►
From YouTube: Velero Community Meeting - July 11, 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
As
with
that,
just
please
follow
the
kind
of
comments
which
boils
down
to
be
nice
to
each
other
without
I'm
gonna
paste
into
the
chat
the
community
minute
minute
notes,
so
it
will
be
super
cool
if
you
feel
like
adding
yourself
there
and
with
this.
Let
me
open
them
and
I
can
see
many
folks
added
themselves,
which
is
great,
and
it's
got
added
some
some
peer
reviews
in
shubham
in
our
show
so
I
think
it
doesn't
make
sense.
A
B
Yeah,
nothing
specific
to
discuss
just
mentioning
that
kind
of
that's
kind
of
my
presentation
this
week
is
I,
did
review
several
PR
as
I
didn't
have
any
other.
You
know
specific
development
progress.
Do
you
expect
to
get
back
to
my
end-to-end
testing
for
the
async
plugin
soon,
but
I've
been
working
on
uhp
issues
this
week,
Beyond
the
pr
reviews
and
I
do
have
a
discussion
topic
later
at
the
end,
but
as
well
but
foreign.
A
C
A
C
A
Mm-Hmm
yeah,
but
do
you
mean,
do
you
need
any
input
from
from
anyone.
D
Like
okay,.
A
Okay,
cool
and
then
control.
E
F
It's
not
a
quick
point
there
if
you
just
scroll
down
a
little
bit
all
in
the
status.
The
right
in
this
document,
hack,
MD
document
I,
saw
that
the
Chinese
team
updated
clear
instructions
on
how
to
test
this
data
mover
support,
actually
I
see
that
you're
not
sharing
the
screen
already.
F
A
F
So
if
you
scroll
down
to
the
previous
status
meeting,
the
chain
is
called
detailed
instructions
on
how
to
test
yeah
go
down
scroll
down
a
little
bit
yeah.
This
is
instructions
to
try
the
CSS,
not
sure
right
at
the
answer.
This
is
what
you're
asking
right.
F
A
Okay,
so
I
should
do
you
want
me
to
stop
sharing,
and
so
you
can
walk
questions.
E
I
think
I
I
think
I
can
just
speak.
I
don't
really
need
to
share.
So
this
adding
support
for
multiple
volumes,
natural
class
in
the
CSA
plugin.
This
had
I
have
raised
in
the
community
meetings
around
three
four
weeks
and
finally,
now
the
design
PR
is
merged
and
we
need
one
more
approval
reviewer
for
the
actual
implementation
PR.
So
if
anyone
has
some
time,
maintainers
please
go
and
check
it
out.
E
Second,
is
the
Json
substitutions
PR
that
I
had
just
basically
given
a
walk
through
on
the
design
change
in
the
last
community
meeting,
so
just
a
heads
up
that
the
implementation
PR
is
up
now
and
also
the
design
PR.
The
changes
that
whatever
had
come
in
I
have
pushed
those
as
well.
Few
folks
are
already
reviewing.
Is
what
I
see
from
the
comments
on
the
PR,
but
just
pitching
it
for
the
broader
Community
to
take
a
stab
and
have
start
reviewing
this
PR,
because
these
are
good
features.
E
Any
insights
you
can
give
any
better
suggestions
on
refactoring.
The
code
would
be
really
helpful,
so
yeah
again
a
request
for
review
on
these
PRS.
The
design
PR
is
also
not
yet
merged
for
Json
substitution,
so
yeah,
please
do
check
that
out
as
well.
It
should
mostly
be
clear
enough
for
you,
folks,
yeah.
That's
it
nothing
to
present
right
now,
almost.
A
Thank
you.
Anyone
you
want
to
comment
on
this
one,
or
do
you
need
some
anything
specific
more
from
usual
or
we
can
move
on
to
the
next
one?
F
Okay,
yeah
before
I
just
go
ahead
with
mine,
I
think
I'm
sure
PR
for
a
short
substitution
right.
We've
seen
at
least
two
users
asked
for
features
in
last
week
alone.
So
I
guess
people
are
really
waiting
for
this.
So
it
should
be
one
of
the
good
features
in
1.12.
G
F
Going
back
to
the
odd
label-
selectors
right
so
I
just
listed
here,
because
we
started
working
on
this
and
in
fact
there
there's
a
PR
from
cloudcaster
from
neeresh
about
this
I've.
Seen
one
user
asked
for
this
few
weeks
back
and
one
another
one
came
back
last
week.
Essentially,
people
want
support
for
our
label
selectors
from
the
Valero
CLI.
So
the
back
of
CR
supports
this.
The
restore
CR
supports
it,
but
the
command
doesn't
so
we're
working
on
it
now
and.
A
I
think
we
can
jump
straight
to
the
discussion.
That's
more
that
thing
over
here.
It's
more
like
a
reminder
for
me
to
fix
the
reminders
after
the
call
yeah
so
yeah
you
can
you
can
go
ahead.
I
can
I'll
stop
sharing,
so
you
can.
You
can.
F
F
So
the
first
thing
is
about
Wonder
eleven
dot,
One
release
date,
I've
seen
at
least
two
or
three
people
asking
for
it
and
I
think
we
are
waiting
for
some
golang
fix
right.
So
is
there
any
date?
I
haven't
really
checked
the
goal.
Language
issue
itself,
I
guess
we're
waiting
for
some
cve
to
be
fixed
in
1.20
or
six
or
something.
But
at
least
two
three
people
ask
her.
H
F
The
release
date,
because
there
are
a
couple
of
important
fixes,
one,
the
backup
status
that
we
fixed
and
the
other
one
that
I
listed
here
is
this
user,
is
mysteriously
seeing
errors
in
the
S3
bucket
region.
Api
I
think
Lyndon
fixed
added
a
log
message,
because
the
error
was
not
being
logged
in.
He
added
a
log
fix
and
these
guys
wanted
to
try
it
in
their
environment
so
because
the
1.11.1
is
not
there
so
I.
What
I
did
is
built
an
image.
F
You
know
ourselves
and
I
I
gave
this
user
the
image,
so
they
they
said
they're
using
it
yeah
and
they'll
get
back.
If
they
you.
F
The
main
problem
here
is
the
earlier
won't
have
to
live
under,
one
is
released,
the
better
I,
guess
and
I,
don't
know
Scott
or
anybody
else
who
is
attending.
The
other
calls
maybe
is.
I
B
Yeah
I,
don't
have
any
specific
Insight
I
also
know
there's
another
bug
in
one
eleven,
not
one
that
we've
seen
at
least
two
people
on
the
red
hat
side
interested
in
us
getting
there's
a
bug
fix
on
The,
Rustic
copia
side,
where
we
fail
a
backup
if
a
volume,
if,
if
a
pod
with
a
battered
volume,
is
not
running
and
instead
of
skipping
the
volume
we
fail
and
as
a
fix
for
that
as
well
and
111
that
one
but
no
I,
don't
actually
know
the
specific.
B
You
know
what
what
the
current
exact
release
date
is.
I,
don't
know
if
Orlin
knows
or
not.
A
Yeah
I
mean
I,
haven't
checked
with
with
deep
on
this
one.
To
be
honest,
so
I
don't
know.
F
Right
I
mean
the
comment
I
have
seen
in
one
of
the
issues.
I
think
it
is
6454
is
we
are
holding
one
because
of
1.20.6
golang
to
fix
some
cves
right,
so
I
mean
probably
I'll
go
ahead
and
check
that
further
I
think
it's
it's
a
little
bit
important.
It
is
important
for
us
to
release
one
level.
One
I
think
because
of
a
couple
of
two
three
important
fixes:
okay,
so
the
other
one
yeah.
F
Can
you
scroll
down
a
little
bit
Yeah
the
CSI
plugin
changes
right
so
Scott,
thanks
for
merging
that
CSA
timeout
PR,
that
is
in
the
main
branch
of
the
CSI
plugin
repo
right,
but
that
is
not
imported.
So
I
don't
know
the
procedure
for
releasing
the
plugins
in
between
I
mean
the
procedure
for
the
main
release
is
kind
of
clear
I
see
labels.
You
know
released
and
1.11
something
like
that.
Yeah.
B
I
I
think
yeah
we're
not
as
consistent
with
the
with
the
plugin
release,
because
sometimes
there's
not
any
changes.
You
know
when
when,
for
example,
the
111
that
one
comes
out
and
if
there's
a
plug-in
that
doesn't
have
any
changes,
you
know
we
wouldn't
release
it,
but
I
think
if
there
are
fixes
we
want
in
there,
maybe
create
a
GitHub
issue
to
say:
can
we
import
export
this
to
the
release
that
corresponds
to
1.11?
B
You
know
so
that
we
can
then,
because
that,
because
I
imagine,
when
we
release
1.11.1,
we
should
be
able
to
release
a
plug-in
that
has
a
fix.
You
know
at
the
same
time
or
similar,
so
I
would
put
a
GitHub
issue
to
backboard
it
and
then
ping
on
slack
to
open
the
discussion
just
to
find
out,
because
I'm
actually
not
sure
right
now
again
whether
we
plan
to
release
all
of
the
plugins
again
with
1.11
that
one
kind
of
an
updated
version
for
those
as
well.
B
You
know
if
they're
go
Lang
related
issues.
Maybe
we
want
to
anyway
for
all
those.
For
that
reason,
I
don't
know
that
CV
effects
the
plugins
or
not,
but
my
physical
thing
level
issue.
So
if
we're
planning
a
release
for
those,
we
definitely
want
to
get
that
backboarded.
So
I
I
would
create
a
GitHub
issue
to
backboard
it
to
111
sorry
to
to
the
release
corresponding
to
one
at
eleven.
You
know.
B
But
you
know
whatever
that
release
is
we
could
do
a
you
know
patch
release
to
that
as
well.
Probably.
F
F
That's
they
call
the
retention
policy
for
the
buckets
right,
but
in
general
that's
a
that's
a
general
issue
for
all
the
crowd
providers
where
object
locks
are
set
on
the
bucket,
so
you
won't
be
able
to
modify
any
objects
and-
and
it
seems
the
Valero
starting
from
1.11-
has
to
modify
so
I.
Think
I
forgot
the
name
but
he's
trying
to
add
this
to
the
documentation
that
villara
will
not
work
with
1.11.
F
F
We
should
probably
discuss
that
how
to
support
you
know:
Valero
backups,
to
buckets
which
have
these
features.
The
object,
lock
is
the
term
used
by
AWS
in
that,
but
that's
there
for
others
as
well.
Gke
and
Azure
both
have
a
similar
functionality
and
definitely
more
people
will
use
that.
F
So
we
need
to
figure
out
how
villaro
can
support
that
I
mean
you
can't
modify
the
objects,
so
you
have
to
do
something
else
in
that
case,
so
it
may
not
be
easy
change,
but
I
think
it
will
be
an
important
thing
and
maybe
something
that
we
can
discuss
in
the
context
of
1.13
right
I've,
seen
recent
message
from
somebody
asking
for
any
suggestions
for
under
13
roadmap.
So
that's
why
I
thought
I'll.
E
F
Don't
really
know
that
that's
easy
to
I
mean
we
can
easily
update
the
document.
That's
happening
right
and
I'm,
reviewing
the
documentation,
PR,
but
I,
don't
think
it
will
be
a
long
term.
It's
not
a
sustainable
position,
because
if
you
are
a
small
shop,
maybe
you
won't
use
this,
but
anytime
you
get
to
I
mean
we
have
this
experience
with
cloudcaster
as
well,
and
we
support
object
logs.
F
So
maybe
we
can
help
a
little
bit
there
from
the
design
perspective,
but
once
you
go
to
decent
sized
customer
I'm
almost
positive
that
they
are
going
to
have
some
policies
on
the
bucket
I
mean
they
want
to
have
control,
especially
from
legal
perspective.
They
want
to
prove
that
the
data
is
cannot
be
removed
before
a
given
recession
date
right.
There
are
legal
requirements
in
some
cases,
so
they
will.
They
will
definitely
have
that.
I
know
it's
challenging
from
a
design
perspective.
You
can't
modify
in
place,
but
there
are
workarounds.
F
G
A
All
right
anyone
else
this
one
or
we
can
move
to
to
the
next
one
actually.
A
F
One
step
ahead
in
this
discussion
topics
part
is
asking:
has
anyone
tested
volume,
mod
mode
block,
support
in
this?
Interestingly,
we
tested
it,
maybe
one
or
two
weeks
back
and
yes,
we
file
system
backups
are
not
supported.
If
it
is
block
block
type
PVC,
it
won't
work
right.
F
B
There's
even
a
question
you
know:
does
it
even
make
sense
for
a
block
volume
to
use
a
file
system
method
on
it?
I
don't
know
if
there's
a
way
to
even
kind
of
hack
it
to
make
it
work.
Part
of
the
reason
I'm
wondering
is
specific
on
The
Daily
member
side,
because
you
know
that's
using
copia
behind
the
scenes,
because
you
know
the
issue
is.
B
If
you
have
block
files,
you
know
if
you're
using
snapshots,
for
example,
you
know
you're
fine,
but
the
problem
with
snapshots
is
you
know
it's
not
affordable
to
other
providers
and
also,
ideally,
something
like
the
data
mover
solution.
We'd
find
a
way
to
work
with
block
mode,
but
I
just
don't
know
if
that's
feasible
with
the
current
design.
F
Yeah,
no,
we,
we
have
actually
some
information
in
this
regard,
because
we
added
that
support
to
Cloud
cloudcaster
supports
block
PVS.
You
can
take
a
snapshot
of
that
and
then
you
can
copy
off
it,
but
we
are
using
copia
on
that,
so
it
for
copy.
It's
just
like
a
file,
but
obviously
it's
not
optimal,
because
it's
going
to
read
through
entire
block
PV
and
even
though
it
may
not
back
up
everything,
because
a
lot
of
blocks
may
be
zero
on
this
device,
but
still
the
read.
F
B
F
B
So
it
sounds
like
we
might
be
able
to
make
this
work
within
Valero
copia
and
that
immover,
but
it
would
just
require
some
changes.
F
I
think
so
yeah
it
should
work
because
the
latest
design,
the
wonderful
design,
is
very
similar
to
what
we
are
doing
in
cloudcaster
already,
so
it
should
definitely
work
but
I
I'm,
really
hoping
for
a
CBT
kind
of
solution.
F
You
know
that
kubernetes
work
group
is
discussing
Forever
at
this
point,
I
mean
maybe
more
than
a
year,
probably
more
than
that
I
understand
it's
not
an
easy
thing
to
do,
because
they
are
worried
that
you
know
the
CBT
data
can
be
big
and
how
best
to
expose
that
information
to
all
the
backup
providers
I
haven't
attended
their
calls
in
the
last
few
times
so
I
don't
know
the
latest
status.
I,
don't
know.
If
there's
anybody
here
who
is
attending
the
work,
you
know
what
data
production
work
group
calls.
Maybe
they
have
more.
F
B
Right,
yeah,
yeah
and
then
there's
actually
an
old
GitHub
issue
where
someone
tried
this
with
rustic
before
we
had
copia
and-
and
that
was
one
of
the
things
mentioned
so
so
I
imagine,
with
data
mover
and
copia
to
support
this
we'd
have
to
you
know,
check
that
if
it's
blocked,
then
we'd
have
to
look
somewhere
else
for
that.
So
that
way,
what
are
the
changes
needed
to
support
that
use
case?.
B
The
thing
is
the
snapshot
data
mover
the
way
because
I
mean
the
CSI
plug-in
by
itself
just
creates
a
snapshot,
but
the
way
the
data
mover
for
one
that'll
12
is
working
is
that
we
take
that
snapshot.
We
create
a
a
temporary
PVC
load.
The
snap
you
know
with
that
snapshot
is
the
the
volume
Source.
B
C
B
Because
of
that,
the
data
mover
copies
all
that
snapshot
data
using
copia
into
the
the
bucket
of
the
object
store
provider.
F
B
F
B
Right
so
so,
data
mover
works
that
way,
basically
there's
a
big
difference
between
data
mover
and
just
just
using
the
copy
of
us
and
backup
directly
is
the
copia
method,
with
Valero
copies
off
the
live
volume
from
the
the
Pod
is
mounting,
and
whereas,
in
the
data
mover
case
we
take,
we
use
the
snapshot
to
mount
the
volume
and
then
copy
it
from
there.
So
we're
not
touching
the
live
volume.
H
F
B
That
would
be
great
if
you
want,
if
you
want
to
look
at
that
too,
you
know,
like
I,
said:
that's
something
that
somebody
we
need
to
look
at
anyway,
but
it
and
there's
really
two
places
where
again
I.
Imagine
the
the
the
changes
are
pretty
similar
is
there's
the
copia
case
for
file
system
backup
and
then
there's
the
data
mirror
case
with
that
temporary
volume
and
basically,
like
you,
said
one
of
the
changes
is
we
have
to
look
at
where
the
volume
is
it's
going
to
be
mounted.
B
You
know
the
block
is
going
to
be
a
different
location
versus
the
file
system
and
then
there's
whatever
other
changes.
We
needed
to
use
Kofi
on
that.
F
B
I'd
say
at
this
point:
since
the
the
intention
is
to
deprecate
Rustic,
you
know
relatively
soon.
I
would
say,
don't
worry
about
rustic
support
for
this.
You
know.
B
I
B
B
Yeah,
that's
good
to
know
because
and
again
from
a
Valero
point
of
view,
rustic
matters
for
two
reasons,
one
because
that's
also
built
into
the
data
mover
solution
for
1.12
and
two
from
a
file
system.
Backup,
non-data
mover
point
of
view.
Rustic
is
the
way
forward.
I'm
sorry
copy
is
the
way
forward
anyway,
we're
trying
to
not
do
additional
work
on
rustic,
because
the
intent
is
once
copia
is.
You
know,
well
tested
and
well
used
to
deprecate
Rustic
and
then
eventually
drop
support
for
it.
F
I
think
Satya
was
talking
about
object
lock,
which
is
the
previous
Topic
in
for
copia
I
mean
slightly
different
yeah
topics
right
object.
Log
we
implemented
using
using
copia,
so
I
mean
blocks,
is
a
slightly
different
topic.
Yeah.
I
Yeah
we're
saying
the
same
thing,
though,
for
both
of
these
I
think
we
should
just
Target
copia
and
forget
about
rest.
A
By
the
way,
there
was
a
command
from
Sean.
If,
if
someone
can
send
that
use
case
over
the
source
code,
but
I'm
not
sure
do
which
exactly
source
code
or
use
case
is
referring
to
Sean,
can
you
can
you
specify.
B
Yeah
yeah
yeah
yeah,
yeah,
yeah
you're,
seeing
how
you
did
it
or
you
know
if
you
want
to
submit
a
PR
to
sh,
to
add
that
to
Valero,
you
know
directly
that'd
be
fine
too
either
way,
I
mean
it's
it's
something
we're
definitely
interested
in
on
the
red
hat
side
for
the
same
use
case
the
Cooper
case.
So
you
know
whatever
assistance
you
guys
could
help.
You
know
whether
it's
showing
us
how
you're
currently
doing
it,
and
then
we
can
work
it
into
Valero
we're
providing
a
PR.
B
You
know
either
way
you
know
just
kind
of
kind
of
helping
along
because
again
that
way
we
can
get
natives
to
work
for
this
end
of
Valero
so
that
you
know
it's
working
there
directly.
F
Sure
I
think
the
quickest
way
is
obviously
to
give
a
description
of
how
we
are
doing
it
now
and
then
then,
depending
on
how
it
goes,
we
can
even
look
at
the
code
and
you.
F
I
B
The
most
important
thing
is
just
you
know:
let's
see
how
this
is
done,
let's
get
some
information,
you
know
the
exact
way
of
communicating
that
is
less
important
to
me
at
this
point.
I
Okay,
Sean
you're
you're
with
radar
team
as
well
just
checking.
I
Okay
cool
so
we'll
we'll
reach
out
and
and
we'll
we'll
schedule
something.
A
Thank
you
and
we
jumped
on
topic
but
materials.
You
wanna
share
your
ideas
for
automations
of
the
trench,
lock.
K
Hi
everyone
I
saw
that
I
have
an
idea
for
that,
but
I
saw
that
should
do
that.
I
would
have
to
change
a
lot
of
files
in
Valor.
So
I
was
wondering
how
we
could
talk
about
this
if
I
open
an
issue
or
or
just
because
this
idea
is
basically
instead
of
when
someone
propose
a
pure
to
add
a
change
log
file
after
the
pr
this
year.
I
would
commit
on
top
of
that
on
the
markdown
file,
for
example,
of
the
released
change
Logs
with,
for
example,
PR,
number
user
and
APR
title.
B
B
The
process
that's
required
for
submitting
a
PR
is
to
either
provide
a
change
log
entry
for
that
PR
or
to
add
the
label
saying
changelog
not
required,
and
it
sounds
like
the
proposal
here
would
be
to
try
to
automate
the
actual
change
log
entry
from
the
commit
message,
rather
than
having
to
put
it
in
directly
I
actually
have
no
idea
whether
that's
something
that
was
considered
in
the
past
or
not
I.
B
I
do
know
that
you
know
we
do
have
that
changeling,
not
required
label,
because
sometimes
you
know
you
make
a
commit
that
doesn't
really
need
to
be
in
the
change
log
you're
fixing
a
typo
you're
changing
you
know,
certain
minor
changes,
don't
need
change.
Log
entries
and
the
way
we
handle
that
now
manually
is
either
to
add
the
changelog
file
or
to
add
the
change
log,
not
required
label
I.
K
For
example,
if
we
just
change
documentation
code,
then
we
don't
trigger
this
job.
For
example,
right.
B
B
Right
right,
the
rules
and
but
the
the
lit,
if,
if
there's
a
way
to
make
use
of
life
because
sometimes
there's
a
change
even
in
the
code
that,
because
you
know
maybe
you're
fixing
a
typo
and
a
comment
or
something
like
that,
so
you
don't
need
to
change
lock
entry,
even
though
it's
changing
a
you
know,
go
file
and
again
the
manual
system
handles
that
by
just
putting
the
label
on
there
so
I'd.
B
A
Yeah
I'm
missing
that
information
as
well.
That's
first
might
ask
if
someone
from
the
so-called
release
team
is
on
the
call.
So
we
can.
We
can
have
some
insights
on
this
one,
but
I've
seen
and
I've
implemented
that
to
other
projects.
A
Do
you
have
the
the
automation
release
notes,
so
we
can
introduce
like
I,
don't
know
five
levels
or
something
which
we
add
to
the
pr
and
then
the
release
notes
are
automatically
built
based
on
this
labels
that
we
can
work
on
this
one,
but
I
think
we
currently
use
this
kind
of
some
something
like
that
so
but
I
kind
of
comment.
L
B
Release
time
that
takes
all
those
unrelated
chains.
Oh
sorry,
all
those
unreleased
change
logs
builds
them
into
a
change
log
entry
and
then
delete
some
of
those
unrelated
unreleased
change.
Logs
files
I
do
believe
that
they
use
the
script
to
generate
the
change
log
for
the
release
from
those
files.
Yeah.
L
So
Mata
was
the
question
here,
for
you
is
in
your
idea.
Would
you
like
to
create
yet
another
PR
for
an
APR
that
requires
same
change,
block
or
Commit
This
code
automatically,
or
do
this
during
the
release
process,
so
help
release
engineering
to
crowd
the
data
and
generate
the
change
lock
when
it's
required?
What's
your
idea
here.
K
K
Will
be
stored
can
be
in
the
same
folder
that,
instead
of
one
file
for
each
user
or
hbr
it'd
be
one
folder
with
all
the
the
formatted
string,
for
example,
PR,
title
pair
number
and
a
user
that
created
it.
K
B
I
mean
if,
if
that
results
in
a
separate
commit
for
each
PR,
then
then
you
know
if
we
need
to
get
logged.
Your
commit
history
is
going
to
have
twice
as
many
entries
one
for
change
logo
and
for
the
and
it
was
after
PR's
marriage.
B
If
we
then
I
mean
I
suppose
if
you
had
an
action
that
modified
the
pr
and
then
you
know
added
them,
basically
where,
where,
when
you
merged
it
would
squash
those,
then
it
you'd
be
fine,
but
well
we
wouldn't
want
a
situation
where
all
those
commit
changelog
additions
or
separate
commits
in
the
release
in
the
commit
history.
L
L
If
someone
is
just
showing
this
is
to
actually
create
an
automated
draft
for
the
release,
which
will
add
those
entries-
and
this
happens
outside
opening
commit,
so
each
company
that
requires
change,
locked
will
update
the
draft
release
and
out
of
the
draft
release,
the
change
load
can
be
created
during
the
new
release
process,
and
this
way
we
skip
the
change
log
folder
and
we
store
everything
as
the
GitHub
draft
release.
Here.
A
It
was
I
was
trying
to
show
exactly
this
one.
We
do
that
for
another
project
called
Harbor,
so
that's
functionality
out
of
GitHub.
You
have
that
template
for
the
for
the
release,
notes,
that's
exactly
what
you
I'm!
Sorry
I
see
your
name
just
as
miji.
A
That's
what
we
do
for
other
projects
as
well,
so
we
kind
of
skipped
the
whole
process
of
getting
all
the
the
information
for
odprs
as
we
pass
that
is
done
by
GitHub
automatically
so
but
materials.
Do
you
mind
if
you
open
an
issue
for
this
question,
so
the
other
folks
from
from
VMware,
can
can
actually
give
some
more
detailed
information.
L
L
A
A
So
I
I
I'll
suggest
you.
If
you
open
an
issue,
a
discussion
issue
or
actually
we
can
we
can
discuss.
We
can
do
that
in
the
discussion.
Let
me
see
if
we
have.
D
I
A
Vote,
yeah,
okay,
I'll,
take
that
I'll
create
the
the
discussion.
I'll,
add
some
some
ideas
on
my
side:
you
cannot
use
you
can
you
cannot
use
and
we
can
figure
out
if
we
first
of
all,
if
we
need
to
change
something
in
the
release
processor
as
of
now,
and
if
we
want
to
update
something
on
the
process,
what
kind
of
approach
we
can
get
so
and
and
and
update
the
the
whole
thing.
H
A
A
K
A
A
Okay,
by
the
way,
I
see
some
many
new
names
on
our
list.
Is
there
anyone
some
brave
soul
that
want
to
introduce
themselves,
and
maybe
the
left
I'm,
not
sure?
What's
your
involvement
in
the
project
and.
L
Yeah
I
can
start.
It's
me
how,
as
you
can
see,
I
didn't
update
my
name
here.
You
know
Bridge
I,
work
for
red
hat
and
I
join
our
ADP
team
recently,
so
I'm
still
new
to
this
project,
bolero
and
I
hope
to
get
on
board
with
this
project
soon.
So
my
contributions
would
be
valuable
rather
than
just
fixing
something
right.
L
I
worked
for
red
hat
for
maybe
seven
years
now
before
that
I
was
involved
in
many
open
source
projects,
including
gnome
and
part
of
open
Solaris.
So
that's
my
involvement
here.
Okay,.
A
Welcome
welcome
to
the
team
then,
just
just
for
your
information
I'm,
the
community
Elite.
D
A
I
work
at
VMware,
so
if
you
have
any
questions
around
the
community,
then
how
you
can
set
up
and
everything
so
just
ping
me
on
slack
or
something
we
can.
We
can
get
the
chat
going
and
work
on
this.
One.
H
A
Nice
welcome:
are
you
all
folks,
under
West
or
I'm,
getting
the
structure
wrong.
A
A
Yeah
about
that
I
mean
folks,
I
suppose
Miho
you're
in
you're
in
Europe,
in
Czech
right.
L
A
I
miss
misread
the
the
accent.
So
sorry
for
that,
so
are
you
folks
playing
to
to
go
to
keep
calm
in
Chicago
or
no
idea
yet
yeah.
C
B
I
have
not
been
told
that
there's
been
a
final
call
on
that.
Yet
so
you
know
unless
isn't
on
the
call,
but
as
far
as
I
know,
a
final
decision
has
not
been
made,
but
I
haven't
been
told.
No
yet
yeah.
A
B
I
guess
when
I
saw
on
the
on
the
on
the
coupon
Shanghai
page,
you've
got
to
be
registered
to
get
the
visa
document
from
them
and
I
we
haven't
registered
yet
because
I
was
waiting,
the
approval
but
I
guess
I
need
to
look
at
the
cancellation,
because
I
know
there's
a
point
where
you
can
cancel
for
a
full
refund.
So
maybe
you
know
at
the
point
now
where
we
should
register
with
the
idea
being.
B
We
need
to
know
the
deadline
to
cancel
if
we're
not
going
to
be
approved
to
go,
but
you
know
I've
definitely
been
thinking
about
that.
I
just
haven't
actually
started
the
process
yet,
but
I
need
to
do
that
soon.
I
Okay,
and
if
for
some
reason
Shanghai
doesn't
happen,
does
it
mean
more
radar
team
will
be
in
Chicago
or.
K
B
On
the
reason,
for
you
know,
if
it's
not
approved
because
Shanghai
is
too
expensive,
then,
but
honestly,
it's
so
expensive
that
maybe
that
means
Chicago
more
likely
it's
not
approved
because
hey
we
don't
have
travel
budget
right
now,
in
which
case
Chicago
would
probably
be
a
similar
situation.
So
I
I,
don't
but
again
I
don't
make
those
decisions
I
just
you
know,
they'll.
Let
me
know
that
I
can
go
or
not
so
I'm,
not
really
able
to
say
for
sure
foreign.
A
Yeah
I
think
you
can
apply
for
for
tourist
visa,
for
Chicago
for
Shanghai.
B
Least
I
know
the
event
does
offer
a
Visa
letters
which
I
guess
helps
the
process.
I
don't
know
I
bet,
but
yeah.
It
sounds
like
if
you're
registered,
then
you
can
do
that.
The
only
thing
to
keep
in
mind
is
you
know
without
a
final
approval.
B
You
know
I
need
to
know
what
the
deadline
is
for
canceling
registration
with
full
refund,
obviously,
which
again
I
know
that's
actually
pretty
late,
because
I
know
you
know
you
can
register,
for
example,
and
pay,
and
then
you
get
your
talk
approval
and
then
you
cancel
the
one.
You
know
because
you
get
the
free
registration
with
the
talk.
B
Yeah
I
I
mean
I
know
you
can
cancel
immediately
because
for
the
last
kubecon
I,
when
I
registered
I
forgot
to
use
the
discount
code
that
I
had
available
so
I
canceled
that
got
the
refund
and
then
re-registered.
You
know
with
the
discount
code,
so
I
know.
If
you
cancel
early
enough,
it's
it's
a
full
refund.
They
have
a
policy,
it's
like
within
so
many
weeks,
it's
full
refund
and
then
after
this
it's
only
partial
refund,
bye.
B
A
I
suppose
the
cost
of
living.
If
you
go
to
the
tourist
areas,
all
right,
it
can
be
on
the
expensive
site
in
in
Shanghai.
D
I
A
All
right,
joking
aside,
I
hope
all
that
can
work
out
and
the
deadline
for
sorry.
L
A
Yeah
I
mean
for
the
Shanghai
event,
not
for
yeah
the
Chicago
one.
Okay,
the.
D
A
I
have
my
Visa,
which
actually
expires
on
on
the
day
that
I
have
to
travel.
I
have
to
slide
really
on
the
same
day,
so
I
have
to
apply
and
get
renewed
of
my
Visa
just
to
be
able
to
get
on
the
on
the
on
the
plane.
So
yeah
I
know,
I
I'll
have
to
check
I
think
they
announced
something
but
I'm,
not
feeling
that
so.
L
Just
sign
in
on
a
page
24
hours
or
48
hours
prior
to
event,
and
that's
it.
A
D
L
L
D
C
I
Yeah
we
we
have
a
couple
of
guys
in
Warsaw
and
Krakow
as
well,
so
you
you
should
have
a
contributor
Summit
in
Poland
next
time.
L
That
will
be
actually
nice
to
at
least
know
them.
If
you
can
just
point
to
for
the
slack
so
maybe
we'll
meet
on
some
Meetup
or.
D
A
C
L
D
G
A
Who's
having
Argentine
cut
Pizza
I'm,
sorry
I
didn't
catch.
That,
though,.
L
Region
out
there
or
a
big
companies
are
paying
less
taxes,
I
think
sometime.
So
maybe
that's
why
and
they
have
quite
big
university
centers
there.
So
lots
of
engineers,
nice
so
yeah.
It
would
be
great
if
you
can
just
buy
me
to
the
people
who
are
involved
in
Bolero
working
there.
So
yeah
as
I
said,
Community
feeling
would
be
great
meet
face
to
face.
A
That's
great
the
European
present
on
in
the
project.
Maybe
we
can
do
something
local,
even
like
European
local
sounds
funny,
but
yeah
there
are
some
some
kcd
coming
up,
so
maybe
we
can
meet
there.
A
A
Sorry
sorry,
yeah
I,
I
yeah.
J
So
I've
joined
last
week
because
I'm
Daniel
gura
I'm
from
Reddit
I'm,
managing
the
quality
engineering
of
oedp
and
also
working
on
Valero
Upstream.
So
some
of
the
guys
here
like
report
as
if
you
ask
regarding
the
structure,
so
the
engineers
are
reporting
to
West
and
the
quality
engineering
is
important
to
me.
We
have
couple
of
guys
here,
while
the
engineering
is
mostly
based
in
the
U.S,
the
quality
engineering
is,
we
have
a
a
lot
of
people
in
Europe
and
mainly
in
Israel.
J
We
are
I'm
in
Israel,
but
we
have
also
in
Germany
and
Spain
and
and
some
in
India
as
well.
So
this
is
why
we
are
joining
this
call,
and
so
it
would
be
nice
if,
if
any
Gathering
will
be
done
in
in
Europe,
so
we
will
be
able
to
join
and
participate.
Actually,
we
sent
those
a
couple
of
guys
to
cubecon
in
Europe
like
a
couple
of
months
ago
and
that's
it.
We
are
looking
forward
actually
to
working
on
on
the
quality
of
of
Valero
and
maybe
submit
collaborating
on
some
tests.
J
Upstream
I
know
that
my
we
have
a
perfect
quality
performance
team.
They
started
already
participating
in
some
talks
about
how
to
to
improve
the
performance
scale
of
Valero
in
the
community,
so
looking
forward
and
hope
we
will
work
together
and
increase
this
area.
J
C
J
L
J
Hope
that
going
forward
you'll
have
more
and
more
capacity.
We
already
started
reviewing
some
of
the
purely
Upstream
features
and
we
will
be
adding
and
I
hope
that
maybe
we'll
start
adding
some
tests
upstream
and-
and
it's
like
talking
about
some
test
procedures,
whatever
yeah.
F
A
F
J
Are
we
are?
We
are
working
both,
but
we
are
mainly
driving
for
automation.
We
have
a
go,
go
Ginkgo
based
automation,
we
have
python
automation,
we
have
ansible
scripts
yeah.
So
obviously,
like
you
know,
we
are
driving
us
to
have
as
much
estimation
as
possible.
Manual
is
just
like
an
intermediate
steps.
F
Right
I
mean
automation,
obviously
is
a
more
long-term
solution
and
I
I
myself
have
to
you
know,
look
into
the
E
to
e
tests
that
velora
has
I,
don't
know
how
comprehensive
they
are,
but
definitely
there's
a
lot
of
benefit
in
installing
Valero
in
a
live
cluster
and
doing
a
lot
of
tests.
You
know
with
scripts
with
code,
for
example
in
Cloud
Casa.
We
use
Pi
test.
You
know
all
our
these
are
like
integration
tests.
These
are
not
unit
tests
right.
So
we
install
the
agent.
J
It's
it's
open
like
yeah.
If
you
go
there
to
the
oedp
repos,
you
you'll
see
our
tests
there,
the
testing
area
so
I
think
yeah,
maybe
some
of
not
not
everything
we
are
allowed
to
share
totally
like
Upstream
for,
like
you
know,
all
our
tests
that
sound
tests
you
can
see,
you
can
see
there,
but
I'll
be
pushing
more
and
more
towards
that's.
D
A
J
J
D
D
A
Nice
so
welcome
everyone
super
happy
to
have
you
actually
I,
really
like
the
excitement
around
the
the
performance
testing.
Is
such
super
super
cool,
so
anyone
else
wants
to
introduce
the
sales
or
I
think
we
can
hit
the
enter
button
for
today.
A
All
righty,
thank
you
very
much.
Everyone
I
hope
to
see
you
in
two
weeks.
If
you
want
to
join
the
other
one,
the
alternating
one,
it's
going
to
be
next
week,
so
yeah
keep
the
good
job
going
and
see
you
next
time.
Thank
you.
Bye-Bye.