►
From YouTube: Brigade Community Meeting -- 2022-04-12
Description
A recording of the Brigade Community Meeting on 2022-04-12
A
All
right,
hello,
everybody
and
welcome
to
the
april
12
2022
occurrence
of
the
brigade
community
meeting.
I
will
share
my
screen
and
we
can
go
through
today's
agenda.
Can
everybody
see
this
and
is
it
big
enough,
yep,
okay,
it's
readable
at
this
size?
Yes,
okay!
So
we'll
we'll
try
to
get
through
this
relatively
quickly.
Oh,
let
me
add
philippe
to
the
list
of
attendees.
A
Okay-
and
he
said
he
was
coming,
but
I'll
probably
get
removed
if
he
doesn't
show
up
okay,
so
today's
agenda,
I
just
structured
it
kind
of
like
almost
like
a
stand-up
or
a
retrospective,
I'm
just
going
to
quickly
cover
what's
happened
in
the
last
two
weeks
and
what
I'm
hoping
gets
accomplished
in
the
next
two
weeks,
and
that's
really
all
I
have
to
discuss.
A
If
anybody
else
has
agenda
items,
we
can
add
them
to
the
end
so
feel
free
to
pop
into
the
the
document
right
now,
if
you
want
to
add
any
agenda
items,
that's
fine
and
please
everybody
always
feel
welcome
to
to
add
to
the
agenda.
This
isn't
just
for
the
maintainers
to
set
the
agenda.
We
want
the
community
to
be
able
to
set
the
agenda
as
well.
A
Okay,
so
what's
happened
in
the
last
two
weeks,
a
lot
of
work
has
gone
into
improving
and
documenting
the
developer
experience.
We
found
a
few
tools
that
have
made
things
a
lot
easier.
We
used
to
have
like
a
lot
of
hacky,
make
targets
and
scripts,
and
things
like
that
that
would
build
from
source
and
deploy
to
a
local
kubernetes
cluster.
Like
a
kind
cluster
or
something
like
that-
and
it
was
really
all
kind
of
you
know-
very
brigade
specific
and
didn't
always
entirely
meet
our
needs.
We
found
some
good
tools.
A
One
of
them
is
called
it's.
It's
ctl
ptl,
it's
up
to
you,
how
you
want
to
pronounce
that
some
of
the
pronunciations
are
like
cattle,
paddle
or
cuddle,
puddle
or
or
cattle
patrol,
but
it's
a
it's
a
nice
way
of
very
quickly
bringing
up
clusters
with
a
local
docker
registry
attached
to
the
cluster,
so
we've
been
using
that
lately
and
we
also
found
tilt,
which
is
just
an
incredible
tool
that
we
discovered
jorge
george.
A
I
don't
know
if
you've
played
with
with
tilt
at
all
before,
but
it's
absolutely
amazing
it.
You
know
it
integrates
with
your
your
docker
files
and
your
charts
and
everything
and
it
just
builds
from
source.
It
will
build
multiple
components
in
parallel
and
deploys
them
and
it
will
watch
the
code
for
changes
and
automatically
refresh
or
the
way
we
have
it
set
up.
A
It
will
watch
the
code
for
changes
and
notify
you
that
there
are
changes
and
give
you
the
option
to
refresh,
because
I
found
that
it
was
a
little
too
much
for
my
cpu.
If
I
just
let
it
build
and
refresh
all
the
time.
So
it's
really
an
incredible
tool
and
it's
it's
making
it
a
lot
easier
to
hack
on
brigade
and
to
hack
on
each
of
the
peripherals.
A
So
we've
started
using
that
in
the
past
couple
of
weeks,
and
we've
also
created
an
updated
contributor
guide
that
that
talks
about
the
tools
that
we
use.
So
all
of
this
is
obviously
being
done
in
the
name
of
growing
a
more
sustainable
community,
more
sustainable
group
of
of
contributors.
We
want
to
make
things
as
easy
and
productive
as
possible
for
everybody.
So
a
lot
of
effort
went
into
that
over
the
last
two
weeks.
A
We
also
added
a
feature
that
was
requested
from
a
community
member.
It
was
the
ability
to
use
an
externally
hosted
mongodb
database-
that's
actually
feature
I
had
wanted
from
the
beginning,
but
it
got
cut
from
2.0
right
before
we
went
ga
because
it
wasn't
tested
well
enough.
We
decided
we
could
go
and
put
it
back
later.
So
now
that
it's
been
requested,
we
went,
and
we
did
that
it's
not
released,
yet
it
will
go
out
in
the
240
release.
A
Also,
since
I
don't
know
two,
two
or
so,
we've
had
something
called
fallible
jobs.
So
we've
had
the
option
that
you
can
mark
a
brigade
job
as
fallible,
and
by
doing
so
what
happens
is
if
that
job
fails.
It
doesn't
make
the
entire
workflow
fail.
A
Well
until
now
the
worker
has
kept
that
information
to
itself.
It
hasn't
shared
it
with
the
api
server
when
it
creates
the
jobs
we've
made.
It
start
sharing
that
information
and
the
benefit
to
that
is
that
now,
if
you
have
gateways
that
create
jobs
and
follow
up
on
those
jobs,
monitor
their
status
and
report,
the
status
you
know
upstream,
to
where
the
event
originally
came
from
good
example,
would
be
the
github
gateway.
A
A
A
Now
the
gateways
have
access
to
that
information
and
because
they
have
access
to
that
information,
they
have
choices
now
about
how
they
report
event
status
or
job
status
upstream
to
the
source
of
the
event.
A
So
in
the
case
of
like
the
github
gateway,
for
instance,
we've
built
in
an
option
that
if
a
job
is
considered
fallible
and
you've
opted
into
this,
it
can
report
failed
jobs
with
a
neutral
status
instead
of
a
failed
status,
which
is
nice
because
when
you
go
and
you
look
at
the
check
suite
in
github-
something
that
you
intended
to
be
allowed
to
fail,
isn't
going
to
make
the
whole
workflow
show
up
with
a
nasty
red
x.
A
So
that's
something
that
we've
done
in
the
last
two
weeks.
Also,
I've.
A
B
A
I
can
tell
you
it
won't
work
with
cosmos
db
because
cosmos
e
cosmos
and
are
only
compatible
like
on
the
wire,
it's
the
same
protocol
between
client
and
server,
but
on
the
server
side.
It
doesn't
support
all
of
the
same
queries
that
that
supports.
So
I
I
do
know
already
that
this
does
not
work
with
cosmos
db.
Unfortunately,
okay.
A
A
If
they
want
to
get
involved,
we
have
to
vote
on
which
of
the
proposals,
if
any
that
we
want
to
accept,
there's
still
no
guarantee
that
those
projects
will
get
selected
because
cncf
as
a
whole
generated
many
dozens
of
project
ideas
and
there's
no
guarantee
that
that
google
accepts
all
of
those
projects.
So
basically
I
can
give
a
recommendation,
but
it
doesn't,
it
doesn't
mean
for
sure
that
anything
will
be
accepted.
A
I
also
see
that
the
hindi
translation
of
the
getting
started
docs
is
almost
complete.
So
hopefully,
that's
merged
within
the
next
few
days,
and
we
can
have
our
first
case
of
documentation
offered
in
in
another
language,
and
I
see
I
think
karen
may
be
jumping
in
and
adding
more
notes.
So
thanks
karen
is
that
you
or
somebody
else.
A
So,
as
for
what
I
hope
we
get
done
over
the
next
couple
of
weeks,
I
have
to
start
work
on
the
annual
project
review
for
the
cncf
technical
oversight
committee
and
karen.
I
might
ask
you
for
a
little
bit
of
help
with
that,
but
I
think
I
can
mostly
do
it
myself.
A
I
again
have
to
continue
reviewing
gsoc
proposals.
I
have
an
in-progress
blog
post
to
write
about
image
security,
so
we
did
talk
about
that
at
the
last
community
meeting.
We've
done
a
lot
in
the
recent
weeks
to
improve
image
security,
so
we're
scanning
images
that
we
create
as
part
of
our
ci
process
where
creating
and
publishing
s-bombs
as
part
of
our
release
process.
A
A
This
goes
back
to
what
I
was
talking
about
earlier
about
improved
support
for
fallible
jobs.
So
we
want
to
get
that
in
there
so
that
we
ourselves
can
take
advantage
of
that
sooner
rather
than
later,
and
then
I'm
going
to
focus
on
finally
releasing
some
of
the
other
peripherals,
probably
the
bit
bucket
gateway
and
the
noisy
neighbor
component,
just
because
those
seem
like
they
require
the
least
effort
right
now
to
get
them
across
the
line.
A
So
if,
if
they
really
are
low
on
effort,
I'm
just
going
to
go
ahead
and
get
them
done.
My
stretch
goal
is
a
video
adaptation
of
our
new
contributor
guide,
which
I
mentioned
earlier,
that
we
updated
a
lot
of
the
tools
that
we're
using
and
I'd
like
to
create
a
short
video
about
that
again,
just
in
the
name
of
building
a
more
you
know,
a
larger
and
more
sustainable
community
of
contributors.
A
C
C
No,
it's
okay!
Okay,
so
I
think
it's
okay
central
european
time
it's
gonna
be
3
30
to
4
15
in
the
afternoon
and
then
for
east
coast
us
time
it's
gonna
be
9
30
to
10
15
a.m,
east
coast
time.
So
ken
I
will
send
you,
I
think,
or
yeah
I'll
I'll,
update
the
calendar,
invite
and
invite
you.
C
But
if
anyone
would
like
to
participate,
you
can
click
on
the
link
and
it
will
have
you
registered
through
cncf's
community
bevvy
platform
that
they
use
and
it's
basically
just
like
google
hangouts
with
the
skin
on
top,
but
yeah
hope
you
can
join
us
spread.
The
word
and
yeah
like
I
said.
A
Karen
one
one
question
about
that:
yeah,
so
you
were
just
doing
the
time
zone
math
for
us
right,
yeah
that
wasn't
that's
not
two
separate
slots
like
one
for
us
right.
B
Ken
for
me,
on
the
next
few
weeks,
if
you
wanna
add
I'm
gonna,
start
the
translation
for
portuguese
philip.
If
you
wanna
help
or
we
need
to
get
started,
then
the
other
people
can
join
and
help
in
some
pace
and
the
chrome
job
blog
see
if
I
can
get
on
the
next
few
days
sure
we
have
a
easter
holiday
coming
australia,
I
have
a
friday,
monday,
public
holiday,
but
try
to
get
the
blog
first
inside
the
translation
and-
and
another
thing
is
another
gateway
that
I
want
to
start
from.
A
B
Yep
start
on
on
this
one
and
now
there's
another
thing:
the
five
minutes
videos
I
can
do
the
chrome
job,
one
okay,.
C
B
A
Oh
sure,
this
this
the
this
occurrence
that
happens
later
in
the
day
well
later
in
the
day.
For
me,
this
is
the
the
better
attended
of
the
the
two
that
we
offer.
B
B
B
A
Oh
yeah
also,
that's
that's
not
now
that
you've
just
said
that
you
reminded
me
of
something
else.
That's
that's
worth
mentioning.
Within
the
past
few
days
we
also
cleared
up
an
issue
that
was
making
it
impossible
to
build
everything
on
an
m1
mac,
but
that
is
fixed
now.
So
if
anybody
has
tried
that
and
had
difficulty,
that's
that's
been
remediated.