►
From YouTube: Kubernetes SIG Node 20210111
Description
Meeting Agenda:
https://docs.google.com/document/d/1j3vrG6BgE0hUDs2e-1ZUegKN4W4Adb1B6oJ6j-4kyPU
A
I
think
it's
kicked
in
so
hello.
It's
january,
11th
2021,
signal
ci
group.
Welcome
everybody.
We
don't
have
much
nets,
we
have
somebody
new
sharma,
you
said
right,
hi,
so
yeah,
let's
start
with
the,
we
don't
have
any
agent.
Oh
yeah,
we
have
one.
We
have
no
agenda
items,
so
I
would
suggest
you
start
with
the
triage
and
see
what's
going
on,
what's
active
and
then
I'll
take
it
from
there.
A
A
B
A
Yeah,
I
have
different
laptops
for
different
purposes,
but
is
it
good
enough.
A
I
wonder
if
it's
belongs
to
our.
B
Oh
this
one
like
they
literally
just
capitalized,
all
of
the
log
strings
on
the
as
you
can
see,
it's
just
expected
or
like
volume
and
like
I
think
these
are
oh.
This
one
is
a
different
one,
but
I've
seen
like
a
number
of
pr's
like
this
recently
and
a
lot
of
them
were
like
just
the
log
strings
in
the
tests
not
even
like
in
production.
A
A
A
I
wonder
if
it
I
mean
it
has
all
possible
areas,
and
so
many
seeks
so
yeah.
A
A
Let's
review
what
is
in
progress,
conformance.
A
A
Fake
numa
enablement,
I'm
not
sure
this
after
the
already
anything
to
me.
A
A
E
Yeah,
I
have
a
pr
for
this
one,
so
basically,
we
are
proposing
another
tab
to
bring
to
bring
this
jobs
copied
for
a
signaled
cost.
A
Roy,
it
will
be
great
if
you
can
take
a
look,
because
this
is
basically
renaming
from
gci
to
course,
and
so
there
is
a
task,
and
there
is
a
pr
for
this
task
right.
Yeah.
A
To
you
and
testing
for
repository,
so
if
you
can
take
a
look
at
the
lgtm,
it
would
be
great
yeah
cool
okay.
So
this
is
actually
oh
yeah.
This.
A
Open
from
rotten,
I
I'm
not
actually
working,
so
I
will
put
it
here
for
now
and
to
do
yeah
anybody
else.
Somebody
working
on
that
is
not
in
progress
in
this
column.
A
A
Useless
parameters
from
end
to
end
test
function-
this
is
just
cleanup.
I
mean
yeah
yeah
just
need
to
take
a
look
renaming
this
one
we
just
discussed.
I
mean
working
on
that
android
will
take
a
look
timeout
config.
So
oh
yeah!
This
is
a
very
old
one.
You
just
yeah.
I
will
ping
jorge
again
asking
what's
going
on.
A
A
Switch
maestro
to
run
yeah
this
one,
this
one
on
me
yeah.
I
will
take
this
one
like
I
really
don't
like.
I
like
the
idea
to
switch
tests
and
do
containerity,
so
you
just
contain
it.
You
bought
better,
but
I
it's
really
blocked
on
on
the
fact
that
we
don't
know
whether
we
have
enough
coverage
for
docker
shim
and
you
don't
want
to
lose
this
coverage
and
break
a
majority
of
customers
running
on
docker
shim
majority.
According
to
a
survey
that
issued
that
was
issues
last
year.
A
What
a
status
of
this
one
misery
base
I'll
put
an
attention
from
author
okay,
so
we
have
six
pr
that
needs
attention
probably
need
to
go
and
ping
somebody
you
wanna!
Do
we
wanna
go
pink
everybody
on
this
list.
B
Six
pr's
yeah
for
the
ones
that
are
like
rotten
or
stale.
Like
those
ones
I
mean
the
bot
has
pinged
them.
Their
pr's
will
be
closed
if
they
don't
respond.
So.
A
B
A
C
A
A
Okay,
so
I
I
think
this
kind
of
like
at
x3
unit
test,
is
kind
of
a
good
way
to
start
so
you,
you
learn
how
to
run
unit,
how
to
run
this
test
and
how
to
and
then
it's
just
straightforward
way
to
start
contributing.
A
Yeah
another
thing
that
can
be
just
another
unit
enter
and
test
that
needs
to
be
added,
but
it's
in
life
cycle
frozen,
so
nobody
was
looking
into
it
for
a
long
time,
so
likely
even
got
immediate
help.
If
you
ask
like
what
exactly
needs
to
be
unit
tested
and
people
may
not
reply
very
fast,
but
if
you
can,
if
you
pick
up
this
one
or
like
that,
one
and
you
don't
have
any
replies,
pingas
and
slack
and
signal
stack
and
we
tried
we'll
try
to
figure
out.
What's
what
to
do
there.
A
So
this
jenkins
thingy,
it's
probably
taken
by
somebody
already.
This
is
upper
case
per
person.
Okay
out
of
there,
okay.
A
And
some
okay-
I
don't
see
oh
yeah.
This
is
morgan,
I'm
not
sure
what
the
morgan
is
working
on
right
now.
A
I
believe
there
were
failing
tests
and
then
we
fixed
the
test,
and
now
the
only
thing
left
is
to
check
change
the
output,
so
it
will
be
more
informative.
B
So
I
think
I
know
what's
going
on
with
this
one,
because
this
was
also
a
problem
in
openshift
ci,
the
so
the
120
regression
where
pods
are
taking
a
super
long
time
to
delete,
are
causing
these
to
flake
and
we're
actually
seeing
in
in
both
cube
ci
and
an
openshift
ci.
So.
A
Okay,
do
you
want
this?
Do
you
want
to
track
it
and
like
being
attend
to
you
in
progress.
B
I
don't
think
so,
because
I'm
not
working
on
the
fix,
like
I
think,
once
we
fix
the
120
like
pod
deletion
regression.
This
should
just
clear
out.
I
noticed
that
those
tests
were
flaking,
but
there
weren't
issues
assigned
for
them,
because
I
think
they
weren't
flaking
enough
to
count
or
something
and
yeah.
I'm
not
sure
if,
if
this
like
report
was
specifically
because
of
that
issue,
but.
B
Yeah,
well,
that
was
around
when
the
the
like
the
sandbox
stuff
was
merged
in
120,
which
then
caused
pod
deletions
to
take
a
lot
longer.
So.
E
Okay,
so
basically,
we
need
to
wait
for
the
fix
to
see
if
stops
flakeness
on
this
one
yeah.
B
A
B
C
A
Yeah
there
is,
there
is
no
like
this
one
is,
is
the
only
thing
that
seems
to
be
critical.
That
needs
to
be
taken
care
of
faster
and
once
you
review
it,
it's
probably
I
mean
I'm
working
on
fix,
I'm
working
on
the
investigation,
I'm
not
like
sure
what
the
fix
will
be.
So
maybe
once
it's
merged
it
will
magically
resolve
itself.
A
So
I
don't
want
to
take
ownership
of
this
particular
like
test
area,
but
let's
review
last
next
week,
if
you
have
a
fix
and
whether
to
keep
failing.
A
Okay,
I
think
we
charged
everything-
let's,
let's
try
to
clean
up
this
in
progress,
thingy
and
start
taking
new
work
items.
Any
other
item
for.
A
A
Okay,
if
no
agenda
item
for
today,
everybody
has
something
to
work
on.
Let's
chat
next
week,.