►
From YouTube: Velero Community Meeting - Feb 1, 2023
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).
A
A
Hello-
and
this
is
meeting
on
February
1st
for
the
overall
status.
A
Started
to
come
back
from
the
holiday
and
some
members
will
be
back
next
week,
so
the
for
wave
1.10.1,
we
have
created
the
RT
build
before
the
holiday,
and
this
week
we
will
prepare
the
ga
and
41.11
development
items
are
basically
on
track
and
for
the
building
data
movements.
We
have
finished
the
first
round
discussion
and
we
have
made
the
agreement
and
on
on
some
key
factors
and
we
will
create
a
well
create
another
discussion
for
the
last
of
my
open
classes.
A
From
my
site,
I
working
on
the
follow-ups
on
the
on
the
data
model
discussion
try
to
follow
up
the
questions
or
concerns
we
have
made
during
the
discussion
and
I
also
create
a
new
version
of
the
Snapchat
backup
and
some
results
here
hope.
Hopefully
it
is
more
generic
and
it
is
more
generic
and
we
can
discuss
that
on
the
next
discussion.
B
Two
atoms,
the
second
one
is
working
on
some
1.11
related
issues
and
the
first
one
is
the
steel
going
on
the
the
design
work
of
the
self-creating
namespace
and
the
cluster
resource
filters.
I
I
saw
Scott's
comments
and
even
also
joined
the
meeting.
Maybe
we
need
to
take
some
time
to
go
through
the
proposals
and
the
events
proposals.
D
A
quick
follow-up
from
my
comments:
I
I
had
some
comments
saying
that
at
an
earlier
version
of
this
design,
we
were,
we
were
suggesting
that
the
include
cluster
or
include
namespace
resources
that
you
would
need
to
specify
star
to
include
them
versus
the
default
and
I
think.
The
last
comment
that
you'd
had
was
that
we
we
can
exclude
everything
with
the
exclude
star
and
I
think
my
only
comment
there
was
with
the
current
implementation.
D
B
The
way
we
may
think
maybe
make
the
included
as
the
default
behavior
is
for
users
to
better
understand
how
it
works
and
it's
getting
reasonable
existing
parameters,
and
for
for
your
question
to
how
to
exclude
resources,
the
change
is
to
make
the
real
card
also
work
for
the
excluded.
Okay,.
D
Yeah,
that
makes
sense,
because,
because
right
now,
it
only
works
on
include,
but
if,
as
long
as
the
implementation
of
this
includes
making
it
work
on
the
exclude
side
as
well,
I
think
that'll
solve
the
the
same
use
case
manually.
Fine,
okay,.
E
How
about
yeah,
yeah
I
think,
like
the
only
question
I
mean
yeah
once
again,
thanks
for
all
the
pointers
and
feedbacks
and
link
to
the
existing
issues,
I
mean,
like
I,
took
a
look
at
all
of
them
like
prior
to
adding
a
comment,
but
I
guess
that's
just
too
much
information
there
again.
Now,
thanks
for
the
refreshment,
I
think,
like
my
my
main,
like
you
know,
like
questions
or
slash
hesitation
around
it
is
like
you
know,
the
deprecation
of
the
existing
like
Flags,
especially
like
around,
include
cluster
resources.
E
How,
like
you
know,
like
users
like
just
get
used
to
like,
depending
on
that
behavior
with
this
Auto
direct
up
inference
of
like
oh
we're,
gonna
go
get
the
PB,
because
the
PVC
is
in
your.
You
know,
namespace
that
type
of
behavior
so
like
I,
know,
like
the
the
belief
right,
I
believe
like
the
default.
If
we
do
deprecating
the
other
extinct
flags
and
Behavior.
D
I,
actually,
don't
think
we're
going
to
deprecate
the
existing
ones.
I
think
the
two
are
going
to
live
side
by
side.
Basically,
you
have
to
use
one
or
the
other,
so
you
can't
mix
all
seven,
but
basically
we'll
have
a
validation
that
says
you
can
either
use
the
old
Flags
or
the
new
flag.
So
existing
users
that
were
happy
with
the
old
functionality,
I
think
I,
don't
think
we
have
to
get
rid
of
that.
I
mean
deprecations,
obviously
a
featured
a
decision
anyway,
because
we
wouldn't
deprecate
it
immediately
and
I.
D
Think
there'd
be
some
discussion
about
whether
we
might
want
to
deprecate
it
or
not.
I,
don't
think
that's
a
closed
discussion.
I
just
was
going
to
say
one
other
thing
about
the
you
know.
The
auto
include
where
you
pull
in.
You
know
just
the
PVS
for
the
PVCs.
D
D
If
you
listed
in
the
include
cluster
resources,
then
we
include
all
of
that
resource
type.
If
you
listed
in
the
excludes,
then
we
include
zero
of
that
type,
but
any
cluster
scope
resources
it's
not
listed
and
either
then
we
follow
the
old
Auto
Behavior,
but
instead
of
being
an
All
or
Nothing
Auto
like
it
is
now
so,
for
example,
you
could
list,
you
know
cluster
roles
as
include
cluster
resources,
and
then
you
don't
list
PVS.
So
that's
going
to
be
the
auto
Behavior
you're
pulling
in
just
the
relevant
ones.
D
That's
a
use
case
that
you
can't
handle
with
the
old
ones
where
the
only
way
to
include
all
cluster
roles
would
be
to
include
all
cluster
script
resources,
including
all
PVS.
You
know
which,
which
we
probably
don't
want
to
do
for
most
cases.
E
Right
so
so,
basically
that
it
really
boils
down
to
like
just
modifying
the
include
Behavior
to
be
more
like
I
guess,
addictive
Edition
kind
of
behavior
instead
of
exclusively
include
what
is
on
the
list
right,
yeah.
D
Yeah,
it's
it's
a.
This
is
an
area
where
we're
going
to
have
to
I
mean
the
implementation
is
going
to
have
to
have
clear
documentation
as
part
of
it,
because
there's
a
there's,
a
distinction.
There's
the
namespace
resources
which
works
just
like
you
would
expect
from
the
old
behavior
you
process
the
includes,
and
then
you
subtract
the
excludes
and
then
that's
your
list
for
a
cluster
scope
resources.
We
actually
have
three
categories.
Anything
in
the
includes
explicitly.
D
We
include
everything
of
those
anything
in
the
excludes
we
completely
exclude
and
any
resources
that
are
not
mentioned
in
either
follows
that
default.
Behavior,
like
we
see
with
you,
know
the
old
parameters
where
we
include
just
relevant
resources,
so
just
the
PVS
for
the
PVCs
just
crds.
If
there's
a
CR
that
kind
of
thing,
okay,.
E
So
in
the
case
of
like
I
believe
like
in
last
week
meeting,
we
talked
about
like
customer
resource
definitions
being
right
that
special
thing.
D
Right,
so
if
you
only
want
to
include
the
crds
or
the
CRS
you're
backing
up,
then
you
would
just
not
list
custom
resource
definitions
in
the
cluster
resources
list.
But
if
you,
if
you
have
a
use
case
where
you
want
to
include
all
of
them,
because
you
know
you're
backing
up
something
and
you
need
your
crds
because
maybe
they're
temporary
resources-
and
so
you
don't
have
any
to
back
up,
then
you
would
list
customer
resource
definitions
in
the
include
list
and
then
that'll
back
up
all
the
crds
in
the
cluster.
E
Okay,
okay,
so
yeah
I
think
the
main
thing
there
is
still
like,
so
that
that
would
that
that
would
like
soft,
like
a
problem
like
70
80
off
the
way
I
think
part
of
it
is
kind
of
going
back
to
our
previous
discussion
around,
like
you
know,
like
I
guess
like
from
you
know,
I
don't
want
to
make.
You
know
basically
like
I'll
use
it
you
like
get
on
the
back
of
all
the
crds
right.
It's
like
okay,.
D
And
that's
where
the
question
of
you
know
adding
some
future
mechanism
of
you
know.
Naming
resources
comes
up,
you
know
and
I
know.
There
have
been
a
couple
of
proposals
around
that
you
know
and
which
is
I,
guess
part
of
what
you're
talking
about
here,
because
because
right
now,
Valera
doesn't
have
anything
like
that.
D
You
know
there's
ways
of
including
and
excluding
by
type
and
namespace
and,
and
you
know
the
label
selector,
which
is
kind
of
difficult
to
work
with,
because
but
and
then
that's
why
we,
you
know
we've
seen
various
proposals,
for
you
know,
I
want
some
way
of
including
Resources
by
name
and
I.
Guess
the
challenge
that
we
have
to
work
through
with
any
of
those
proposals.
Is
you
know
again?
How
do
we
do
that
in
a
way?
That's
not
confusing
to
the
user
in
the
way
it
interacts
with
the
existing.
D
You
know,
resource
based
and
namespace
based
and
all
that,
and
if
there's
a
way
of
doing
that,
you
know
that
we
can
come
up
with
they
kind
of
can
and
I.
I
would
think
that
would
be
in
addition
to
this,
because
these
use
cases
that
we're
kind
of
trying
to
solve
here
don't
go
away.
If
we
add
the
name,
you
know
named
resources,
because
I
think
there
are
valid
cases
that
this
solves
that
we
need
this
for.
D
But
you
know
to
your
point:
there
are
other
cases
where
we
want
to
say:
hey
I
want
just
this
list
of
crds
whether
or
not
we
have
resources
you
know,
CCR
is
for
them
and
that
that's
kind
of
would
call
for
a
a
named
resource
inclusion,
so
not
just
the
type
but
the
type
and
the
name
and
the
namespace
or
whatever
and
I,
think
we've
had
some
issues.
You
know
and
we've
had
some
discussions
around,
starting
that,
but
again
that's
another
case
that
it's
going
to
be
like
this
one.
D
E
Actually,
like
you
know,
I,
you
might
remember,
I
think
that's
actually
an
existing
issue
around
that,
like
the
resource
by
name.
D
Yeah
yeah
I
know
if
someone
that
was
mentioned
last
week
and
I
think
I'd
even
commented
on
that
one
last
year
when
it
came
out
it's
kind
of
initially
and
then
there
hadn't
been
a
lot
of
activity
on
that
lately.
So
that
might
be
an
issue.
That's
you
know,
can
be.
What
could
be
brought
up
again
is
something
to
think
about
in
a
sense
again.
D
D
I
mean
I,
mean
I,
think
we
can
do
it
at
the
same
time,
I
I,
I,
don't
know
that
necessarily
means
I
mean,
maybe
maybe
there's
a
case
to
be
made
to
say:
hey
it's
easier
to
do.
One
kind
of
you
know
Grand
unifying
design
here,
that's
easier
to
to
work
with.
It
also
might
be
easier
to
at
least
complete
this
design,
get
it
merged
and
then
follow
on
PR
that
modifies
this
you
know
design
as
to
it
I
just
I,
wonder
that
we
have.
D
We
already
have
so
much
in
play
with
this
design
and
all
these
changes
back
and
forth.
We've
been
working
on
this.
You
know
for
a
couple
months
now
that
I
think
to
throw
a
whole
bunch
of
new
scope
into
this
I
think
might
kind
of
derail
a
bit
of
the
work,
yeah
I
think
I'd
like
to
see
this
design
merged
and
then
a
follow-on
design.
Maybe
even
the
modifies
this
document.
You
know
that
kind
of
kind
of
adds
in
the
other
use
cases
like
that.
E
Yeah
I
think,
like
I,
fully
agree
with
you
on
that
I
actually
added
a
comment
to
like
that
issue
around
like
the
name
featuring
and
stuff
like
that
yeah
definitely
like
I.
You
know
like
this.
This
proposal
right
here
on
the
screen
like
if
I
saw
this
already
like
I'm,
pretty
involved
I,
don't
think
like
we
miss
you
later.
E
G
That
sorry,
but
I
guess
doing
the
implementation
together
makes
sense
right.
The
design
can
follow
one
after
the.
D
Other
yeah,
yeah,
probably
again,
I
think
that
probably
depends
on
you
know
what
design
we
come
up
with,
but
you're
right,
there's
a
you
know:
it
might
be
easier
to
implement
it
as
one
big
PR
that
does
everything
and,
depending
on
how
the
designs
interact,
it
might
be
easier
to
implement
the
PRN
sages.
You
know
I
think
that's
a
decision
we
can
make
after
we
have
an
approved
design
as
to
you
know,
hey
do
we
want
to
implement
this
thing
as
one
one
big
chunk,
or
do
we
wanted
to
phase
it
in?
D
You
know,
from
an
API
change
point
of
view,
one
change
and
one
release
to
get
it
all
together
makes
sense,
because
then
users
only
have
to
react
to
one
change.
Whether
we
get
there
incrementally
within
one
release,
cycle
I.
Think
is
a
question
we
can.
You
know
figure
out
later,
but
I
do
think
that
since
it's
a
related
functionality
change,
we
probably
want
to
implement
all
of
this
in
the
same
release
cycle
so
that
we
don't
have
to
you
know,
educate
users
on
a
new
way
of
selecting
items
for
backup
twice
right.
E
Right,
hopefully,
not
not
a
PR
with
like
two
two
thousand
lines
of
code
change
right,
so
so
yeah
I
I
noticed
that
you
know,
as
we
pointed
out,
that
that
issue
has
gone
coincale
like
is
it
okay?
If
I
reach
out
to
you,
like
I'm
offline,
to
see
if
I
mean
you
know,
I'm
happy
to
like
to
take
it
on
to
take
that.
D
On
yeah
and
I'll
move
that
forward.
That
makes
it
and
I
think
a
lot
of
times.
These
issues
go
stale,
because
you
know
someone
in
the
community
will
submit
an
issue
and
there's
an
active
discussion
back
and
forth,
and
then
maybe
the
original
you
know
reporter.
You
know,
priorities
changes,
no
longer
matters
to
them,
and
so
they
stop
responding
to
the
issue,
and
so
since
it
wasn't
something
that
Upstream,
you
know
that
we're
in
the
in
the
maintainers
you
know,
are
actively
working
on.
D
You
know
if
the
person
that
submitted
the
issue
has
stopped
responding,
those
issues
kind
of
stopped
getting
activity.
So,
if
you're
interested
in
this
now
it's
kind
of
taking
it
on
again
or
to
talk
about
it.
You
know
I,
think
I,
think
commenting
on
that
issue
probably
would
be
a
good
place
to
to
to
continue
that
discussion,
and
ultimately,
once
this
design
gets
merged,
it
might
make
sense
to
have
a
design
PR.
D
You
know
linked
to
that
issue,
but
actually
possibly
a
lot
of
possibly
modifying
this.
This
document,
or
or
an
add-on
document
I'm,
not
sure
it
kind
of
it
probably
depends
on
what
direction
you
know
we're
going
with
the
design
as
to
what
makes
sense
there,
but
even
if
it
is
a
separate
document,
we
just
need
to
make
sure
that
the
use
case
is
covered
in
terms
of
what
kind
of
API
changes
we
want.
You
know
handle
all
the
use
cases
here.
You
know
smoothly.
E
Yeah
for
sure,
okay,
yeah
I
actually
added
a
comment
to
the
issue.
Maybe
like
I'll
reach
out
to
you
like
offline,
to
go
into
more
detail
discussion
there
about
that
one
yeah
sure
so,
yeah
I
think
like
going
back
to
John
going
back
to
your
earlier
question
like
I.
Don't
at
this
point
like
yeah
yeah
like
I've,
my
I
guess,
like
my
only
like
I
guess,
like
ask
some
questions
around
this
proposal.
E
Right
now
is
just
like
if
we
do
end
up
like
deciding
to
deprecate
like
or
not.
If
we
do
decide
to
deprecate,
like
you
know,
Flags
like
include
like
class
of
resources
and
stuff
like
that.
Just
maybe
like
want
to
be
clear
about
like
how
the
application
process
is
going
to
be,
like
probably
mainly
take
slightly
more
than
two
releases,
because
like
user
has
gone
so,
like
you
know,
depending.
D
On
that,
we
actually
had
a
PR
out
to
a
design
PR
to
discuss
the
deprecation
for
us,
because
that
one
of
the
issues
is
that,
at
this
point,
I
don't
think
we
actually
have.
D
You
know
that
process
designed
explicitly
but
but,
like
you
said,
part
of
the
the
idea
here
is
that
you
know
when
you
you
know,
and
we
need
to
finalize
that
process.
You
know
in
the
governance
just
to
make
sure
that
we're
going
to
stick
with
it,
but
but
I
think
the
intent
is
exactly
that.
So
so
at
some
future
release
we
say:
okay,
you
know
in
1.12,
for
example,
we
deprecate,
you
know
the
old
parameters.
D
E
Know
yeah,
yeah
and
part
of
that
also
like
young.
Just
like
you
know,
coming
up
with
an
upgrade
path
for
the
user
right
I
mean
it
could
be
as
simple
as
just
changing
the
flag
on.
E
But
part
of
it
is
like
what
what
about
the
API
level
right,
like
you
know
meanwhile
like
and
what
is
that
translation.
D
Look
like
yeah
and
I.
Think
too,
and
this
is
another
process
question,
but
just
a
kind
of
specific
to
this
particular
issue.
I
think
I
think
it
makes
sense
to
release
them
side
by
side
and
get
feedback
from
users
to
say:
hey
just,
do
the
new
parameters
meet
your
needs?
Are
there
still
users
that
need
the
old
ones?
Let's
have
that
discussion
before
we
deprecate
anything
to
make
sure
that
you
know
if
we
are
planning
on
deprecating
and
we
find
a
way
that
you
know
to
you
know
to
meet
their
needs.
Yep.
E
So
yeah,
okay,
so
on
guess
what
yeah
so
like,
whatever
the
flex
and
options
are
going
to
look
like
I
guess
at
this
point
like
it,
it
really
doesn't
matter.
I
think
what
Islam
proposal
have
already
is
really
good.
I,
think
that
will
help
us
and
like
on
auto
user
as
well.
E
E
What
does
I
I'm
just
curious
like?
What
does
the
restore
generator?
Generally
look
like
you
know
like
what,
if
like,
if
especially
around
crd
right,
you
know
how,
like
someone
sometimes
like
they
have
like.
We
want
Alpha
One
in
one
backup
and
then
there's
a
we
want
beta
one
in
another
backup
and
then
during
restore
like
I
guess.
E
Would
they
just
end
up
like
stepping
on
each
other's
toes
like
yeah,
you
know
how
like
the
Clusters
scope,
resources,
they're
gonna
affect
the
entire
concert
right
and
then
it's
up
to
the
user
to
know
oh
yeah
in
this
backup.
You
know
taken
two
days
ago.
My
crd
is
V1
beta
one,
the
other
one
is
we
want
Alpha
One
I,
better,
not
roll
back
to
the
older
one
during
restore
right.
D
And
then
no
no
do
keep
in
mind
that
the
default
Behavior
with
the
phone
restore
is
that
Bolero.
If,
if
the
resource
exists
out
in
the
cluster,
it's
not
going
to
override
it.
You
know
if
so
polaro
only
restores
things
that
don't
exist
in
the
cluster
by
default.
If,
if,
if,
if
that
object,
is
there
already,
then
there'll
be
a
warning
on
the
restore
saying
the
resource
was
already
in
the
cluster?
There
is
an
option
to
get
Valero
to
try
to
patch
resources
unrestored,
but
that's
not
the
default.
D
G
There
are
a
lot
of
others
right
click.
Cluster
scope,
resource
that
will
get
over,
which
have
been
deleted
would
get
restored,
which
user
doesn't
want.
So
just
what
to
give
background
right
for
our
product
right,
customers,
backup
at
namespace
level,
that
is
the
unit
of
protection
yeah,
and
they
want
all
the
applications,
the
name
space
to
protect
it.
But
then
there
are
some
dependencies
at
the
cluster
scope
right
resources
so
do
I
have
to
backed
up
which,
which
is
more
or
less
solved,
with
that
nil
option
default
or
auto
option.
G
D
Yeah
yeah
right
right,
I,
mean
I,
think
that
was
kind
of
one
of
the
things
that
we
identified.
You
know
and
that's
why
we
have
that
other
issue.
Then
it's
the
summer
that
issued
in
it
before,
where
you
know
being
able
to
identify
specific
named
resources,
is
certainly
something
that
you
know
a
number
of
users
have
asked
for,
and
you
know
the
fact
that
the
existing
issue
is
kind
of
it'll
related
to
that
and
I
think
there
is
a
need
to
pursue
that
to
design
that
you
know.
D
E
Yeah
that
sounds
good,
okay,
yeah.
Why
don't
we
do
that
then
I'll,
just
you
know,
really
talk
to
you
offline
and
like
continue
to
comment
on
the
other
issues,
and
we
can
see
like
what's
the
next
step,
for
that
particular
feature.
E
Meanwhile,
like
once
again
yeah
this
one
this
this
proposal
here
on
the
screen
that
looks
great
thanks
for
all
the
hard
work,
I
think.
That's
all
for
me.
B
Thank
you
anyway.
My
my
part
is
thought.
F
Yeah
I'm
working
on
on
11
issue,
I'm
trying
to
add
the
detail,
the
resource
list,
when
user
run
the
Valero
will
still
describe
Dash
detail
command.
Just
we
did
for
the
Valero.
F
D
Okay,
so
yeah
the
I've
been
working
on
the
async
plugin
work
kind
of
the
controller
side
in
terms
on
the
backup
side.
I
don't
have
a
PR
yet,
but
I've
been
kind
of
putting
that
together,
I'm
hoping
to
get
a
PR
by
the
end
of
the
week,
the
you
know,
at
least
at
least
the
draft
ER,
if
it's
not
ready
to
review.
Yet
there.
C
C
A
And
father
did
start
topics.
I
just
have
a
little
query
that
for
the
second
round,
the
discussion
on
the
beauty
needed
more.
What
about
the
time
Netflix
like
at
the
same
time
and.
A
At
the
same
time,
with
the
community
Chinese
community
meeting
next
Wednesday
or
next
Thursday
is
the
time
okay,
for
you.
D
I
would
say
the
the
Wednesday
time
works
better
from
the
Thursday
time,
but
but
either
could
work.
Okay,.
A
So
that
happens.
Okay,
that's
I
will
talk
with
other
members
for
the
time
and
hopefully
we
will
we'll
have
the
meeting
last
week.
So
yes.
A
Not
I
think
we
can
finish
off
a
meeting
right
here,
thanks
for
don't
have
a
good
day
and
evening.
Thank
you.
Bye-Bye
thanks.