►
From YouTube: Kubernetes Sig Node 20180306
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).
C
D
So
I
guess
what
I
was
thinking
today.
We
have
the
two
topics
from
first
with
a
crime
update
from
it
all
and
Josh.
You
had
your
topic
and
talk
through
and
then
I
want
to
reserve
time
for
any
PRS.
That
folks
want
to
bring
up
that
needed
to
get
review
or
attention
for
one
time
just
to
make
sure
it
was
on
everyone's
radar.
So
Rinaldi
you
want
to
chew
on
food
update
on
cryo
now
and
yeah.
D
E
Okay,
can
you
see
my
slides
yeah?
Okay?
So
it's
a
very
simple
and
boring
update
for
crowd,
so
one
9x
is
being
updated
based
on
feedback
that
we
get
from
production
like
minor,
bugs
and
feature
requests
that
would
improve
it.
So
we've
made
several
minor
releases
on
the
one
nine
train,
one
nine
eight
was
the
last
release
for
110.
Cryo
master
is
rebase
to
release
one
ten
branch
of
cube
cuban.
It
is
upstream
and
we
should
be
able
to
cut
the
release
on
our
c0
by
the
end
of
the
week.
E
We
are
passing
all
the
tests
with
110
our
cube
dashboard
result.
Reporting
is
currently
blocked
on
an
upstream
issue
and
in
the
test
and
flow
repo
for
for
a
couple
of
months.
So
we
will
appreciate
some
help
in
pushing
that
forward.
So
I'll
put
a
link
to
that
PR
over
there,
so
some
changes
basically
need
to
be
made
so
that
our
results
could
be
parsed,
because
we
are
not
running
on
GC
p,
we're
running
on
AWS
and
try
to
publish
results,
and
the
second
part
is
community,
so
community
seems
to
be
thriving.
E
We
are
getting
more
and
more
contributors.
We
are
up
to
74
contributors
right
now
and
if
you
take
a
look
at
the
person
last
week,
yeah
or
27
commits
going
in
when
we
are
having
a
hard
time,
keeping
up
with
PR
review
so
I
think
that's
a
good
sign
and
one
more
thing
I
wanted
to
bring
up
was
cutting
the
releases
of
cry
CTL
that
correspond
to
cube
version.
So
it
is
easy
for
someone
to
pick.
E
B
So
what
we're
going
to
be
doing
is
when
we
need
to
make
a
change
or
when
we
need
to
clarify
something
that
already
existed,
but
people
weren't
sure
how
to
use
that's
going
to
be
published
with
at
least
three
days
and
in
most
cases
like
a
week
or
so
of
notice.
On
the
contrabass
sigelei
days
and
kubernetes
that
mailing
lists,
we
will
also
bring
it
in
bring
it
up
on
the
thursday
community
meetings
to
make
sure
that
everybody
knows
about
it.
B
So
some
someone
or
several
someone
from
signo
D
as
many
people
should
be
on
one
of
those
forums.
Probably
the
community
meeting
kubernetes
dev
just
to
make
sure
that
you
hear
about
these
things
in
advance
and
if
they're
going
to
break
a
workflow
for
you
that
you
can
bring
it
up
in
advance
before
we
actually
push
the
change
through
automation.
B
If
you
are
an
owner,
there's
listed
in
the
owner
file
and
you
make
a
PR
and
has
automatically
approved
if
ult
TM
is
automatically
approved
and
not
everybody
wants
it
to
be
automatically
approved,
but
for
the
who
do
want
it
to
be
automatically
approved,
they
don't
still
want
to
take
the
extra
step,
and
so
that's
gonna
need
a
community
decision
to
settle
because
having
two
different
workflows
within
kubernetes
kubernetes
would
be
even
worse
than
either
the
drawbacks
of
either
workflow.
B
B
The
second
thing
a
contributor
died.
We
have
a
new
and
updated
version
of
the
contributor
guide.
This
is
a
document
on
how
to
contribute
to
kubernetes
in
terms
of
clas
and
contributions,
and
how
do
you
announce
things
and
that
sort
of
thing
and
other
stuff
please
review
and
make
comments
on
that
document
again
linked
from
your
meeting
minutes,
we
want
to
make
sure
that
what
we're
saying
in
there
does
apply
equally
to
all
SIG's
and
we
need
your
help
to
make
that
happen.
B
The
now
we
have
several
different
mentoring
programs,
and
so
mentoring
is
how
we
get
from
having
kubernetes
users
to
having
kubernetes
maintainer
x'
and
all
of
those
sort
of
steps
along
the
way,
and
one
of
the
biggest
news
of
the
project
has
right
now
is
to
move
people
sort
of
up
from
being
contributors,
I
to
being
reviewers
and
eventually
owners
the.
So
we
have
a
number
of
different
mentoring
programs,
the
two
that
I've
highlighted.
We
call
them,
you
know-
is
an
easy
participation.
One
called
meet
our
contributors
and
that
takes
place
once
a
month.
B
It's
got
currently
two
different
time
slots
where
it's
a
combination
of
slack,
what's
the
slack
Twitter
and
a
zoom
meeting,
so
that
we
can
reach
as
many
people
as
possible
for
people
who
are
interested
in
contributing
and
that
sort
of
thing
it's
really
helpful
to
have
representative
small
things.
We've
really
hopefully
have
a
representative
from
sig
node
on
meet
our
contributors
for
a
couple
of
reasons.
One
is
you
can
talk
about
what
you
do
and
what
signal
does
the
second
is
people
do
show
up
there
who
are
saying
hey?
B
We
have
had
a
lot
of
new
contributions
from
the
virtualization
folks,
based
on
where
those
projects
are
going.
Those
contributions
are
only
going
to
increase
the
and
I
just
linked
from
the
I
just
linked
a
dev
stats
chart
a
new
one
from
the
minutes
on
this
has
a
calculation
of
relative
cig
workload
based
on
incoming
PRS.
B
B
B
So
this
is
just
KK
now
sig
note
is
number
four
on
here
right
now,
I
actually
originally
created
this
chart
because
of
what
I
was
seeing
in
terms
of
signal
workload
about
three
or
four
weeks
ago,
when
signaled
was
number
one
on
the
chart
and
by
a
pretty
significant
margin,
things
that
actually
slowed
down
somewhat.
Do
you
guys
have
gotten
through
some
of
your
backlog?
Now
we
have
a
whole
bunch
of
virtualization
folks
coming
in
one
to
combine
virtualization
in
kubernetes.
D
I
think
you
know
for
that
topic
in
particular.
I
guess
one
of
the
things
I'd
be
curious
about
with
contra
backs
is
I.
Think
a
lot
of
the
groups
that
are
coming
to
Sigma
about
their
various
interest
areas
may
actually
have
their
work,
get
focused
more
on
a
particular
project
and
not
necessarily
cube
cube
proper
itself.
D
B
B
B
By
by
taking
people
who
are
contributors
now
who
are
who
are
good,
you
know
what
are
contributors
now
and
giving
them
that
knowledge,
particularly
the
knowledge
of
how
signal
does
things
and
what
issues
you
need
to
look
at
and
what
other
you
know,
sort
of
side
effects.
You
need
to
look
for
and
all
kinds
of
other
things
in
order
to
be
good.
A
F
C
A
Done
a
lot
of
review
and
work
in
like
wider
ecosystem
with
docker
and
stuff
like
that,
is
there
any
kind
of
way
that
I
can
help
kind
of
like
fast-track
to
help
with
reviews
and
stuff
like
that
or
you?
Would
you
recommend
that
I
go
through
the
process
of
going
through
from
contribution
and
becoming
a
reviewer
I.
D
I
actually
think
special
hardened
class
number
is
difficult.
One
of
the
things
I
think
that
we
should
do
as
a
cig.
Here
is
when
we
talked
about
this
last
week.
We
need
to
write
our
charter
for
the
sake
that
kind
of
describes
way
we
work
now
and
I
guess
the
way
we
aspired
to
work
in
the
future.
Okay,
I
think
a
key
element
in
that
charter
should
be
like
how
how
what
measure
of
contribution
you
need
to
to
appear
in
either
reviewer
or
an
ownership
role,
I'm
curious.
D
D
Like
the
actual
template,
I
know
that
Phil
Rock
has
put
together
I,
don't
think
it
goes
into
like
the
depth
of
you
reviewed,
X
number
of
PRS
I.
Don't
even
think
we
have
the
tooling
in
place
to
know
how
many
like
passive
reviews,
an
individual
has
done
and
I
definitely
don't
think
it
went
into.
Like
may
may
not
have
said
anything
more
than
just
like
demonstrated
sustained
contribution,
but
yeah
like
the
the
cubelet,
was
going
through
a
lot
of
changes,
obviously
early
in
its
formation,
and
now
the
changes
are
much
more
targeted.
D
B
B
So
in
terms
of
guideline
or
the
amount
of
review,
activity,
etc,
that
somebody
needs
to
have
I.
B
B
B
D
To
use
that
I
don't
know
how
everyone
else
feels,
but,
like
I,
think
I've
been
talking
about
this
with
Dawn,
probably
for
a
while,
throw
that
both
river
pool
and
the
owners
place
responsibly
and
like
Steven.
Your
experiences
is,
is
wide
and
deep
right,
so,
like
it'd
be
great.
If
we
can
account
for
that
when,
when
looking
to.
A
B
C
So
one
example
is:
there
is
an
owner
aliases,
there
is
signal
reviewers,
but
there
is
no
signal
approvers,
for
example.
So
if
we
go
through
the
hierarchy
of
directories
and
sprinkle
them
with
signal
reviewers
and
approvers
in
different
places,
then
you
know
the
people
who
are
coming
in
will
automatically
get
pinged
with
all
the
Box
picking
off
stuff
from
here.
B
D
So
if,
if
folks
have
have
interesting
ideas
on
how
you
can
look
to
do
that
unique
to
the
sick,
I
think
we
should
have
that
discussion
in
the
mailing
list,
or
maybe
we
can
dedicate
a
time,
maybe
when
Don
gets
back
to
get
that
tied
into
the
Charter
other
other.
The
last
topic
I
went
up
real
today
was
if
there
are
any
PRS
that
required
review
that
we're
not
getting
attention
for
one
time
that
people
wanted
to
bring
up
al.