►
From YouTube: Kubernetes SIG Node 20210104
Description
Meeting Agenda:
https://docs.google.com/document/d/1j3vrG6BgE0hUDs2e-1ZUegKN4W4Adb1B6oJ6j-4kyPU
A
Now
you
been
recorded
and
it's
2021
first
meeting
of
2021
january
1st.
Fourth,
it's
a
ci
signaled
meeting
welcome
everybody.
We
have
very
short
agenda.
Let's
I
mean
you
added
your
agenda
item
first,
so.
B
Should
okay,
hello?
This
first
task
is
about
the
renaming
dashboard
from
gca
to
cause.
So
I
start
the
pr-
and
I
start
some
document
to
describe
what
we
should
do.
B
This
is
something
we
need
to
discuss
first,
because
there
is
no
real
gain,
just
moving
everything
to
a
new
tab.
We
are
only
renaming
things
that
people
from
see
the
infra
was
trying
to
do
so
for
signal.
That's
our
part.
B
A
So
so
you
want
like
this
test
logically
belongs
to
both
containers,
because
it's
the
test,
continuity
and
the
cause,
because
it's
cost
specific.
Is
it
the
case?
Yes
yeah.
I
wonder
what
this
test
does,
because
I
mean
everything
in
course
needs
to
be
cost
specific.
I
think,
and
I'm
wondering
what
what
it
will
do,
but
I
mean
yeah.
Logically,
it's
fine.
B
I
I
think
we
we
need
roy
to
review
this,
this
vr,
as
well
so
being
here
on
slack.
B
Okay
and
you
have
another
one-
no
other
one
is
a
flag
that
I'm
adding
on
the
node
e2e
test.
So
basically,
we
need
a
way
to
restart
cubelet
or
control
the
restart
of
cubelet
in
the
test.
B
B
A
B
Yes,
yes,
we
are
graduating
some
flags
on
cubelet
for
the
log
contention
and
I
need
to
not
restart
the
cubelet
numeric.
So,
let's
see
whether
it's
on
our
project
board.
A
Let's
say
queries
and
are
these
queries?
You
see
it's
open
a
sig
note
and
in
testing,
for
there
is
no
test
area,
so
there
is
no
need
for
label
tests.
I
just
get
project
like
I
exclude
everything
that
belongs
to
the
project
already,
so
that's
already
added
to
this
project,
so
you
see
like
there
is
no
zero
issues
here
and
then
I
did
this
for
all
like
prs
issues
on
test
infrared
and
pr's
interest
on
kubernetes.
A
Here
I
do
area
test
also
signaled
and
then
issue
the
same
and
then
a
failing
test
again
signaled
and
they
all
empty.
Now,
because
I
moved
all
issues
into
triage
space,
I
mean
everything
that
wasn't
added
yet,
and
I
discovered
that
some
of
them
are
like
not
related
to
our
group,
and
what
I
want
to
do
like
is
out
of
48
right
now,
like,
for
instance,
this
one
is
drop,
docker
shim,
it's
doesn't
belong
to
our
group
at
all.
So
what
I
will
be
doing
is
just.
A
Am
I
signing
wait?
Okay,
it's
after
location,
I'll,
just
computer,
I'm
not
signed
in.
A
Okay,
sorry,
so
some
of
them
clearly
not
belong
to
us
and
what
I
will
be
doing
is
just
go
ahead
and
archive
it,
so
archiving
will
make
it
disappear
for
from
this
dashboard
and
also
disappear
from
add
button.
So
if
I
will
try
to
add
this
issue,
it
wouldn't
show
me
anymore
any
longer.
So
there
is
a
add
cards.
I
can
just
query.
I
will
see
nothing
so
before
every
meeting.
A
A
I
don't
know,
I'm
not
sure
if
you
want
to
three
hours,
47
issues
right
now
together.
It
may
be
not
a
bad
idea,
but
if,
if
you
want
to
do
it
offline,
we
can
do
it
fine.
I
wonder
if
everybody
here
has
a
permissions
to
triage.
A
Yeah,
I
don't
know.
C
A
First
meeting
after
weekend,
long
quantities.
D
Yeah,
all
I
can
do
is
do
the
copy.
I
can't
don't
touch
them
at
all,
so.
A
A
D
Me
in
there,
but
a
not
version
of
me
b-a-u-e-r.
D
C
A
A
A
So
now
we
can
move
like
the
ashes
and
yeah
it's
47
issues.
I
I
hope
that
every
new
meeting
it
will
be
way
less
issues,
so
let's
triage
them
offline
and
try
to
like
at
least
archive
everything
that
doesn't
belong
to
our
group
and
now
I
want
to
look
at
the
issues
that
I
in
progress
or
my
pr
is
so
we
have
this
one
well,
this
is
supposed
to
be
done.
A
Oh
no!
This
is
something
that
who
is
working
on
this.
A
D
Yeah,
I
think
those
are
related
and
they're
the
same
kind
of
thing
and
I
need
to
review
the
existing
pr
because
it's
complicated.
A
Oh
yeah,
there
is
a
restart
plug
that
I
mean
we'll
be
talking
about.
A
I
can't
find
it
very
quickly,
so,
let's
maybe
later
so
this
is
I
mean
working
on,
and
this
is
roy.
A
So
those
issues
are
currently
in
progress.
I
think
we
have
more
so
if
you're
working
on
something-
let's
put
it
here,
do
we
have
anything
missing.
B
We
have
the
serial
the
serial
tab,
the
flake
one-
I
don't
know
if
it's
in
progress,
also,
no
someone
is
looking,
but
this
is
too
flaky.
B
A
Yeah
because
request
last
missing
when
we
discussed
how
to
triage
it,
I
think
we
said
that
we
want
to
put
everything
that
is
in
progress
into
the
start,
so
we
know
what
we're
working
on
and
we
can
understand
what
is
stored
and
what
we
need
to
like
how
this
okay.
So
let's
find
this
blanket
test
and
put
it
here
and
reviews
and
progress
like
there
are
three
two
tabs.
I
renamed
this
tab
into.
A
It
used
to
be
reviewer
approved
and
we
don't
have
reviewer
approved
it's
like
non-existent
in
kubernetes,
so
I
put
reviewed
lgtm
and
all
these
issues
already.
All
sprs
now
already
have
lgtm
tab.
So
this
is
something
that
we
need
to
bring
to
signal
group
tuesday
and
tell
them
please
go
ahead
and
approve
this,
because
it's
also
lgcm
by
our
group
or
by
somebody
and
now
like
most
complicated
npr's
currently
in
progress,
is
review.
A
This
is
what
I'm
even
working
on
timeout.
It's
here,
I
think
driving
this
forward.
C
D
A
C
A
A
So
let
me
move
it
into
this.
A
Thank
you
for
joining
and
welcome
to
the
new
year.
B
A
C
C
A
And
this
one
is
jim
started,
I
I
need
to
take
a
look
and
listen
whether
you
want
it
yeah.
I
think
it's
all
been
tracked
yeah.
I
will
double
check
with
maybe
next
time
with
jorge
whether
he's
driving
this
timeout
thingy,
because
it's
very
old
and
you
may
not
need
it
any
longer
yeah.
I
think
that
the
way
we
can
triage
everything
so
on
the
left
top
is
like
done
so
maybe
you
can
like
put
like.
A
I
have
I
archived
everything
that
was
here
and
now
we
can
like
drop
everything
here
and
then
celebrate
and
we
achieve
a
lot
yeah.
Do
you
like
this
way
to
triage?
Do
you
think
it's
reasonable
to
like
flow?
Oh
yeah.
A
A
Okay,
let's
finish
them
for
today,
any
other
topics,
no
okay,
no
nothing.