►
From YouTube: Kubernetes SIG Windows 20180417
Description
Kubernetes SIG Windows 20180417
A
I
have
hello
everybody
and
welcome
to
another
sick
windows
meeting.
Thank
you
all
for
attending,
not
a
lot
of
topics.
Today.
It
looks
like
you
know.
We
are
you
know
we
don't
have
a
lot
of
things
in
flight.
There's
a
couple
of
updates.
I
know
from
from
Bob
and
I
leaned
on
the
ovm
kubernetes
work
that
they're
doing
it
looks
like
we
got,
got
some
great
progress
there,
so
I'll.
Let
them
speak
to
that
and
we
do
also
have
a
proposal
and
I
want
to
start
with
that
on
the
board.
A
So
we
would
like
to
nominate
a
second
lead
for
the
sick
windows
effort
and
unless
anyone
objects
effective
immediately,
Patrick
Lang
will
join
me
as
a
second
lead
for
sick
windows
that
will
have
one
lead
from
a
printer
and
one
for
Microsoft
to
help
us
load
balanced
vacations
and
also
it's
good
for
every
seek
to
have
a
secondary
lead.
Anybody
opposed
to
that.
A
All
right:
congratulations!
Patrick!
Your
your
cig
lead
on
Windows,
we'll
figure
out
how
to
make
it
official
I
think
all
you
have
to
do
is
just
edit
a
couple
of
pages
in
the
kubernetes
github
repo,
and
your
name
will
show
up
there
we'll
figure
out
how
to
do
that
tomorrow,
all
right,
let's
go
around
and
do
some
updates.
Bob
humor
get
started
on
the
ovm
kubernetes
journaling.
B
Sure
I've
been
doing
some
testing
on
some
of
the
changes
that
Alan
had
made
and
had
a
few
fixes
that
I
fear
for
I
know
that
there's
this
to
realize
on
the
PR
for
culet,
that's
still
outstanding
has
emerged
in
yet
it's
just
waiting
on
a
review,
a
wink
seven
here
but
and
I'm
also
working
on
a
guide,
just
kind
of
like
we
used
to
have
with
the
the
inner
games.
Cluster
walk
through
one
of
those
or
the
newer
go
controller.
C
E
So
the
two
biggest
issues
are
related
to
metrics
I,
see
they
in
the
agenda.
They
were
so
basically
in
their
batches
of
tests,
where
some
of
them
tried
to
create
a
autoscaler
and
the
test
image.
Basically
just
generates
cpu
load
and
the
point
of
the
test
would
be.
The
reactor
hits
a
certain
threshold.
They
would
spawn
another
pod,
which
never
happens
because
the
metrics
are
not
are
not
collected.
A
F
G
A
E
E
Looked
over
we're
starting
to
get
the
and
find
bugs
a
lot
and
basically
we're
trying
to
figure
out
how
much
we
dig
into
them.
How
much
of
them
are
known
issues
and
what
we're
doing
is
a
triage
like
a
hey,
I
know
this,
and
this-
and
this
is
happening
and
do
you
guys
know
anything
about
it?
That's
because
otherwise
we're
spending
a
lot
of
time,
figuring
out
issues.
They
may
be
known.
A
A
Hopefully,
if
those
are
known
issues
or
not
just
to
be
safe,
if
you
can't
find
a
known
ticket,
always
ticket
things
that
way,
we
have
four
positions
and
for
us
for
documentation,
have
them
if
there's
things
that
are
blockers
like
the
high
priority
ones,
the
ones
that
just
prevent
you
from
doing
progress,
wanna
raise
them
up,
and
maybe
every
time
we
have
the
seek
meeting
every
week,
maybe
show
us
those
two
three
tickets
or
really
that
we
can
see
see
if
we
can
find
resources
to
work
on
them.
First,
yeah
yeah.
F
F
So
there
were
some
DNS
changes
made
in
the
patch
that
went
out
last
Tuesday
that
should
have
fixed
that
stuff
between
DNS
and
Service
Phipps.
So
if
you're
still
seeing
problems
there,
let
me
know:
I
talked
with
that.
Only
earlier
today,
it
sounds
like
there's
still
some
problems
in
the
ACS
deployments,
so
I'm
raising
that,
with
with
the
windows
Deb's
that
we're
working
on
that
fix
as
a
next
step.
So
I'm
sorry
go
back
to
test
ass.
So.
A
B
Had
an
issue
deploying
the
dashboard
to
test
it
out,
making
the
request
for
the
desk
board,
it
wasn't
getting
all
of
the
response
back
now.
They've
done
some
work.
Some
fixes
with
alienates
I
could
try
again
to
see
if
that,
with
the
updated,
again
related,
offense
part
of
the
issue,
maybe
not
I
did
see
a
ticket.
That
was
a
similar
problem
where
they
wouldn't
get
the
for
response
back,
but
that
was
there
was
no
resolution
to
that.
So.
F
F
B
It's
essentially
what
I
was
doing.
I
know
that
there's
a
there's
two
others
one
or
two
others
that
people
that
work
on
Windows
related,
but
like
ash
and
maybe
another
one
that
people
wanted
edit
in
I,
can
see
if
I
can
find
the
issue
for
that.
My
bra
was
just
actually
testing
it
and
trying
to
get
the
dashboard
says
the
dashboard
that
I
feel
actually
what.
F
F
F
Lost
track
of
it,
cuz
I
put
him
in
the
meeting,
that's
ahead
of
time
and
then
promptly
forgot
what
what
they
were.
Okay,
so
one
of
them
they
wanted
to
check
on
was
about
some
some
demo
stuff.
So
I
was
talking
to
Cindy
over
at
Huawei,
and
they've
actually
got
a
preview
offering
available.
That's
able
to
run
Windows
containers
and
then
the
other
thing
that
I
was
thinking
about
was
we've
got.
The
work
pain-free
did
to
get
some
alpha
support
for
using
hyper-v
isolation.
F
F
Okay,
we
already
talked
about
the
metrics
off.
Do
some
follow
up
on
that
and
then
same
thing
with
secrets?
We've
got
some
fixes
coming
in,
though
the
Windows
Server
version
1803
release
should
have
the
fix
for
symlinks,
plus
I,
can't
remember,
which
other
ones
specifically,
unfortunately,
I
don't
have
a
specific
date.
I
could
share
on
that
yet,
but
it
should
be.
You
know
within
the
next
few
weeks
because
they
called
it
1803
and
it's
still
not
out
yet
so
as
soon
as
that's
out
I'll,
everyone
know
if
you're
using
the
Windows
Server
insider
builds.
F
You
know
later
this
year,
so
I
think
that's
another
discussion
that
we
need
to
to
have
is
sort
of
formalized
what
what
those
requirements
are
than
steps
are
and
get
those
reviewed
by
sig
node.
The
first
thing,
of
course,
is
that
we
need
to
get
the
get
the
tests
running,
and
so
that's
that's.
What
been
the
other
major
push?
F
There's
I've
been
focusing
on.
You
know
long
with
the
folks
at
cloud
base.
You
know
we've
got
most
of
the
key
functionality.
That's
there.
We
we
just
need
to
fix
or
Tum
you
know,
set
the
bar
on
that
and
get
it
reviewed.
So
I
think
that's
something
I'd
like
to
probably
try
to
get
buttoned
up
in
the
probably
during
the
may,
in
terms
of
getting
the
criteria
done.
Does
that
does
that
make
sense
to
you
Michael
and
others?
Yeah.
H
H
A
We
already
have
most
of
the
high
level
features
from
the
hyper-v
isolation
staff,
the
CRI,
the
the
resource
control.
Most
of
those
things
are
there
now,
it's
just
a
matter
of
you
know
is
the
overall
quality.
They
are
good
enough
that
people
can
install
this
and
run
containers
of
scale,
and
if
we
can
assess
that
and
say
yes,
these,
then
we
can
probably
start.
You
know
marching
towards
our
GA.
That
line.
F
D
F
D
J
D
B
A
A
F
H
F
H
F
A
All
right,
I,
don't
have
anything
else
in
legend.
Anything
anybody
else
wants
to
ask
any
questions
concerns
all
right.
Well,
I
really
have
a
great
rest
of
your
day
and
a
good
week
and
congrats
to
Patrick
for
for
being
part
of
our
part
14
here
and
all
his
contributions
was
from
human
in
the
rest
of
the
Microsoft
group
and
we'll
see
you
guys
next
Tuesday
bye,
all.