►
From YouTube: Velero Community Meeting - November 23, 2022
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
Okay,
hello,
everyone:
this
is
with
our
community
meeting
on
November
23rd.
First,
let
me
do
some
status
update
for
1.9.3.
We
have
G8
8
on
November,
11th
and
for
way
one
per
10.
We
are
planning
to
create
the
rt2
today
November
23rd,
and
we
have
made
some
very
small
training
and
some
new
tested
cases.
A
So
we
want
to
wait
and
more
interested
in
the
pipeline
today,
I
mean
the
nightly
papad
today,
I
know
if
everything
is
okay
with
a
great
attitude
and
for
for
the
net
not
release
way
1.11,
we
are
reviewing
the
candidates,
I
mean
of
all
the
issues
that
always
think
it
is
valuable
and
suitable
to
make
it
into
the
natural
level
put
the
candidate
labels
into
that,
and,
of
course,
on
top
of
that,
the
data
moment
it's
it's
with
the
high
priority.
A
And
for
individual
update
myself
with
prepared
in
one
content
release
and
also
we
have
started
another
turn
of
manual
tests
and
I
have
done
some
many
tasks
after
rc1.
That's
from
my
side
and.
B
D
And
I
had
to
fix
some
issues
regarding
documentation
to
make
sure
that
you
know
merged
into
one
hand.
In
addition,
here,
I'm
working
on
the
White
Album
any
of
the
team,
so.
B
My
part
is
I've,
been
working
on
a
poster,
I've
C1
menu
test
and
do
some
nightly
on
1.10.
That's
all
for
me.
E
Yeah
so
I've
updated
the
item.
Action
plug-in
design
based
on
some
of
the
discussion
a
couple
of
weeks
ago,
to
kind
of
add
some
flexibility
around
different
plugins
in
terms
of
being
able
to
report
progress,
invites
or
other
units.
So
we
added
a
field
for
units.
E
We
renamed
the
fields
to
make
it
less
ambiguous
around
that
because
it
was
confusing
before
with
you
know,
items
to
complete
and
you
know
water
items
and
and
then
we
also
added
a
description
field,
because
there
may
be
some
plugins
that
aren't
able
to
give
their
progress
in
a
percentage
basis.
E
You
know
maybe
they're
interacting
with
a
an
external
controller
that
just
has
you
know,
uploading
phase
and
clean
that
phase
or
whatever
you
know,
just
some
text
descriptions,
and
so,
though,
that
can
be
placed
into
that
text
field
the
description
and
if
a
plug-in
is
not
using
the
numerical
progress
and
those
will
just
be
left
blank,
so
there'll
be
zero,
so
it'll
say
zero,
zero,
meaning
we
have
no
entire
Nation.
E
You
know
in
terms
of
progress,
but
there
will
be
that
description.
Field
available.
I
have
also
updated
the
design,
API
design
PR's,
to
reflect
that
one
change
as
well.
Again,
let's
get
the
the
overall
design
wrapped
up
in
the
food
first
and
then
we
can
look
at
the
individuals
and.
B
E
D
Yeah
yeah,
we
have
resolved
most
of
the
comments
and
I
think
I
have
one
or
two
regarding
the
candle
I'll
go
through
them
today
and.
E
Okay,
hopefully
yeah
that's
fine
and
cancels
was
when
we
haven't
talked
about
it
as
much
so
there's,
maybe
some
room
to
make
changes
there
as
needed
and
we'll
I
linked
the
other
ones
there,
mainly
because
I
did
make
the
changes
around
the
operation.
Progress,
but
again,
let's
focus
on
the
main
one.
First,
because
once
that's
approved
and
we
got
all
the
changes
in
and
then
those
get
updated,
I
think
those
more
detailed,
API
specific
designs
should
be
relatively
straightforward
to
you
know,
review
and
get
going
on.
E
E
And
then,
like
I
said,
if
there's
additional
comments
you
have
on
cancel,
we
can
talk
about
that.
If
there's
anything
else
needed
there,
but
we
can
go
through
that.
E
Yeah
yeah,
the
U.S
offices
that
we
have
Thursday
and
Fridays
is
holidays,
so
I'm
here
well
yeah.
So
so
tomorrow
you
know
for
me.
So
I
have
one
more
day
this
week,
I'm
here
so
anything
you
get
to
today.
I
can
respond
to
tomorrow
and
then
I'll
I'll
next
be
back
on
Monday.
A
Okay,
okay,
thanks
God
sure,
and
do
we
have
any
other
things
to
update,
probably
from
anyone.
A
So
first
one
is
it's
about
the
two
data
moment
issues.
So
as
I
just
mentioned
of
way,
one
point:
eleven,
the
data
moment
is
still
with
the
the
the
top
priority.
A
So
we
we
discussed
it
locally
locally
and
we
are
so
we
discussed
it
locally.
So
I
created
two
issues
about
the
discussion
about
further
meeting
part
of
the
design
that
maybe
we
are
not
currently
covered.
So
maybe
we
can
cover
it
with
1.11.
A
The
first
one
is
it's
a
data,
it's
a
data
path.
That
is
how
the
how
we
move
the
data
for
velard's
data
moment
and,
for
example,
here
we
have
listed
some
questions
that
need
to
cover
need
to
be
covered
in
the
in
the
design
and,
for
example,
after
some
Charlie
taken
how
to
expose
the
sample
to
the
data
movement
pod
and
how
many
Central
will
be
exposed
to
one
that
amount
part
that
will
decide
the
the
parallel.
A
What
is
the
parallel
processing
like
and
after
that?
How
do
we
call
the
copia
uploader
and
call
the
the
unified
repo
and
and
then
write
the
data
that
should
be
something
covered
in
the
in
the
in
the
univariable
design,
but
we
need
to
clarify
it
overall
from
this
new
CR
scenario
and
also
like
for
for
the
file
system
data
movement.
A
How
do
we
organize
the
data
in
the
in
in
the
Repository
and
how
do
I
connect
or
initialize
the
repo
before
or
before
the
backup
result,
and
also
we
have
another?
Besides
the
file
system
data
Movement,
we
we
will
also
create
the
block
that
blocked
level
data
moment.
So
so
I
believe
this
will
be.
A
You
know
you
know
that
the
design,
but
we
need
to
mention
it
and
in
the
current
in
the
next
design,
and
and
also
make
a
leave
a
open
door
to
it
in
the
design
and
the
result
part
is
similar
to
the
backup
part.
That
is,
for
the
data
path
of
the
data
moment,
and
the
other
thing
is
something
something
like
that
way.
A
A
We
further
create
a
workload
pause
to
be
resolved
and
then
we
we
need
to
prevent
PVS
or
like
a
lot,
the
the
system,
the
community
system
present
QA.
All
we
created
people
ourselves
and
then
we
hold
the
workflow
are
parts
and
then
explode
is
post.
The
pause
to
the
data
movement,
the
the
expose,
the
PVS
to
the
data
movement
pause,
and
here
we
may
have
two
candidates.
Like
the
part,
one
backup.
A
We
assess
the
pause
on
the
host
that
is
through
the
host
pass,
or
we
attach
the
the
part,
the
PVS
to
be
resolved
to
the
data
one
part
and
then
inside
the
data.
One
part
we
call
the
related
uploader
to
copy
the
data
from
the
backup
repository
and
finally,
we
release
the
the
the
workload
pause
and
and
the
result
will
be
finished.
A
That
is
a
very
brief
steps
and
we
have
quite
a
lot
of
details
to
cover
if
we
want
to
go
to
this
approach,
and
currently
there
are
maybe
two
candidate
approaches
to
fulfills
or
implement
this
generic
volume
restore
the
first
one.
A
Is
we
create
our
own
like
that,
should
be
something
similar
to
The
Current
Port
volume,
backup
output
volume
result
merely
the
positive
volume
result
is
using
the
first
method,
that
is
through
the
host
path,
and
that
will
require
the
the
route
privileged
and
the
other
part
the
other
candidate
is,
is
accumulative
volume
populator.
This
is
still
in
the
beta
phase
of
I
mean
this
feature.
So
if
we
take
this
method,
it
means
we
will
only
support
the
newer
versions
of
the
kubernetes.
A
So
why
do
we
want
to
this?
Generic
generic
volume
result
so
some
reasons
some
benefits
the
first
one
is.
This
solution
could
work
for
all
kinds
of
backup,
for
example
the
CSI
backup
and
phone
volume
backup,
and
also
we
may
create
some
direct
access.com.
For
example,
we
read
data
directly
from
the
AWS
EBS
API
and
then
create
the
restore
and
this
for
this
back
kind
of
backup.
We
can
also
use
this
generic
restore
so
any
kind
of
backup,
and
we
are,
we
will
add
in
the
future.
A
The
second
second
one
is
the
second
one
is
because
this
generic
volume
result
doesn't
really
rely
on
any
online
story
to
snapshot
mechanism,
for
example,
about
40
000,
because
we
will
not
need
to
create
the
the
CSI
volume
assumption,
content
and
volume,
some
short
objects.
So
it
is
clear
and
windows
and
Below
do
need
to
maintain
that
object,
and
also
we
up
also.
A
We
don't
need
to
to
leave
that
object
in
the
in
user's
production,
environment
after
result
and,
finally,
for
the
same
same
reason,
because
it
doesn't
rely
on
any
online
stories
and
from
mechanism,
theoretically,
the
suppose
heterogeneous
result.
That
is,
we
can
back
up
in
one
environment
and
we
saw
it
in
a
very,
very
different.
C
C
C
C
D
Yeah
I
have
some
problem
about
his
microphone.
C
A
Okay-
okay-
sorry,
sorry
so
I
I
have
almost
a
down
the
introdu
Youtube
division
for
the
the
two
issues.
So
I'm
not
sure
if
you
have
started
the
discussion,
so
are
there
any
questions?
A
All
I
think
we
I
just
want
to
make
a
start
for
this.
This
to
I
mean
issues
or
or
requirements
for
the
data
movement.
We
can
make
the
discussion
anytime
offline
or
we
if
we
want,
we
can
revisit
this
this
in
the
next
community
meeting,
but
right
now,
if
we
want,
if
we
have
some
questions,
we
can
start
here.
A
Yeah,
actually
it's
the
volume
online,
the
volume
populator
Works
in
the
same
way
as
the
first
approach
I
mean
the
the
way
we
create
in
our
own
in
about
ourselves,
because
it's
just
something
like
the
the
about
the
the
steps
like
this,
the
volume
the
population
just
help
us
to
do.
This
two
points
on
hold
the
workload
and
release
the
workloads
and
also
it
transfers
the
how
to
transfer
the
right
or
access
to
to
to
the
to
the
populator,
Plugin
or
public
library.
A
That
will
do
the
data
copy.
So
it
is.
It
helps
us
to
to
be
to
to
implement
this
stuff
more
easier,
but
we
will
have
to
wait
for
its
ready
in
cumulative.
E
So
this
is
more
of
a
general
comment:
I
don't
have
detailed
feedback
here
at
this
point,
but
I
know
shubham
had
produced
PR
for
a
design
around
a
high
level
kind
of
discussion
around
design
for
data
movement.
That
I
think
focus
more
on
the
CSI
use
case.
I,
don't
know
if
you've
looked
at
that
one
to
make.
D
E
See
if
this
is,
if
this
is
this,
going
to
be
something
that's
in
addition
to
that,
or
would
this
also
meet
those
requirements?
It's
a
pull
request.
Number
4768.
It
has.
It
hasn't
been
changed
most
recently,
but
I
think
the
original
pull
request
was
from
maybe
back
in
March.
That
was
in
March
on
not
in
the
office
this
week,
so
he's
actually
gone
until
10th
of
December.
That
I'm
sure
he'll
be
interested
in
looking
looking
at
this
design
when
he
does
return
so.
E
Was
that
was
the
the
thinking
and
we
see
we
I
guess
there's
two
there's
two
things:
we
we
have
a
kind
of
interim
data
mover
Solution
on
the
as
part
of
oadp,
that's
built
on
top
of
Valero,
but
not
you
know,
obviously
upstream
and
then
we,
the
hope,
is
eventually
to
take
to
to
rely
on.
You
know
upstream
and
Valero
functionality,
and
then
we
wouldn't
need
that.
E
You
know
long
term
and
I
think
that
was
kind
of
part
of
what
was
driving
his
original,
the
design
doc
that
was
posted
in
that
PR.
But
again,
but
yeah
he's
he's
going
to
be
back
in
the
office,
I
think
on
the
10th
of
December.
D
E
Yeah
exactly
that
was
that
was
my
concern
is
to
make
sure
that
you
know
if
there's
two
different
things
here,
we
we
kind
of
get.
You
know
everything
covered,
and
you
know
if
there's
use
cases
that
you
know
that
his
approach
solves,
and
this
isn't
that
we
need
to
find
a
way
that
kind
of
resolves
all
that
you
know,
because
obviously
the
the
CSI
snapshot
use
case
has
got
different
needs
versus
the
native
snapshots
versus
copia,
and
so
it's
just
a
question
of
figuring
out
the
best
way
forward.
D
I
think
the
priority
is
the
same.
I
think
the
first.
If
we
want
to
handle
specific
or
concrete
one
is
the
CSI
yeah,
but
Young
play
is
talking
about
some
idea
that
we
don't
need
to
create
the
the
PV
in
order
to
restore
it,
we
can
just
move
the
data
into
what
people
and
things
are
done.
I
think
that
may
work
yeah.
A
Yeah,
so
for
the
for
the
so
I
think
these
two
issues
or
or
these
two
designs
is
it's
exactly
compliant
with
the
the
the
the
the
requirement
to
PR
that
has
been
merged
or
submitted
by
high
subang,
and
this
is
some
sub
work
items.
That
is,
we
make
a
concrete
design
for
the
requirement
that
I
think
the
relationship,
and
it
definitely
like
this.
The
two
ideas
work
for
all
the
of
all
kinds
of
data
movements.
A
So
it
will
work
for
the
CSI
data
moment
and
then
what
we
will
I
think
and
maybe
when
suban
comes
back
and
we
will
discuss
to
create
some
more
dedicated
work.
Item
designs
like
this,
like
these
two
ones,
for
example,
hallway
grade
and
start
the
backup
data
movement,
controller,
I,
I,
think
that
had
been
covered
by
by
by
the
demo
that
had
been
demonstrated
by
subang
a
couple
of
amounts
ago.
A
So
and
we
will
have
some
more
design,
I
mean
the
sub
item
designs
and
besides
these
two
ones,
for
example
the
backup
controller
and
something
like
that.
And
finally,
we
will
we
have
the
requirements
PR
and
we
have
the
sub
item
design
PR,
and
then
we
will
be
able
to
start
work.
I
think
that
is
the
relationship
it
doesn't
have
any
confliction
with
the
existing
are
coming
to
work
out.
The
data
movement
and
a
slightly
different
is
here.
A
I
think
this
one
is
slightly
different
with
the
demo
from
subang
and
that
the
result
will
not
require
on
the
the
CSI
some
shot
and
that
part
I
think
when
subang
comes
back,
we
can
have
more
discussions
and
to
decide.
You
know
whether
it's
it
is
suitable
to
use
this
kind
of
idea.
D
A
This
one
is
yeah,
and
this
one
we
will
need
to
have
more
discussion
and
because
that
is,
you
know,
approach
selection
and
we
use
the
CSI
way
or
we
use
a
generic
way
right.
A
Okay,
so
let's
continue
this
discussion
and
and
offline
and
especially
when
subang
comes
back,
and
maybe
we
no,
for
example,
we
can
create
some
design
and
in
the
way,
1.11
not
only
these
two
ones,
but
all
the
sub
items
that
is
required
by
the
by
the
data
movement
right
and
then
we
make
things
to
be
concrete,
and
then
we
can
start
the
coding
or
work
after
1.11
consider
the
time
frame.
Consider
11.
A
B
A
Okay,
that's
from
my
part
and
so
the
next
one
Daniel
yeah.
D
That's
what
I
want
to
discuss
with
I
want
to
discuss
with
Scott
that
I
thought
you
guys
mentioned.
You
want
a
relatively
short
release
in
1.11
to
make
sure
the
item.
The
progress
monitor
is
delivered
sooner
right.
So
do
you
have
any.
E
Yeah
I
mean
that
that
was
kind
of
the
the
main
thinking
you
know
just
kind
of
at
a
high
level
of
if
we're
talking
about
you
know
the
number
one
thing
we
talked
about
recently
was
you
know
about
you
know
we
we
want
to
get
to
the
point
where
we're
having
you
know
three
Belair
releases
a
year.
So
that's
about
four
months.
You
know.
B
E
This
is
going
to
be,
you
know,
one
that
might
be
a
little
bit
shorter
than
that.
You
know
that
would
be
good
in
our
case,
just
because
you
know
we're
going
to
be.
You
know,
given
that
the
full
data
mover
is
not,
you
know
completely
in
place,
and
you
know
all
that
yeah
the
data
mover
that
we
that
we're
building
If
part
of
oadp
on
top
of
Valero
is
going
to
be.
E
You
know
we
need
that
item
action,
progress,
work
so
that
we
don't
have
to
block
you
know
on
the
backup
and
restore
you
know,
while
we're
waiting
for
those
uploads
and
things
to
finish
so
that's
kind
of
what
our
thinking
was.
E
You
know
if,
if
we
could
scope
through
at
least
certainly
include
that
and
whatever
else
you
know
we
fit
into
in
the
right
time
frame,
but
that
it
would
be
better
to
get
the
release
out
relatively
quickly
rather
than
adding
a
whole
bunch
of
extra
scope,
and
you
know
dragging
it
out
longer.
D
So
so
do
you
have
any
deadline
in
your
mind
or
any
proposed
date,
I'm
I
used
to
think
we
want
to
have
it
like
February,
early
March,
but
but
yeah
for
deep.
The
PM
told
us
that
you
are
also
okay
in
April
May,
so
I
think
February
or
March
is
better
right.
E
Yeah
I
wanted
to
talk
to
Dylan
about
that,
because
I'm
he's
clear
on
you
know
kind
of
product
deadline,
sides
of
things
and
I'm
not
as
much
off
on
that
at
this
point.
But
I
could
talk
to
Dylan
and
we
can.
You
know
either
get
back
to
you
offline
or
next
meeting,
though
yeah.
D
A
Okay,
thanks
and
anything
else,
we
want
to
discuss.
A
Okay,
so
if
now,
we
I
think
we
can
finish
up
here
today
and
thanks
for
attending
and
have
a
good
day
and
a
good
evening
thanks.