►
From YouTube: Velero Community Meeting - June 29, 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
B
A
Okay,
so
thanks
ming
and
everyone
for
the
contribution
in
the
you
know,
valero
one
nine.
Also,
the
biggest
change
in
one
line
is
we're
gonna,
you
know,
provide
official
support
for
the
csi
plug-in
for
aks
and
eps,
and
in
addition
to
that,
there
are
several
interesting
enhancements
to
the
velaro,
like
the
existing
resource
policy
and
the
resource
status,
flag,
etc.
A
Thanks
for
everyone
in
the
community
for
the
support
and
contribution
from
my
part.
In
addition
to
the
you
know,
the
the
ga
effort
with
the
team,
I've
been
doing
a
one-time
plan
and
reviewing
the
design
for
copia
integration.
We're
gonna
talk
about
the
110
plan
later
like
how
we
execute
on
the
key
milestones.
C
D
I'm
review
reviewing
the
backup
item
action
reflecting
pr,
and
I
think
this
is
the
part
of
the
work
to
support
the
plugin
version,
and
I
have
added
two
comments
and
I
also
do
some
issues
in
kitchen
for
the
copyright
integration
work.
E
Oh
okay,
currently
I'm
working
on
migration
in
test
and
and
it's
caught
up
by
some
other
work,
and
I
will
keep
on
working
it
later.
A
Okay,
so
I'm
gonna
quickly
move
to
the
discussion
topic.
First
thing:
I'm
gonna
discuss
the
110
plan,
so
we've
been
have
a
few
rounds
of
internal
discussion
so
for
one
hand
we're
gonna
execute
a
little
differently
from
a
previous
one,
nine
or
one
eight.
I'm
not
gonna
write
the
a
big
table
to
include
all
the
planned
issues.
A
We're
gonna,
give
more
flexibility
in
the
execution,
we're
gonna
make
sure
the
major
features
are
delivered
and
after
that
there
are
a
few
candidates,
we're
gonna,
keep
watching
them
and
making
adjustments
to
the
plans
until
the
feature
free
state
which
is
planned
at
august
in
end
of
august.
A
I
will
double
check
with
other
maintainers
around
the
state,
the
so
so
so
before
the
state.
We
may
add
new
issues
or
to
this
milestone
of
one
hand
and
potentially
may
move
something
out,
but
after
that
date
we're
gonna
focus
on.
What's
in
this
110
miles
long,
we're
gonna
use
this
one
as
a
single
source
of
truth.
As
for
you
know
what
we're
going
to
deliver
in
110,
so
you
won't
see
that
issue
list
in
this
row
map.
A
Instead,
you
see
a
few
items
so
for
one
hand
the
major
feature
we're
gonna
deliver.
The
anchor
feature
is
the
copia
integration
we
have
been.
You
know
discussing
around
the
design
for
a
while,
but
we
figured
because
we
want
to
support
both
rastic
and
cobia
for
a
few
versions,
so
the
refactor
work
and
the
upgrading
process
is
tricky.
We
feel
there
are
some
potential
issues.
We
still
need
a
bit
more
investigation,
but
we've
made
the
decision
that
we
want
to
deliver
this
cochlear
integration,
so
without
big
change.
A
A
There
is
a
pr
by
phone
to
do
the
initial
refactor
as
mankind
mentioned,
and
we
we're
gonna
double
check
with
someone's
guy
like
what
they
plan
to
add
into
one
hand
for
the
new
versions
of
plugins.
I
know
phone
is
planning
to
add
a
new
version
of
backup,
item
action
and
stuff
scott
mentioned
he
wanted
to
add
a
new
version
of.
I
think
it's
restored
as
an
action
plugin,
but
we'll
double
check
with
them,
and
in
addition
to
that,
I
I
know
scott
has
been
submitting.
A
Oh
it's
good,
I'm
gonna
change
the
name
later
I
I
saw
scott
has
submitted
a
pr
to
add
credentials
for
vsl
yeah.
Are
we
gonna
chat
with
the
design
and
see
if
scott
wanna,
you
know,
have
that
track
in
one
pen
plan
and
in
the
due
diligence
to
that?
There
is
an
important
design
work,
which
is
the
data
mover.
We're
gonna
continue
reviewing
the
pr
for
shubham
and
per
discussion
with
dylan
and
shibang
there.
A
We're
gonna
also
try
to
do
some
poc
to
make
sure
that
the
generic
data
mover
workflow
work
for
the
major
csi
drivers
in
our
case
from
vmware
we're
gonna
make
sure
that
works
for
vcr.
A
We
know
that
the
generic
code
path
may
not
be
very
efficient
or
for
a
vcr,
so
the
data
moving
workflow
may
be
slow,
but
we
no
one
make
sure
that
it
works,
so
so
that
so
with
that
we're
gonna
deliver
a
workflow
that
works
for
all
the
major
satellite
drivers
and
in
addition
to
that
billion
also
mentioned,
they
want
to
make
some
progress
in
concurrent
backup
and
restore
as
the
groundwork
for
the
multi-tenancy
support.
A
So
since
dylan
and
scott
or
siobhan,
it's
not
there,
we're
gonna
check
with
them
later
and
in
addition
to
that,
we're
gonna
continue.
Our
plan
to
you
know:
refactor
the
controllers
in
valero,
we
we
have
been
covering,
you
know,
more
than
half
of
the
controllers,
so
we're
gonna
handle
the
rest
of
them,
including
the
backup
scene,
controller,
pc
controller
and
ether.
A
But
I
want
to
point
out
that
we
we
will
treat
backup
and
restore
control
as
opportunistic,
because
in
the
refactor
of
cochlea
there
are
expects
to
be
a
lot
of
change
in
these
controllers.
We
don't
want
to
have
a
lot
of
code
conflict,
but
you
know
if
time
permits
and
we're
in
good
condition.
A
It's
possible
that
we
also
handle
the
refactor
for
these
two
controllers
and,
last
but
not
least,
we're
gonna
continue
enhancing
our
past,
as
stanford
already
mentioned,
and
she's
already
started
the
work
now,
we're
gonna
add
the
migration
test
to
android
and
test
suite
we're
gonna.
You
know
automatically
test
the
migration,
a
scenario
where
you
do
backup
in
one
cluster
and
do
restore
in
the
other.
That
will,
you
know,
save
us
a
lot
of
a
manual
testing
effort,
hopefully
in
the
future
releases
and
again
this
all
these
works.
A
I
mentioned
I
have
put
in
the
one-time
milestone,
which
is
a
single
source
of
truth
for
what
we're
gonna
deliver
and
there
is
still
110
candidates.
A
So
the
idea
is
that
if
we,
I
think
there
are
more
time
to
handle
additional
issues,
we're
gonna
move
these
issues,
some
of
them
into
the
110
milestone
before
the
feature-free
state.
A
Also,
we
we
also
allow
there
are
new
proposed
features
and
that
we're
going
to
put
in
the
110
before
this
feature
phrase,
but
after
that
we're
going
to
focus
on
what's
in
the
milestone.
A
Yeah,
I
want
to
nominate
you
as
a
maintainer
of
valero
young,
who
has
been
doing
a
lot
of
great
work
in
valero
and
he
really
educated
the
team
a
lot
about
this
with
his
rich
experience
in
the
you
know,
br
area
of
the
industry.
A
So
I
I
want
to
nominate
you
know
the
maintainer,
so
he
can,
you
know,
have
higher
privilege
to
you
know:
adding
labels
and
triaging
issues,
changing
assignments,
etc.
I'm
gonna
double
check
with
scott
and
other
scott
shubong,
dylan
and
other
maintainers
in
the
channel
so
yeah.
If
there's
no
disagreement,
I'll
already
have
a
pr
in
who
I
think
you
can
be
really
briefly
talking
about
the
changing
copier
integration
design,
because
we
need
to
think
with
other
folks.
You
know
via
the
channel
anyway.
A
Oh
yes,
so
let's
stop
stop
sharing.
C
C
Actually,
the
change
is
not
about,
after
on
the
higher
level,
actually
on
the
architecture
or
on
the
higher
workflow
level
are
all
on
the
interface
experience
actually
happened
on
the
on
the
operational
level.
We
tell
the
operation,
one
of
the
trainings
is,
like
you
know,
other
requirements.
Some
parts
just
want
to
restore
the
all
data
that
backed
up
by
russia,
even
under
the
copier
patch
or
under
the
new
path.
C
So
that
means
we
need
to
allow
users
to
specify
the
path.
At
the
I
mean,
the
backup
result
level
actually
only
on
the
result
level,
because
there
is
no
requirement
for
backup.
That
means
we
need
to
have
a
way
to
allow
users
to
specify
the
path
for
each
of
the
results.
C
C
This
this
one
is
decided
at
the
time
of
the
installation,
and
this
means
the
default.
Fixed
decision
indicates
that
the
default
pass
when
creating
a
new
backup
and
also,
we
will
add
the
the
same
value
uploaded
type
into
the
port.
One
dot
common
result.
Pr,
and
also
in
the
backup,
repository
sorry
backup,
resource
cr
and
also
in
the
backup,
repository
cr
and
nearly
the
the
name
and
the
backup
repository
is
repositive.
But
it
means
the
same
thing
and
how
do
we
do
that?
C
How
do
we
use
these
values
to
fulfill
the
dynamics
page,
let's
take
a
overall
backup
and
result
workflow,
for
example.
Firstly,
when
you
will
create
a
new
backup
and
the
backup
controller
will
track
the
uploaded
parameters
from
the
vlog
server,
because.
C
C
So
just
these
two
crs
are
created
with
the
value
of
jurassic
and
then
the
backhoe
repository
controller
sees
the
newly
created
backup
repositor,
and
it
will
get
that
the
current
pass
is
that
is
decided
by
the
repetitive
is
racist,
so
it
will
be
able
to
start
its
internal,
its
internal
legacy
value.
So
at
this
time
it
is
with
the
lux
values
force.
It
means
the
rustic.
C
Oh
sorry,
it
loves
it
likes
is
true,
it
means
rustic
right.
So
the
backup
repository
controller
will
go
into
the
russian
way.
Just
following
this
diagram
and
finally
created
the
russian
mylar
and
just
similarly
when
the
pawn
was
pulled.
One
back
up
for
receives
the
newly
created
pokemon
back
up.
Cr
is
well
got
to
the
same
value
of
the
oblique
type
from
the
cr
and
then
follow
and
then
may
decide
its
value
electric
value
to
two
and
follow
this.
The
same
workflow
here
to
create
the
backup
result.
C
And
when
you
want
to
restore
from
the
backup
and
feel,
very
importantly,
the
restore
doesn't
track
the
other
types
from
the
vlog
server.
That
is
the
static
value
it
doesn't
calculate.
It
will
track
from
the
backup
from
which
it
is
going
to
result.
C
It
means
actually,
there
is
no
no
uploaded
type
value
inside
the
backup
cr
directly,
but
it
can't
cut
the
product.
One
backups
here
from
the
backup
crm
then
got
its
value.
So
in
this
case
the
value
is
relative,
so
it
is
going
to
create
the
0.1
resource,
er
and
backup
repository
with
uploaded
type
value
of
jurassic,
and
then
everything
goes.
Similarly
with
battery
repository.
D
C
Problem
will
resolve
computer,
and
so
their
parts
will
be
decided
and
then
when
and
then
for
another
time,
user
may
want
to
change
the
path
by
resolve
the
arrow,
and
at
this
time
it
the
the
arrangement
type
specified,
develop,
server
is
copia
and
then,
when
you
create
a
new
backup,
the
w
type
is
copia
and
it
will
create
the
two
crs
with
the
value
of
copia
and
where
everything
goes.
C
Similarly
to
the
backup
report,
on
one
backup
controller
with
copia
and
and
now
this
time,
when
user,
okay,
oh
users
say
I
don't
want
to
restore
from
the
second
backup
I
want
to
resolve
from
the
first
backup
that
is
created
by
russia.
Oh,
it
is
also
okay,
because
the
same
thing,
the
result,
controller
gas,
the
the
value
from
the
backup
or
actually
from
the
port,
one
backup,
cr,
and
that
value
is
relative.
So
everything
goes
with
the
graphic
path
conclusively.
C
We
have
one
static
value
that
decided
by
the
by
the
installation
such
that
as
a
default
value
when
creating
new
backup.
Only
that
value
is
static
and
for
other
cases
everything
is
driven
dynamically
or
every
other
cases.
Path.
Selection
is
driven
by
the
by
the
the
path
values
inside
the
crs,
not
by
this
static
value.
So
in
that
way
we
will
be
able
to
implement
the
dynamics
range.
A
I
have
a
small
comment,
so
I
I
get
that
in
some
way
you're
gonna
try
to
match
the
uploader
type
to
the
repository
type
in
the
controller
right,
because.
C
You
know
to
have
the
controller
to
track
the
the
cr,
the
crs
type,
so
that
it
make
sure
that
it
it
is
processing
or
hardly
the
ground
cr,
but
the
first.
A
But
we
still,
I
think
we
still
need
to
check
sometimes
because
when
you
in
the
volume,
vr
controller,
for
example,
you
need
to
get
the
information
of
the
repository
so
that
you
can
trigger
copy
already
to,
for
example,
upload
the
data
to
the
repository.
C
Something
like
a
a
backup
or
restore
or
uploader
right,
and
here
we
call
it
a
repository
so
reports
here.
It's
very,
I
think,
a
little
bit
weird
that
we
have
something
called
uploader
type
because,
as
you
can
see
here-
and
we
have
small
usage
for
this
repository
in
future-
and
currently
we
have
the
only
reactive
agriculture
in
future.
You
know
the
unified
repository
may
support
multiple
backup
repository
storages,
so
there
will
be
multiple
values
other
than
russia
or
copia.
C
A
C
Yeah
that
will
happen
in
future.
So
currently,
since
there
are,
you
know
only
two
values
for
both
the
uploader
and
the
repository.
So
so
it's
it's
bite-wide
right.
So
when
we
choose
a
copy
uploader,
it
implies
that
it
will
choose
copia
copy
of
repository,
because
it
cannot
do
the
radical
repository
right
so,
but
it
doesn't
mean
that
the
the
the
copy
uploader
must
bound
to
the
same
copy
repository.
A
I'm
not
sure
if
that's
correct,
I
think
we
can
discuss
offline.
I
get
a
highly
high
level
idea
because
in
the
design
you
you
know,
you
have
two
components
as
the
repository
part
and
uploader
part.
So
in
that
sense
it
may
make
sense.
You
have
two
type
but
you're
thinking
about
matching,
I'm
not
sure.
If
we
go
this
way,
we're
gonna
add
more
attribute
into
the
volume
vr
crs
so
that
we
can
find
the
right
type
of
repository,
but
yeah
we
can
discuss
offline.
I
think
later
yeah.
C
Yeah
offline-
actually
here
I
just
add
one
more
slider
actually
for
the
postmodern
backup
controller
it
it
doesn't
need
to
charge
the
the
the
repository
or
the
backup
repositories,
repository
type
values.
You
just
need
to
check
the
uploader
type
value
from.
Is
it's
a
polygon,
backup
cr?
C
So
maybe
things
if
not
that
that
complex,
but,
let's
discuss
later
that
is,
you
know
not
deciding
or
not
very
important
for
the
or
changing
thing
for
the
currency
workflow
yeah,
that
is,
that
is
the
pathwage
and
dynamics
page,
and
one
thing
we
want
to
add
is
like
I
I
want
others
like
this
is
based
on
one
requirement
from
you
know
some
other
part,
it's
not
something
that
we
want
to
you
know,
so
we
have
to
do
the
dynamics
right
and
for
the
valencia
name
is
the
same
thing.
C
We
didn't
change
the
principle
that
we
want
to
change
the
name
of
the
retro
project
to
a
more
generic
one
like
backup
repository
so
and
the
rules
is
not
changed
either
like
way
the
new
controllers
or
the
controller.
You
need
the
newer.
C
The
new
name,
only
the
crs
with
the
old
name,
will
be
ignored.
So
here
I
just
add
some
details
about
why
we
have
problems
from
either
the
repository
side
and
from
the
backup
or
resolver
flow
side.
We
can
check
these
taxes
offline
if
there
is.
C
Leave
comments
and
the
strong
story
configuration
is
something
I
knew,
but
actually
it's
not
a
new
part
in
it.
Yeah
just
you
know
we
make
something
more
clear,
you
know,
so
I
thought
I
I
I
mean
it
is
necessary,
so
we
just
make
it
clear.
It's
like
you
know
for
for
different.
You
know,
backup
storage.
We
need
very
different
parameter
to
connect
to
it.
For
example,
the
parameter
to
connect
to
s3
compatible
story
will
be
quite
different
from
the
the
ones
to
connect
to
another.
C
Sorry
right
so
actually
at
present.
Rashford
is
also
have
this
kind
of
parameters
in
the
in
the
backup,
repository
called
modern
backup
and
the
product
resolutions.
But
the
the
parameters
are
formatted
into
a
single
screen
or
we
call
it
ripple
identifier,
but
that
spring
is.
It
has
been
good
enough
to
call
russia
command
line,
but
it
is
very
limited
and
it
is
not
very
easy
to
extend
so
we
we
want
to
change
this.
You
know
stream
formatted
value
into
an
infrastructure,
and
everything
is
very
clear
to
understand.
C
Actually
no
significant
thing
in
the
meanings
like,
for
example,
we
have.
We
have
a
provider
to
indicate
the
provider
of
the
storage
and
we
have
some
parameters
for
connecting
to
this
cell
storage.
C
So,
even
though
that,
even
though
for
wave
1.10
we
will
have,
we
will
now
support.
We
have
no,
you
know
immediate,
like
requirement
to
support
an
fs,
but
we
think
this
is
enough
to
have
this
structure
to
support
nfs
in
the
future.
A
C
If
it
is
okay
for
the
later
animation,
because
at
presence,
it
is
a
line
this
way
that
is,
we.
C
This
report,
identifier,
if
it's
you
know
it
gives
reporting
by
106
yards.
So
I'm
not
sure
right
now,
but
if
we
want
we
can
do
more
investigation
on.
A
A
C
So
yeah
actually
originally
this
this
information.
Oh
this
information
is
not
from
it's
not
originated
in
from
this
racial,
it's
oriented
from
the
bsl
right
because
sorry,
location
and
then.
C
This
screen
and
then
put
into
this
vcrs
so
yeah.
So
of
course
we
have.
Actually.
We
have
three
versions
of
the
redundant
new
information
from
the
bsl
and
from
yeah.
We
can.
D
C
You
know
effects
also,
but
we're
gonna
not
cover
this
in
this
vr.
I'm
sorry
in
this
pr
and
if
we
want
to,
we
can
do
along
with
the
change.
Okay.
C
Installation
there's
no
special
changes
and
upgrade
upgrade.
Actually
it's
it
is
also
the
things
mentioned
here
is
also
necessary.
We
we
just
want
to
put
it
out
into
this
pr,
so
that
you
know
we
don't
ignore
them
during
our
development.
C
Upload
the
type
value
of
the
relative
server
during
installation.
That
is
also
true
for
upgrade
right.
So
if
you
upgrade
the
same
as
installation,
we
need
to
change
the
window.
Servers
uploaded
type
value
by
smashing
specifying
this
flag,
but
currently
I
don't
think
the
for
either
the
cri
installation
and
the
hard
time
installation.
We
have
the
way
to
change
the
windows
servers
command
line
because
I
think
currently,
if
the
the
server
is,
there
will
do
nothing.
C
So
we
will
need
to
apply
the
change,
the
toolbar
server
for
both
the
cri
and
the
hard
part
installation,
and
also
we
we
want
to
add
some
latency
to
the
relational
part,
so
that
there
is
an
easy
way
to
query
the
current
path.
A
Yeah,
but
I
just
I
just
want
to
point
out
for
the
in-place
upgrade
where
you
don't
need
to
uninstall
the
lateral.
I
think
we
don't
need
to
introduce
new
submac
sub
command
to
cli
to
support
that,
but
we
need
to
write
documentation
because
currently,
if
you
check
the
official
website,
it
suggests
users
use
control
to
upgrade
to
bump
up
the
image
version.
A
So
I
think
we
need
to
provide
the
same
user
experience,
but
just
a
little
bit
more
complicated,
because
user
are
expected
to
execute
more
command
or
add
more
parameter
to
control
so
that
he
can
finish
finish.
This
upgrade
okay,.
C
Okay,
okay,
so
it's
like
when
you
we
will
not
use
the
cri
to
for
the
in
place
upgrade,
so
we
will
use
a
good
control
and
yeah.
A
C
Yeah,
I
I
recall
that,
and
I
will
modify
it
here.
A
C
And
another
thing
is,
we
all
also
change
the
name,
we're
asking
the
demonstration
to
the
learning
node
and
we
have
a
comment.
This
is
disgusting
and
we
also
talked
with
the
other
teams
and
the
fifth
one
is
okay,
but
we
the
same
thing.
We
need
to
apply
this
name
change
after
the
upgrade
yeah.
A
A
C
Yeah
and
here's
the
cli
actually,
this
is
not
something
new,
but
we
just
want
to.
I
mentioned
it
obviously
in
this
vr
that
we
have
some
changes
in
the
below
cri,
for
example,
without
backup.
This
describes
how
I
need
to
at
least
indicate
the
current
pass
for
the
backup
and
the
same
thing
for
the
stone
and
another
very
important
one.
Is
the
lara
russet
the
ripple
guide?
I
think
here
the
atmosphere
needs
to
output
all
the
repositories
if
both
the
radical
reporter
and
the
unified
representative
exists
at
the
same
time,
and
also.
A
I
think
I'm
not
sure
the
restore
I
mean
the
backup
currently
in
the.
I
think
we
only
need
to
handle
that
in
the
dash
dash
with
the
when
the
detail
flag
is
specified
right,
yeah,
because.
C
A
Because,
currently,
without
the
detail,
you
you
don't
tell
if
it's
recipe
based.
C
A
C
Yeah,
let's
leave
comments
on
this
another
name.
You
see
yes
and
finally,
we
want
to
mention
that
when
I
back
up
great,
we
will
not
have
to
leave
any
option
to
specify
the
past
because
we
didn't
have
any
requirements,
but
it
doesn't
mean
that
we
don't
support
it
from
the
underlying
controller.
Actually,
with
the
dynamics
range,
the
online
controls,
the
modules
are
very
flexible,
so
it
is
flexible
enough
to
support
any
anytime
because
which
is
called
any.
C
You
know
up
level
commander
for
dynamics,
ratio
dynamically
specified
path,
but
we
don't
want
to
make
all
the
things
happen
for
all
the
commands
that
go.
You
know
to.
C
That
release
many
of
the
you
know,
sorry
that
it's
like
to
make
pieces
simpler
right.
C
That's
it!
I
think,
that's
all
for
the
changes
and
we
and
currently
we
are
still
something
still
under
the
discussion,
so
we
may
make
some
money
in
future,
but
this
is
what
they
are
like
now
and
we
can,
you
know,
leave
any
comment
theories.
You
know
if
there's
any
questions
or
suggestions.
C
A
Okay,
so
I
see
schumann
you're
joining
there
are
a
few
items
related
to
a
red
hat.
I
wanna
follow
up
since
steven
or
scott
is
not
here.
F
Yeah
sure,
what's
up.
G
F
A
So
I
I
have
drafted
this
110
road
map
and
I
I
explained
how
we're
gonna
execute
so
so
you
can
watch
the
video
shubham,
but
there
are
a
few
items,
the
new
versions
of
plugins.
I
think
phone
and
scott
have
been
working
on
that
for
a
while,
and
please
follow
up
with
scott,
like
what
new
version
plugin,
he
wanna
you
know
implement
and
if,
if
that's
needed,
please
ask
him
to
write
up
a
an
issue
and
put
in
the
110
a
milestone.
A
We
have
two
levels
like
we
put
issue
in
the
one-time
milestone.
It
means
we
need
to
handle
them
before
the
fc
of
one
tank.
If
there
are
issues
left
in
this
milestone
very
likely,
we
will
not
have
c
and
there
is
the
110
candidate,
a
label
we
added
to
the
issue.
Okay,
which
means
that's
a
lower
priority,
but
we
think
it's
opportune.
They
are
opportunistic.
A
The
idea
is
that,
before
the
feature
free
state,
we
are
making
adjustments
to
the
plan.
But
after
that
we
will
remove
all
those
110
candidate
labels,
so
either
they
will
be
part
of
milestone
or
they
will
be
pushed
to
other
releases.
Okay,
that
makes
sense
yeah,
so
the
one
thing
I
want
to
follow
up.
I
wish
you
helped
to
follow
up
with
the
plugins.
The
other
one
is.
Scott
has
opened
the
pr
to
add
credentials
to
the
bsl.
Let's.
D
D
A
The
design
work,
I'm
sorry
that
we've
been
focusing
on
the
one-nine
delivery
and
your
pr
has
been
there
for
a
while
and
we're
going
to
continue
discussing
on
that
and
that's
a
very
high
level.
I
mean
high
priority
work,
we
plan
for
one
time.
We
want
to
have
a
concrete
design
for
the
data
mover
in
one
hand,
so
we
can
execute
in
111
and
we
I
also
discussed
with
dylan
he
mentioned.
There
may
be
some
groundwork.
We
can
merge
into
main
into
110
and
we'll
see
yeah.
G
So,
regarding
the
data
mobile
design,
we
are
working
on
a
new
plug-in
type
design
which
will
enable
async
processes.
So
there
will
be
a
pr
coming
up
so
that
we'll
get
that
in
place
first
and
then
we
can
use
async
processes
to
regarding
stuff.
G
Totally
agreed
alien,
daniel
I'll,
just
post
up
here
in
this
week,
so
you
guys
can
take
a
look
at
the
design.
It
will
be
great.
A
Yeah,
thank
you
and,
as
for
there
are
a
deal,
I
had
a
few
issues
saying
that
red
hat
is
more
interested
in.
You
know
doing
that,
I'm
not
sure
if
he
has
discussed
with
your
staff,
he
want
to
work
on
multi-tenancy
and
concurrent
backup
and
restore,
and
that
has
offline
discussion
with
bill
and
dylan,
and
we
decided
that
we
can
only
focus
on
one
of
them
and
we
chose
the
concurrent
backup
and
resource.
A
But
I
marked
that
as
p1
because
I
don't
think
that's
necessarily
a
blocker
for
rentham,
but
if
you
have
resources
from
red
hat
to
initiate
some
discussion
we
are,
we
are
happy
to.
You
know
continue
with
that,
because
that's
the
groundwork
for
the
multi-tenancy
and
yeah.
G
Okay,
so,
basically,
to
sum
up,
you
want
inputs
regarding
110
milestones
for
vsl
credentials,
plug-in
versioning
and
multi-tenancy
support
right.
A
Yeah
yeah
yeah.
Let
me
explain
a
little
bit
because
I
think
that's
happening
earlier.
A
discussion
between
eleanor
and
dylan.
G
A
So
I
think
dealer
added
a
few
added
this
label
to
a
few
issues,
so
the
csi
ga
and
data
movement
they're,
essentially
the
same
thing
and
the
plugin
versioning,
the
one
I've
been
talking
about.
I
think
what
we
have
some
disagreements
between
I
and
dylan
are
these
two.
I
don't.
C
D
A
Propose
we
can
only
you
know,
start
working
on,
one
of
them
so
didn't
think
they
are
intertwined,
but
later
we
decided.
Maybe
we
should
focus
on
this
one
and
you
can
double
check
with
bill
if
he
doesn't
care,
we
I'm
okay.
We
move
this
out
to
110.,
but
if
he
thinks
it's
important,
I
I
think
you
guys
can
start
yeah
initiate
some
effort
to
investigation.
I.
G
Get
that
so
yeah
we'll
let
you
know
if
we'll
answer
resources
to
work
on
multi-tenancy
yeah.
A
G
A
So
yeah,
I
think
we're
gonna
have
a
longer
discussion,
but
in
order
to
you
know,
contain
the
guys
from
west
coast.
The
time
is
difficult
for
us,
but
we
are
okay
to
attend
apm
meetings.
So
if
you
are
okay
to
you
know
attend
the
meeting
in
the
early
morning.
Maybe
we
can
adjust
the
meeting,
move
it
a
little
bit
to
a
few
hours
earlier,
so
that
we
can
have
a
better
chance.
G
Yeah
sure
thing
I'll
just
ask
around:
what's
the
input
and
I'll,
let
you
know
about
the
meeting
timing.
A
Okay,
but
but
this
time
it's
usually
it's
difficult
for
all
of
you
right,
because
previously
I
mean
dylan
and
scott
often
show
up
in
the
same
time,
yeah.
D
G
A
G
A
Okay
and
what's
the
other
guy
yoon
chan,
please
also
check
with
fong
like
what
he
thinks
and
we're
gonna
discuss
in
this
slack
about
you
know
if
we
can
make
it
in
a
new
time
for
the
community
meeting
yeah.
H
Totally,
that
would
be
great
yeah.
Let's
do
that
because
it's
only
western
coast,
so
I
think
at
least.
A
F
Yes,
exactly
red
hat
is
from
eastern
coast.
So
yes.
A
Yes,
yes,
yes,
but
but
for
this
one,
because
we're
gonna
go
through
the
one
hand
roadmap.
So
I
was
thinking
you
guys
are
coming,
but
what
shows
up
is
a
little
yeah
strange.
A
Okay,
yeah
and
I'll
be
more
actively.
You
know,
you
know
talking
about
the
community
meeting
time
to
make
sure
you
guys
can
join
okay.
C
I
have
a
difficulty
exactly
first
bound
and
when
we
say
the
the
asynchronized
plugin,
it
is
just
for
the
crocodile
backup
and
result
no.
A
A
G
So,
basically,
to
give
you
an
idea.
Currently,
if
you
trigger
any
plug-in
process
in
backup,
is
your
workflow
right?
You
have
to
block
on
the
plugins,
so
yeah.
If
you
see
the
csi
plugin,
we
block
on
it
right
so
yeah.
So
in
order.
G
That
dependency,
we
are
coming
up
with
a
new
plug-in
type,
which
will
help
kick
off
the
async
plug-in
processes
and
let
the
backup
restore
workflow
continue
not
block
on
it.
Something
like
that.
A
Okay
but
there,
but
there
are
different
way
of
doing
that.
I
see
what
you
want,
but
I
used
to
think
the
flow
is
like
we
trigger
another
cr
to
handle
the
data
movement
and
we
let
the
backup
mark
itself
as
completed
yeah.
G
But
what
we
are
trying
to
propose
is
there
will
be
a
method
with
the
plugin
type
which
will
enable
you
to
see
the
pro.
What
do
we
say
status
of
the
process
which
was
triggered
so
based
on
that
you
can
mark
the
backup
complete
not
without
checking
on
the
process
which.
A
G
A
Something
that
was
was
merged
before
I
joined
this
project.
Okay,
upload
progress.
This
one
was
designed
by
dave.
I
think
it
served
the
same
purpose
and
he
already
has
a
lot
of
pr
a
big
pr
here.
I
think
we
have
some
back
and
forth,
but
it's
relatively
done
so
that
it
can
handle
this
upload
progress,
but
but
I
I
I
double-
checked
dave
he
he
mentioned.
He
may
not
have
enough
time
to
continue
working
on
this
one.
Okay,.
G
A
G
Monitoring
discuss
this
about
this
once
once
I
get
the
design
pr
like
when
you
guys.