►
From YouTube: Kubernetes SIG Windows 20200505
Description
Kubernetes SIG Windows 20200505
A
A
B
Yeah,
it's
not
so
much
updates.
It
was
so
basically
I
realized
that
I
had
a
number
of
repos
up
that
people
were
using
for
for
Docs
and
samples
that
are
getting
a
bit
stale,
and
so
basically,
what
I
wanted
to
do
was
get
see
if
I
could
get
some
help
moving
those
somewhere
else.
B
B
And
so
this
one
I,
don't
think
Michaels
on
the
call,
so
this
one
I
had
done
with
with
the
ham,
charts
and
stuff
like
that
and
so
we're
actually
mr.
pirate.
So
I
guess
the
question
is
like:
do
you
think
it
makes
sense
to
move
a
collection
of
like
windows,
specific
samples
and
docks
to
like
another
repo
under
kubernetes
SIG's?
Or
is
there
somebody
that
could
help
change
stuff?
So
it
fits
the
style
of
kubernetes
website,
because
I
know
that
they
have
a
very
specific
way.
B
They
want
stuff
organized
and
that
would
become
probably
a
bigger
project
than
I
am
able
to
do,
and
so
I'm
thinking
that
out
of
these
Doc's
that
we
want
to
keep.
They
probably
need
to
go
to
one
of
those
two
places
unless
someone's
got
another
place
for
them.
So
has
anybody
looked
at
that
at
this
list
so
far.
A
C
But
ultimately,
what
the
thing
is
that
you
know
it
makes
sense
for
us
to
actually
maybe
create
the
sequin
those
samples
project
under
kubernetes
6
and
and
drop
them
in
there.
So
now
we
have
the
cig
windows
tools.
It
doesn't
really
make
sense
to
actually
have
it
under
the
tools
or
in
Windows
testing
either.
So,
as
you
might
make
sense,
to
actually
create
a
sequin.
Those
samples
I
can
kick
off
the
workflow
to
see
how
to
data
create
it.
Okay,.
B
C
B
B
D
B
D
Basically,
I
was
trying
to
look
for
a
Windows
example
and
examples.
I
don't
see,
one
I
mean
there's
no
way
for
us
to
run
any
example,
app
any
move
your
fabric
em
into
the
IO,
not
examples
and
say
that
this
is
Windows.
It's
a
kind
of
a
cross,
first,
application
that
works
for
both
Windows
and
Linux
right.
B
B
D
B
A
B
B
They
had
some
pretty
specific
style
feedback
and
I.
Don't
think
I've
got
the
time
for
that
kind
of
back
and
forth
on
it
right
now,
and
so
I
could
do
the
first
step
to
get
it
like
up-to-date
and
working
and
move
it
to
another
repo,
but
I
think
I
would
want
someone
else
to
own
the
PR
into
website
so
that
way
update
they
could
do
the
updates.
Does
that
make
sense.
A
B
B
A
B
B
C
B
B
C
C
B
B
B
E
A
A
F
It's
still
have
time:
Oh,
nobody
have
each
like
topics.
I
have
a
kind
of
a
like
question.
Is
that
okay,
sure.
F
So
this
make
today
inaccessible-
and
this
is
common
case
when
you
clone
this,
could
I
say
you,
take
snapshot
of
alternatives
and
then
create
another
disk
based
on
the
from
the
snapshots.
They
will
have
the
same
signature
and
so
for
a
second
disk.
If
attached
to
the
same
node,
you
have
to
manually
make
it
online
I'm
just
wondering
anyone
have
some
knowledge
about
this
issue.
F
So
I
think
they
notice
this
issue
or
like
especially
when
they
clone
the
disk
right.
So
if
you
take
that
out
and
then
create
another
disk
from
the
snapshot
and
then
you
happen
to
attach
the
disk
to
a
same
node
yeah,
so
you
know
this
right
then
these
two
days
who
will
have
the
same
fiscal
signature
exactly.
E
Yeah
that
makes
yeah,
so
this
is
done
specifically
for
the
reason
that
you
just
mentioned
about
these
signatures,
conflicting
so
Windows
I.
Think
the.
If
you
look
at
the
storage
stack
in
Windows
I
believe
it's
the
partition
manager
that
cannot
really
handle
two
disks
that
have
the
same
MBR
or
GPT
signatures
and
since
the
rest
of
the
storage
that
kinda
depends
on
each
disk
getting
its
own
individual
signature.
E
What
the
partition
manager
does
is
it
assigns
one
of
the
snapshot
disks
so
that
you
do
not
end
up
in
this
conflicting
scenario
where
we
have
multiple
online
disks
with
the
same
signature?
So
if
you
look
at,
if
you
just
like
search
online
for
something
called
San
policy,
I
think
that
was
the
policy
that
was
specifically
implemented
to
deal
with
this
scenario,
where
you
know,
if
it's,
if
it's
enabled,
which
is
the
standard
for
server,
then,
if
like
whenever,
the
partition
manager
sees
a
new
disk
device
popping
up
in
this
in
in
the
storage
stack.
E
If
it
detects
like
there's
a
duplicate
signature
about
to
be
online,
then
it
goes
ahead
and
basically
off
finds
the
disk.
So
that's
what
does
an
policy
cover
governs
and
that
mode
I
think
is
called
offline
shared
and
you
can
also
set
it
to
something
called
like
online,
all,
basically
which
says:
hey
partition
manager?
If
you
see
like
this
a
separate
this
device
popping
up
and
it
does
see
the
same
signature
is
something
that's
existing.
It
will
automatically
go
and
assign
me
a
fresh
new
signature
to
the
disk.
F
E
F
F
E
E
E
A
E
G
E
Yeah
so
I
think,
like
long
back,
I
was
looking
at
it
from
like
you
work
for
VSS,
and
basically
that
was
like
a
snapshot
service
that
Windows
used
to
implement
and
what
the
storage
vendors
would
do
is
as
part
of
the
snapshot
call
when
they
are
taking
this
replica
of
the
disk.
They
would
make
sure
that
the
replica
has
a
different
signature
so
that
they
do
not
hit
this
duplicate
signature
issue
so
like
in
the
context
of
CSI
jiaying.
E
It
would
be
that
whatever
is
a
CSI
provider
handling
the
snapshot
call
that
provider
can
make
sure
that
the
snapshot
is
generating
in
the
context
of
Windows
at
least
has
a
different
signature
who
has
a
different
signature
from
the
original
disk,
so
everything
else
is
obviously
same
kind
of
fight
for
right,
but
the
signature
needs
to
be
changed.
If
that
is,
if
you
want
to
kind
of
make
it
come
online
by
default.
I.
H
Don't
know
if
it
helps
Cobb
is
a
neat
already
sets
their
descent
policy,
so
it's
already
something
that
can
be
done
there
and
in
order
to
keep
it
says,
keeping
an
XML
executable
or
calling
some
powerful
API.
We
went
directly
with
a
native
API
that
the
operating
system
offers.
So
this
is
some
theater,
of
course,
come
for
free
if
you
use
like
class
the
API
after
that
with
Club
is
in
it
or
it
could
be
just
lifted
off
from
there
and
reused.
In
any
other
context,
oh
cool.