[m] has quit [singleton.oftc.net synthon.oftc.net]
Mithrandir has quit [singleton.oftc.net synthon.oftc.net]
FLHerne has quit [singleton.oftc.net synthon.oftc.net]
gkiagia has quit [singleton.oftc.net synthon.oftc.net]
kusma has quit [singleton.oftc.net synthon.oftc.net]
fahien2 has quit [singleton.oftc.net synthon.oftc.net]
dcbaker has quit [singleton.oftc.net synthon.oftc.net]
blue_penquin has quit [singleton.oftc.net synthon.oftc.net]
bnieuwenhuizen has quit [singleton.oftc.net synthon.oftc.net]
hikiko has quit [singleton.oftc.net synthon.oftc.net]
danvet has quit [singleton.oftc.net synthon.oftc.net]
jstein has quit [singleton.oftc.net synthon.oftc.net]
mupuf has quit [singleton.oftc.net synthon.oftc.net]
psychon has quit [singleton.oftc.net synthon.oftc.net]
Brocker has quit [singleton.oftc.net synthon.oftc.net]
DragoonAethis has quit [singleton.oftc.net synthon.oftc.net]
MrCooper has quit [singleton.oftc.net synthon.oftc.net]
r00tobo has quit [singleton.oftc.net synthon.oftc.net]
Thymo has quit [singleton.oftc.net synthon.oftc.net]
ocrete has quit [singleton.oftc.net synthon.oftc.net]
tomeu7 has quit [singleton.oftc.net synthon.oftc.net]
dev[m] has quit [singleton.oftc.net synthon.oftc.net]
romangg has quit [singleton.oftc.net synthon.oftc.net]
arekm has quit [singleton.oftc.net synthon.oftc.net]
pq has quit [singleton.oftc.net synthon.oftc.net]
vsyrjala has quit [singleton.oftc.net synthon.oftc.net]
whot has quit [singleton.oftc.net synthon.oftc.net]
heftig[m] has quit [singleton.oftc.net synthon.oftc.net]
demarchi has quit [singleton.oftc.net synthon.oftc.net]
gitlab-bot has quit [singleton.oftc.net synthon.oftc.net]
V has quit [singleton.oftc.net synthon.oftc.net]
jadahl has quit [singleton.oftc.net synthon.oftc.net]
bilboed has quit [singleton.oftc.net synthon.oftc.net]
tintou has quit [singleton.oftc.net synthon.oftc.net]
Haaninjo has quit [singleton.oftc.net synthon.oftc.net]
Sheogorath[m] has quit [singleton.oftc.net synthon.oftc.net]
mceier has quit [singleton.oftc.net synthon.oftc.net]
nroberts has quit [singleton.oftc.net synthon.oftc.net]
tanuk has quit [singleton.oftc.net synthon.oftc.net]
zzag has quit [singleton.oftc.net synthon.oftc.net]
tinywrkb has quit [singleton.oftc.net synthon.oftc.net]
scorpion2185[m] has quit [singleton.oftc.net synthon.oftc.net]
nirbheek has quit [singleton.oftc.net synthon.oftc.net]
bentiss has quit [singleton.oftc.net synthon.oftc.net]
cwabbott has quit [singleton.oftc.net synthon.oftc.net]
narmstrong has quit [singleton.oftc.net synthon.oftc.net]
eric_engestrom has quit [singleton.oftc.net synthon.oftc.net]
jjardon has quit [singleton.oftc.net synthon.oftc.net]
yk has quit [singleton.oftc.net synthon.oftc.net]
tobtoht has quit [singleton.oftc.net synthon.oftc.net]
bcarvalho has quit [singleton.oftc.net synthon.oftc.net]
immibis has quit [singleton.oftc.net synthon.oftc.net]
alatiera has quit [singleton.oftc.net synthon.oftc.net]
karolherbst has quit [singleton.oftc.net synthon.oftc.net]
pendingchaos has quit [singleton.oftc.net synthon.oftc.net]
shadeslayer has quit [singleton.oftc.net synthon.oftc.net]
veganaiZe has quit [singleton.oftc.net synthon.oftc.net]
pzanoni has quit [singleton.oftc.net synthon.oftc.net]
keithp has quit [singleton.oftc.net synthon.oftc.net]
emersion has quit [singleton.oftc.net synthon.oftc.net]
reillybrogan has quit [singleton.oftc.net synthon.oftc.net]
marcheu has quit [singleton.oftc.net synthon.oftc.net]
glisse has quit [singleton.oftc.net synthon.oftc.net]
rg3igalia has quit [singleton.oftc.net synthon.oftc.net]
Kayden has quit [singleton.oftc.net synthon.oftc.net]
alanc has quit [singleton.oftc.net synthon.oftc.net]
jcline has quit [singleton.oftc.net synthon.oftc.net]
kem has quit [singleton.oftc.net synthon.oftc.net]
agd5f has quit [singleton.oftc.net synthon.oftc.net]
anholt has quit [singleton.oftc.net synthon.oftc.net]
sumits has quit [singleton.oftc.net synthon.oftc.net]
halfline has quit [singleton.oftc.net synthon.oftc.net]
dri-logger has quit [singleton.oftc.net synthon.oftc.net]
ebassi has quit [singleton.oftc.net synthon.oftc.net]
SanchayanMaity has quit [singleton.oftc.net synthon.oftc.net]
melissawen has quit [singleton.oftc.net synthon.oftc.net]
__tim has quit [singleton.oftc.net synthon.oftc.net]
daniels has quit [singleton.oftc.net synthon.oftc.net]
samuelig has quit [singleton.oftc.net synthon.oftc.net]
Ford_Prefect has quit [singleton.oftc.net synthon.oftc.net]
blue__penquin has quit [singleton.oftc.net synthon.oftc.net]
Seirdy has quit [singleton.oftc.net synthon.oftc.net]
mixfix41 has quit [singleton.oftc.net synthon.oftc.net]
aswar002 has quit [singleton.oftc.net synthon.oftc.net]
eck has quit [singleton.oftc.net synthon.oftc.net]
shbrngdo has quit [singleton.oftc.net synthon.oftc.net]
sauce has quit [singleton.oftc.net synthon.oftc.net]
jrayhawk has quit [singleton.oftc.net synthon.oftc.net]
asdflkj has quit [singleton.oftc.net synthon.oftc.net]
Lyude has quit [singleton.oftc.net synthon.oftc.net]
ds` has quit [singleton.oftc.net synthon.oftc.net]
austriancoder has quit [singleton.oftc.net synthon.oftc.net]
thaytan has quit [singleton.oftc.net synthon.oftc.net]
ceyusa has quit [singleton.oftc.net synthon.oftc.net]
tchar has quit [singleton.oftc.net synthon.oftc.net]
swick has quit [singleton.oftc.net synthon.oftc.net]
eh5 has quit [singleton.oftc.net synthon.oftc.net]
kchibisov has quit [singleton.oftc.net synthon.oftc.net]
ishitatsuyuki has quit [singleton.oftc.net synthon.oftc.net]
jstultz has quit [singleton.oftc.net synthon.oftc.net]
ivyl has quit [singleton.oftc.net synthon.oftc.net]
orumin has quit [singleton.oftc.net synthon.oftc.net]
robclark has quit [singleton.oftc.net synthon.oftc.net]
Ekho has quit [singleton.oftc.net synthon.oftc.net]
benh has quit [singleton.oftc.net synthon.oftc.net]
ezequielg has quit [singleton.oftc.net synthon.oftc.net]
znahry610 has quit [singleton.oftc.net synthon.oftc.net]
tobtoht has joined #freedesktop
Brocker has joined #freedesktop
swick has joined #freedesktop
ceyusa has joined #freedesktop
ezequielg has joined #freedesktop
tchar has joined #freedesktop
znahry610 has joined #freedesktop
jrayhawk has joined #freedesktop
ds` has joined #freedesktop
Lyude has joined #freedesktop
thaytan has joined #freedesktop
austriancoder has joined #freedesktop
yk has joined #freedesktop
danvet has joined #freedesktop
asdflkj has joined #freedesktop
aswar002 has joined #freedesktop
mixfix41 has joined #freedesktop
eck has joined #freedesktop
shbrngdo has joined #freedesktop
blue__penquin has joined #freedesktop
sauce has joined #freedesktop
Seirdy has joined #freedesktop
Kayden has joined #freedesktop
pepp has joined #freedesktop
agd5f has joined #freedesktop
kem has joined #freedesktop
immibis has joined #freedesktop
jcristau has joined #freedesktop
ishitatsuyuki has joined #freedesktop
mceier has joined #freedesktop
mupuf has joined #freedesktop
anholt has joined #freedesktop
psychon has joined #freedesktop
alatiera has joined #freedesktop
eh5 has joined #freedesktop
Thymo has joined #freedesktop
DragoonAethis has joined #freedesktop
keithp has joined #freedesktop
r00tobo has joined #freedesktop
rg3igalia has joined #freedesktop
ocrete has joined #freedesktop
ndufresne has joined #freedesktop
tomeu7 has joined #freedesktop
dcbaker has joined #freedesktop
kusma has joined #freedesktop
jpnurmi has joined #freedesktop
bnieuwenhuizen has joined #freedesktop
blue_penquin has joined #freedesktop
gkiagia has joined #freedesktop
hikiko has joined #freedesktop
Sumera[m] has joined #freedesktop
gnfzdz[m] has joined #freedesktop
Mithrandir has joined #freedesktop
FLHerne has joined #freedesktop
jadahl has joined #freedesktop
fahien2 has joined #freedesktop
heftig[m] has joined #freedesktop
gitlab-bot has joined #freedesktop
demarchi has joined #freedesktop
whot has joined #freedesktop
vsyrjala has joined #freedesktop
Sheogorath[m] has joined #freedesktop
romangg has joined #freedesktop
dev[m] has joined #freedesktop
V has joined #freedesktop
tintou has joined #freedesktop
arekm has joined #freedesktop
pq has joined #freedesktop
bilboed has joined #freedesktop
pendingchaos has joined #freedesktop
tinywrkb has joined #freedesktop
shadeslayer has joined #freedesktop
nroberts has joined #freedesktop
scorpion2185[m] has joined #freedesktop
narmstrong has joined #freedesktop
jjardon has joined #freedesktop
nirbheek has joined #freedesktop
zzag has joined #freedesktop
eric_engestrom has joined #freedesktop
cwabbott has joined #freedesktop
bentiss has joined #freedesktop
sumits has joined #freedesktop
halfline has joined #freedesktop
reillybrogan has joined #freedesktop
glisse has joined #freedesktop
marcheu has joined #freedesktop
ebassi has joined #freedesktop
dri-logger has joined #freedesktop
kchibisov has joined #freedesktop
benh has joined #freedesktop
SanchayanMaity has joined #freedesktop
orumin has joined #freedesktop
Ekho has joined #freedesktop
Ford_Prefect has joined #freedesktop
melissawen has joined #freedesktop
ivyl has joined #freedesktop
__tim has joined #freedesktop
jstultz has joined #freedesktop
robclark has joined #freedesktop
samuelig has joined #freedesktop
daniels has joined #freedesktop
pzanoni has joined #freedesktop
NoGuest17 has joined #freedesktop
alanc has joined #freedesktop
Haaninjo has joined #freedesktop
i-garrison has joined #freedesktop
jstein has joined #freedesktop
veganaiZe has joined #freedesktop
infernix has joined #freedesktop
MrCooper has joined #freedesktop
bcarvalho has joined #freedesktop
jcline has joined #freedesktop
karolkro has joined #freedesktop
emersion has joined #freedesktop
bl4ckb0ne has joined #freedesktop
tanuk has joined #freedesktop
karolherbst has joined #freedesktop
<mupuf>
bentiss: interesting! I think I'll make it the default then
<bentiss>
mupuf: also I was wondering if gitlab.fd.o/freedesktop/boot2container would not be a better place than gfx-ci
<mupuf>
why freedesktop?
<bentiss>
cause it's more general purpose than just gfx
<mupuf>
that is absolutely true
<mupuf>
So, I don't think we want another group for it, and it should not be staying under my name, so either it goes to gfx-ci or it goes to freedesktop
<mupuf>
I guess one or the other really depends on how you want to see it: a testing component, or an infrastructure tool
<mupuf>
that's the beauty of it, it is so generic :)
<bentiss>
daniels: any opinion on the above? ^^
alanc has quit [Remote host closed the connection]
alanc has joined #freedesktop
veganaiZe has quit [Quit: veganaiZe]
yk has quit [Remote host closed the connection]
yk has joined #freedesktop
<bentiss>
mupuf: also, what's the recommended way for b2c to retrieve the pipeline status? with a post-container? or can we make qemu exit with a different eroor code?
<mupuf>
bentiss: maybe we efi we could inspect a variable?
<mupuf>
but what I have been doing for the test suite of b2c is just to grep the output
<mupuf>
which is what I do on real machines too... using a serial console
<bentiss>
hmm... I'd rather not grep
<mupuf>
yep :D
<bentiss>
well, I'll probably just use a post container that takes the environment variable and writes it to a file
<mupuf>
yeah, that sounds like a better option
<bentiss>
talking about that, I forgot to forward the environment from the job to qemu/b2c
<mupuf>
actually, if we you were to expose the cache to the container, you could write the result there of the execution
<mupuf>
but that may also not be the best
<mupuf>
at least, if you re-use the same container, but with a different entrypoint, then you should be able to just write the status to the file
<mupuf>
to a file*
laguneucl has joined #freedesktop
<bentiss>
well, I'd rather use a small container (alpine) that I know what tools it has to dump the value in a file on the cache. Given that I sync the cache with the caller of the b2c script, I can get that quite easily
<mupuf>
ah, of course!
<bentiss>
because re-using the same container means that you might not even have sh in it
<bentiss>
like some containers only have one binary
<mupuf>
yes, it makes a lot of sense for a generic solution like you are trying to achieve!
<bentiss>
mupuf: also I was wondering if we could not think at a generic way in the templates to run b2c on bare metal
<bentiss>
it would be nice if we could just include ci-template, and use fdo.b2c.bare-metal and done
<mupuf>
runner <-> serial console <-> b2c <-> test container?
<mupuf>
or do you want to have the runner straight in b2c? :D
<bentiss>
yeah, but maybe that would require fdo to have a generic runner or farm to contact that the bare metal test HW would contact
<bentiss>
I like b2c being small: it's just the initrd
<mupuf>
you "yeah" was for the runner straight in b2c?
<mupuf>
your yeah*
<bentiss>
no, for the first thing
<bentiss>
assuming the following:
<mupuf>
ah, for the first thing, you pretty much need what I am making, don't you?
<bentiss>
yes, but I wonder how much maintainance it would be to have a freedesktop scheduler (or runner) for that
<mupuf>
A gateway, serial consoles, dnsmasq, b2c, containers
<mupuf>
right now, we have the runners exposed directly by the gateway.... but this is not a requirement
<bentiss>
don't know if it is feasible, but it would be nice if we could have a common gateway, and people (from fdo) could connect their farm/HW to it without having to invest in a dedicated gateway server
<mupuf>
well, the executor now is running dnsmasq directly, but if we were to use the old boot service, then the executor could be anywhere
<bentiss>
nnot sure why the executor needs to run dnsmasq and what the 'old boot service' is
<mupuf>
people will need a gateway locally, no matter what. They need storage for caches, and they need something to read the serial consoles
<mupuf>
but it does not have to be a big machine, indeed
<bentiss>
mupuf: I am just brain dumping my thoughts here
<mupuf>
well, I am glad you are dumping your brain, because what you want to achieve is exactly what I want: Make testing as simple and cheap as possible!
<bentiss>
storage could be in the machine itself, no?
<mupuf>
and seriously, I would *LOVE* for us to find a cheap recipe to deploy HW testing!
<mupuf>
storage could indeed be mostly in the test machines, depending on the type of testing
<mupuf>
but if you have 10 machines, it is a little stupid that they would all need to download the same containers
<mupuf>
when a proxy registry is so simple to deploy
<mupuf>
speaking about deployment, we have been working hard on making it as simple as possible: Literally just plug a USB stick in a gateway, and everything gets set up
<mupuf>
and machines are auto-discovered and tested before being exposed in gitlab
<bentiss>
agree, but I am thinking more of the personal hobbyist who just wants his machine to be used for testing (probably on his own project only)
<mupuf>
yeah, I think we need a mode with a RPi as a gateway
<bentiss>
and yeah, you will surely say that we still need a controller/gateway locally to poll for tasks and assign runs locally
<bentiss>
exactly what I was thinking :)
<bentiss>
actually, that mode could come as a ci-template job
<mupuf>
well, you *need* a gateway for *one* thing: the serial console
<bentiss>
the rpi is setup as a gitlab runner attached to your project (and taged)
<mupuf>
yep :)
<mupuf>
oh, another thing needed though: power cutting
<bentiss>
and when ever you start the job, the rpi pulls the gateway container, and set ups the serial console and boots up the attached machine(s)
<mupuf>
we can always use a simpler power cutter, like my previous project: just press the power button
<bentiss>
PoE is an other option
<mupuf>
PoE works for boards, indeed.
<bentiss>
cause generally, when you have a PoE capable switch, you can control whether or not to enable it per port
<mupuf>
or Ikea smart plugs
<mupuf>
well, our executor has an abstraction for PDUs, so we can just add a class that implements the needed methods
<mupuf>
and we can add different switches there
<mupuf>
I think the cheapest solution might be the Ikea smart plugs though
<mupuf>
for small scale deployments
<bentiss>
that's why you went to Finland?
<mupuf>
Valve and Intel are using datacenter-grade PDUs
<bentiss>
it was all on purpose!
<mupuf>
lol, Ikea is swedish :p
<bentiss>
I know :)
<mupuf>
but the Ikea I would buy the stuff from is the first one ever constructed outside of Sweden :D
<bentiss>
see???
<mupuf>
yep ;)
<mupuf>
But seriously, I think there is a loooot of potential here, and the team has been super excited about it too \o/
<mupuf>
but of course, the more, the merrier!
<mupuf>
bentiss: do you think this is something I should talk about at XDC? I don't want to be boring people out with my CI talks every year
<bentiss>
mupuf: why not? and this year XDC is virtual, so if people don't want to attend, they can do so :)
<mupuf>
good point
<bentiss>
but honestly, if we can simplify setting up farms, and manage to get a nouveau farm out of it, you'll get interest I think
<mupuf>
yeah
<mupuf>
and I think the potential of self-discovery and self-healing would make a compelling case for "make testing cheap and easy"
<mupuf>
because hiring a technician is expensive
<mupuf>
especially keeping them on call :D
<bentiss>
yep
<bentiss>
mupuf regrarding tradfri (ikea), is it using zigbee or some known or unknown protocol?
* bentiss
can't find a simple "here is ahow to control tradfri from rpi)
<mupuf>
zigbee
<mupuf>
I found one such tutorial
<bentiss>
ok, so you need a zigbee gateway
<bentiss>
s/gateway/module
<mupuf>
zigbee2mqtt is apparently what's needed if you don't want to use the gateway
<mupuf>
and yeah, you need an adapter
<bentiss>
ok, found some blogs about it too know
<bentiss>
now
<mupuf>
tradfri only makes sense for a couple of machines. After that, a second-hang PDU is actually better value
<bentiss>
FTR, I haven't realized those plugs were that cheap, 8€ here
<mupuf>
yes, and Ikea is *everywhere*
* mupuf
has been working on remotely-controlling PCs since 2013
<mupuf>
and custom PCBs are just a giant pain
<mupuf>
although now, with PCB assembly for cheap on JLCPCB and others, the problem isn't as bad anymore
<bentiss>
oh, realy? I am shocked :)
<mupuf>
:p
<mupuf>
live and learn.... about supply chains and reproducibility!
Sheogorath[m] has quit []
heftig[m] has quit [Quit: Bridge terminating on SIGTERM]
scorpion2185[m] has quit []
dcbaker has quit [Quit: Bridge terminating on SIGTERM]
demarchi has quit [Quit: Bridge terminating on SIGTERM]
kusma has quit [Quit: Bridge terminating on SIGTERM]
nirbheek has quit [Quit: Bridge terminating on SIGTERM]
tinywrkb has quit []
blue_penquin has quit [Quit: Bridge terminating on SIGTERM]
dev[m] has quit []
gkiagia has quit [Quit: Bridge terminating on SIGTERM]
tintou has quit [Quit: Bridge terminating on SIGTERM]
Sumera[m] has quit []
gnfzdz[m] has quit []
MrCooper has quit [Remote host closed the connection]
<bentiss>
shadeslayer: IIRC the attachements are uploaded only in case of failures
<bentiss>
but I honestly never used that feature personally, so can't be of much help here
MrCooper has quit [Quit: Leaving]
MrCooper has joined #freedesktop
gitlab-bot has quit [Ping timeout: 480 seconds]
yk has quit []
<shadeslayer>
bentiss: you mean the attachments are only rendered in case of failures?
<shadeslayer>
or uploaded as in uploaded to gitlab? because I definitely see the images in the artifacts
<daniels>
bentiss: yeah I'd be fine with freedesktop/ namespace, or we could just rename gfx-ci, since it's really at its core about doing complex test execution on real hardware devices, which ... isn't gfx-specific, it's just that we're almost the only ones doing it. GStreamer are starting to do a bit more tho which is really nice :)
<daniels>
I haven't seen any of the ci-templates b2c work since I'm not subscribed, but if it lets me run a container image inside qemu without having to manually invoke qemu, prefix all my commands with 'vmctl' and have to do double shell quoting inside YAML as well, then WOOHOO yes I'm all for it
<bentiss>
(it is probably not the final API from ci-templates point of view)
arifu has quit [Remote host closed the connection]
<daniels>
bentiss: hmmmmm, I wonder if we can get around that by having vmctl be the container entrypoint, and forward the cmd into the VM
<mupuf>
daniels: do you have an example where you are using vmctl and shell quoting?
<daniels>
no, we got around it by using virtme instead :D
<bentiss>
daniels: the thing is you don't have vmctl in the container started by b2c: this is literally the container from your regular CI that you use
<bentiss>
I have been contemplating *a lot* virtme and b2c. And virtme has a lot of benefit, but is "just" a wrapper around qemu (with probably as many options as qemu has)
<mupuf>
yeah, I was confused why one would need to use vmctl
<bentiss>
while b2c allows for a regular environment building (a container), and then you use b2c to get a safe environment
<bentiss>
I still need to pull that yaml file in the chart though
ximion has joined #freedesktop
<shadeslayer>
bentiss: you were right btw :)
<bentiss>
shadeslayer: I like that :)
<bentiss>
not sure why though
<shadeslayer>
yeah, seems odd
<bentiss>
shadeslayer: well, I wasn't sure why I was right, but the fact that the artifacts are uploaded only on failure is a way for the mesa team to not use useless storage AFAICT
<shadeslayer>
bentiss: I'm pretty sure we still artifact on success, unless my MR for lava duts was merged
<daniels>
bentiss: hey so for the CI data writing into Influx, I think I've kind of come full circle ... would you have any objection to exposing Influx2 svc out to public HTTPS and the data being written from a CI job where the token is just a secret var in that repo?
<bentiss>
daniels: which CI data?
<daniels>
bentiss: the perf-tracking stuff I added a bucket for a couple of months ago?
<bentiss>
daniels: can we get a separate influx for that?
<daniels>
but you've done about 7 driver migrations since then so I don't blame you ;)
<daniels>
yeah sure, separate Grafana as well or same?
<bentiss>
same grafana should be fine, I think, as most things happen locally
<daniels>
s/driver/storage/
<daniels>
ok, no prob :) thanks!
<bentiss>
but I am slightly worried having the same influx we use for general k8s used and presented externally
* daniels
nods
tobtoht has quit [Remote host closed the connection]
<daniels>
wfm
<daniels>
thanks :)
<bentiss>
k, thanks
<daniels>
seemed easier to reason about than RBAC ...
<danvet>
bentiss, daniels thanks a lot
tobtoht has joined #freedesktop
<bentiss>
danvet: is there a fdo board meeting today?
* bentiss
promised to Lyude a status report for fdo infra, but completely forgot about it
<Lyude>
bentiss: there is, we've still got 4 hours before it happens
* bentiss
doesn't have 4 hours tonmight :(
<bentiss>
Lyude (and daniels) email in your RH inbox. daniels: feel free to add anything I might have forgotten
<Lyude>
bentiss: oh I didn't think you needed to present it :P
<Lyude>
I was just expecting you give me a summary and I mention it
<bentiss>
Lyude: oh, no, I never intended to attend the meeting. But my evening (starting from now) is pretty much packed, with my kids having their year end sport events
<Lyude>
ahhh okj
<Lyude>
bentiss: you can take your time with that btw, it's fine if you can't make the report every now and then :P
<bentiss>
Lyude: well, this one was important enough to mention I think
<daniels>
bentiss: could you please push the ceph+opa & kilo RBAC changes to helm-gitlab-config?
Kayden has quit [Quit: to the wasteland!]
<daniels>
bentiss: (also, do you have more unpushed changes which move svcs from default -> packet-HA? with what I have, logging/influxdb2 is only deployed in the default env, but that doesn't seem right since 'default' is wildly out of sync with what's now on packet-HA i.e. the new cluster?)
<mupuf>
bentiss: jesus, I wonder what you are having for breakfast, but I want to eat the same! You are so fast!
<bentiss>
daniels: hmm, everything I have is pushed, but the logging part is using the default environment
<bentiss>
helmfile -f logging/helmfile.yaml diff (on the branch packet-HA)
<daniels>
bentiss: ooh OK, I was just doing it from the root helmfile.yaml as that includes the logging one too
<bentiss>
yeah, sorry, got caught up by time, and I also wonder if it makes sense to have environments
<bentiss>
for the k3s deployment
<bentiss>
(though being able to have separate networks was a requirement)
<bentiss>
mupuf: for breakfast, whenever I can: pains au chocolat!
<mupuf>
nice!
* mupuf
became unfrench: porridge!
<daniels>
bentiss: yeah, just merging the two might make sense
<daniels>
and np, was just confused :)
<daniels>
just need to import nginx + ca-injector + etc
<bentiss>
mupuf: noooooooo
<mupuf>
bentiss: hehe
<bentiss>
daniels: nginx is in the omnibus chart, no?
<bentiss>
or are you doing some cleanup?
<daniels>
bentiss: also, isn't it called a 'chocolatine'? O:)
<mupuf>
daniels: ha ha, someone likes to torture the frenchies :p
* mupuf
would call it a chocolatine, for sure!
<mupuf>
especially since I love both chocolatine and chocolate in bread
<bentiss>
daniels: I spent 21 yearss of my life calling it pain au chocolat, then 12 chocolatine, and since then it's pain au chocolat
<daniels>
hehe
<daniels>
bentiss: indeed ingress is in omnibus, I've just never done cross-namespace ingress before
<bentiss>
daniels: anyway, the deployment should be mostly packet-HA except when there is no such environment in the config. But everything is on the master of omnibus and packet-ha of -config
<daniels>
right
<daniels>
it was just using default from the root-level helmfile rather than logging/helmfile
<bentiss>
as for the rbac kilo change, that's just becuse I haven't locked the kilo version and we got a change in the deployment since we did it
<bentiss>
sorry, got to go pizzA HUNTING
<daniels>
anyway, should I just copy the ingress template from the omnibus chart and stick that in influxdb2?
<daniels>
hehe, np
<bentiss>
damn caps lock
<daniels>
bon appetit!
<bentiss>
daniels: ideally, I'd like to have a separate deployment (helmfile) for your influxdb2 and ingress, if possible
<bentiss>
well, we can always split it out later
<bentiss>
bbl
<daniels>
bentiss: yeah, I don't mind doing that, but it does mean copying the influxdb2-init templating across as well I think?
<daniels>
or I guess maybe I could cross-reference it
<daniels>
I don't really mind how it's split up, happy to go with whatever you think is best
<daniels>
anyway, gogogo, talk tomorrow :)
veganaiZe has joined #freedesktop
ngcortes has joined #freedesktop
Kayden has joined #freedesktop
blue__penquin has quit [Quit: Connection closed for inactivity]