►
From YouTube: Working Group: 2022-03-29
Description
No description was provided for this meeting.
If this is YOUR meeting, an easy way to fix this is to add a description to your video, wherever mtngs.io found it (probably YouTube).
A
A
All
right,
I
think
we
can
start
well
welcome
everyone
to
the
pacquiao
working
group
meeting.
I'm
really
glad
to
see
you
here
all
right,
so
let
me
share
my
screen
feel
free
to
add
yourself
to
that
in
this
list
totally
optional,
but
it
will
be
nice
to
see
you
there.
A
Okay,
here
again,
we
will
name
for
the
notes
in
case
you
don't.
A
All
right
new
faces
well
not
much.
So
let's
get
started
review
outstanding,
rsp.
A
Yeah
seven
days
ago,
it
has
now
six
approvals
so
yeah
according
to
the
the
rfc
process
that
will
be,
we
will
consider
approved
right,
ready
to
be
merged.
A
All
right
cool
just
wondering
what
would
we
I
mean
how
someone
will
know
that
this
is
well
yeah.
D
A
It's
merged,
but
I
was
wondering
if
we
want
do.
We
want
to
have
labels
or
use
projects
just
to
tag
rfcs
for
this
specific
stage.
They
are
yeah
just
a
suggestion
for
especially
for
folks
out
there
to
know
what
the
current
status
of
an
rfc
just
looking
at
the
beaver
okay.
So
this
one's
ready,
your
runtime
environment,.
C
E
I
had
some
notes
on
this
one
about
like
the
number
of
images
that
be
retaining
based
on
kind
of
current
trends
and
usage
if
we
kept
the
images
for
two
years.
So
that's
something
that
you
were
waiting
to
see
that
that's
in
there
now.
B
C
E
Yeah,
I
think
I
think
we
said
last
week
I
needed
to
rewrite
a
few
things
I
just
haven't.
I
haven't
got
to
that
yet.
C
A
C
A
Okay
seems
like
no
more
rfcs
before
so
next
step
will
be
we'll
leave
cmb
to
updates
wondering
if
there's
any
update
or
questions
are
on
big
cmb.
A
If
you
have
anything
else
to
share
this
because
yeah,
please
add
it
here
for
now,
I
I
have
a
couple
of
well
an
update
and
a
couple
of
questions.
First,
update
with
the
mailing
list
is
a
live
it.
It
was
alive
since
very
long
time
since
like,
but
we
were
not
using
it.
A
So
I
I
completed
a
couple
of
verifications.
This
tool
is
under
linux
foundation
in
general.
It's
you
know
it
has.
It
has
several
features.
You
know
it's
way
better
than
google
groups,
so
yeah
feel
free
to
join.
If
you
want
to
start
a
thread
there,
you
just
need
to
send
a
message
to
main,
at
least
that
pocateo.io
and
it's
a
good
way
to
start
discussions,
and
eventually
it
will
be
the
vehicle
to
release
the
newsletter.
A
But
yeah
first
thing
was
to
have
the
proper
tool
to
do
it.
Now
we
have
it,
I
will
be
sending
the
pr
to
update
the
community
repo
to
point
to
the
million
list
by
reading
the
ribbon
of
that
I
had
a
couple
of
questions.
A
A
You
know
to
fix
a
typo
in
documentation
that
makes
this
person
a
contributor
new
contributor.
Does
it
require
to
pass
through
an
election
process
by
the
maintainers.
F
So
the
contributor
is
actually
a
like
elevated
status,
above
just
contributing
to
the
project.
This
person
is
like
considered
to
be
connected
to
a
particular
language
or
project
group
within
picato,
and
they
have
the
ability
to
do
things
like
help
to
triage
issues
or
open
pull
requests
against
the
repository
without
having
to
create
like
a
fork
of
the
repo.
F
So
they
have
like
actually
like
more
elevated
permissions
than
say
like
an
outside
contribution.
There
is
no
election
process
or
any
the
only
real
barrier
to
someone
contributing
from
the
outside
is
signing
the
the
cla
that's
kind
of
the
the
one
thing
that
is
maybe
a
barrier,
but
this
is
actually
like.
You've
got
outside
contributor
which
isn't
really
enumerated
in
the
list
here
and
then
you've
got
this
contributor
status
and
then,
above
that,
you
have
maintainers.
A
A
A
So
the
first
question
that
that
I
had
you
know
by
looking
at
the
kettlebell
packs
repo
is
that
we
have
136
ripples
there
and
I
don't
have
a
clue
of
which
repo
to
track
contributions.
A
So
I
was,
I
was
trying
to
infer
it's.
For
example,
it
could
be
something
related
to
this
organization.
I
mean
there
are
know:
repos
maintain
net
repos
maintained
by
you,
know
these
users
and
a
forest
frankie.
So
here
the
document
mentions
that
some
teams
maintain
some
set
of
repos
but
don't
specify
which
repos
so
I'm
trying
to
wonder
which
repos
to
track
for
in
terms
of
contributions
and
which
are
the
relevant
ones.
So
probably
you
don't
have
an
answer
for
that
just
now.
I
will
probably
file
an
issue
for
that.
A
C
Like
the
way,
the
project
works
with
all
the
tiny
component,
build
packs
being
added
up
into
larger,
composite,
build
packs
and
then
builders,
we
end
up
like
splitting
the
work
across
many
many
repos,
but
there.
C
A
C
C
Feel
like
they,
they
do
have
steps
for
how
you
could
sort
of
use
your
own
instance
of
dev
stats,
but
it's
not
like
as
delightful
to
just
stand
up
and
use
as
I
was
hoping
it
would
be.
When
I
first
read
that
you
could
set
up
your
own
instance,
but
I
do
think
it's
possible
if
we
wanted
to
go
that
direction
so.
F
It
was
my
understanding
that
cloud
foundry
was
using
this,
although
I
just
went
to
the
link
where
I
thought
that
their
their
actual
stats
page
was,
and
it
appears
that
the
dashboard
has
been
decommissioned
sometime
in
the
last
month
since
I
last
looked
at
it.
So
I
don't
know
if
they're
tracking
this,
but
all
the
paquetto
stuff
was
actually
tracked
under
that
instance,
you
could
actually
filter
for
paquetto,
and
then
you
could
just
see
piqueto
stats
by
themselves
separate
from
the
rest
of
cloud
foundry.
F
A
Yeah
I
tried
to
access.
I
found
that
dashboard
last
week
and
yeah.
It
was
same
for
me.
I
wasn't
able
to
access
that,
but
I
will
work
with
this
csf
thing
to
see
what
happened.
Probably
there.
I
will
find
the
answer
for
which
repos
they
are
actually
tracking,
because
what
we
have
in
hand
right
now
here,
at
least
from
the
vmware
point
of
view-
is
bittersweet
instance
and
we
could
track
all
these
kind
of
metrics.
But
again
the
the
question
will
be
which
requests
so
yeah.
A
E
Yeah,
I
just
wanted
to
raise
this
up
because
we
had
a
particularly
rough
week
last
week
and
I'm
curious
how
much
this
is
impacting
other
people
and
looking
back
at
their
history
of
the
status
page.
It's
there's
quite
a
few
incidents
over
the
last
couple
months
and
I'm
just
wondering
what
other
people
other
people
feel
about
this.
E
E
No
no
svn,
but
I
am
curious
if
if
people
have
used
other
ci
systems
or
if
there's
anything,
you
know
any
thoughts,
people
have
on
that.
We
don't
have
to
get
into
it
too
deep
now.
But
if
you
have
thoughts
about
that,
you
know
feel
free
to
reach
out
to
me.
I'm
kind
of
investigating
exploring
options.
C
Github
actions
is
so
convenient
because
it's
so
easy
for
people
to
find
and
it's
easy
to
integrate
with
the
repos,
and
we
don't
have
to
pay
a
lot
of
money
for
it.
So
there's
so
many
things
I
like
about
it,
but
if
it's
unreliable
and
continues
to
be
unreliable,
that's
you
know,
get
it
exact.
A
pretty
high
toll
in
the
long
run,
so
yeah,
that's
yeah
and
nothing
to
add.
E
Yeah,
I
think,
there's
other
things
about
it
too.
That
are
not
summed
up
in
this.
You
know
I
mean
I
get
emails
from
all
of
our
our
jobs
that
that
fail,
and,
I
would
say,
like
weekly,
there's,
easily
20
or
30
or
more
jobs
that
just
fail
for
random
networking
reasons
and
other
random
reasons,
and
it's
it's
a
pain
to
have
to
look
and
see.
Oh
why'd,
that
fail
this
time.
You
know
it's
just
wasting
a
lot
of
time,
but
you
can't
necessarily
ignore
it,
because
you
know
one
in
50.
C
E
E
So
anyway,
if
people
have
thoughts,
feel
free
to
reach
out
partly
venting,
partly
interested
in
what
other
options
are.
D
Always
this
has
to
do
with
the
the
actual
usage
of
pochetto,
like
I
know,
other
projects
that
use
with
her
actions
but
like
not
quite
to
the
same
extent
that
the
ghetto
does
so.
I
think
that
also
magnifies
the
issue
like
there's
a
lot
of
automation
built
into
the
project
around
github
action
so
like
even
if
1000
jobs
are
failing,
you'll
notice
it
more
often
here
than
any
of
the
other
projects
too.
D
C
We
used
to
have
a
self-hosted
concourse
instance
back
in
the
day,
except
sort
of
the
lineage.
A
lot
of
the
vmware
folks
come
out,
of
which
I
think
we
were
able
to
use
pretty
efficiently,
because
most
of
the
people
here
are
very
familiar
with
concourse.
But
it's
not
great
for
large
distributed
projects
that
are
doing
pr
validation
and
you
have
to
you
know,
maintain
your
own
instance
and
it
uses
a
bunch
of
bespoke
technologies.
So
I
wouldn't
want
to
go
back
in
that
direction,
but
maybe
like
hosting
a
techton
instance.
C
F
H
We're
doing
all
of
our
internal
buildback
work
on
concourse
it's.
It
is
working
quite
nicely,
but
a
lot
of
there's
been
a
lot
of
development,
sort
of
investment
going
into
creating
custom
resources
and
whatever
else
to
to
kind
of
do
what
we
needed
to
do,
but
it
really
sucks
with
pull
request.
Workflows.
That's
that's
where
it's
hurting
us
at
the.
D
Moment
think
github
is
also
scheduled
to
release
like
a
blog
post
on
why
this
happened
right
a
few
threads,
and
they
said
that
this.
This
is
like
they'll
try
to
contain
this,
but
this
all,
like
they
march
out,
is
at
least
one
from
one
instance
that
just
had
sort
of
a
loop
with
things
retrying
their
database,
just
not
being
able
to
cope
up
with
the
load
and
just
failing
over
and
over
again
with
their
mitigation
processes,
not
being
able
to
cope
up
with
the
load,
and
they
did
say
that
they'll
resolve
this.
D
I
think
natalie
would
probably
be
a
better
person
sponsor
this,
but
we
just
host
self-hosted
drummers
on
equinox,
I
believe
and
then
just
use
github
action
circuit
of
actions.
Allow
you
to
configure
self-hosted
runners.
We
just
use
that
that
does
mean
that
the
self-hosted
one
was
a
bit
flaky
every
now
and
then
the
life
cycle,
pr
will
just
not
build
for
arm
and
then
nashville
has
to
go.
Kick
it,
but
I'm
guessing
with
the
increased
arm.
Adoption
github
at
some
point
has
to
provide
something
like,
but
for
now
the
only
option
is
self-hosted.
D
Funnels
cncf
has
also
got
this
like
cluster
of
machines
that
they
offer
to
projects
which
they
are.
What.
H
Arm
also
used
to
offer
resources
to
open
source
projects
that
wanted
to
validate
that
they
were
built
on
interesting.
F
I
think
for
us,
like
the
biggest
thing,
is
mostly
that
we
can
build
armed,
docker
images,
I'm
not
entirely
convinced
that
we
actually
need
runners
running
on
arm
mostly
because,
like
the
things
we
end
up,
building
are
things
you
can
like
cross
compile.
So
it's
we
could
cross,
compile
like
our
go
binaries
into
arm
64
like
architectures.
F
C
C
A
D
C
D
Yeah
yeah
yeah-
I
forgot
about
that.
So
I
was
saying:
cncf
has
this
cluster
offering
which
any
open
source
project
can
apply
to
and
they
do
provide
equinox
runners
there.
So
you
can
also
just
get
stuff
from
cncf
like
even
if
you're,
not
a
cncf
project,.
D
There
are
thinking
just
on
board
using
an
issue.
I
know
a
bunch
of
non-cncf
projects
that
have
converted
to
it.
Now
at
some
point,
there
were
plans
to
create
an
automation
system
around
the
same
repository
that
would
automatically
manage
github
runners
for
you,
it's
just
very
low
on
their
priority
list.