►
From YouTube: Velero Community Meeting - Nov 12, 2019
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
Hi
everyone
and
welcome
to
this
week's
Caballero
community
meeting.
My
name
is
Jonas
Rosslyn
and
I'm
here
with
the
community,
the
Bolero
team
and
community
members
super
happy
to
have
you
all
here.
Let
me
share
my
screen
with
the
agenda
and
if
you
have
anything
that
you
would
like
to
add
to
the
agenda,
I'm
gonna
share
that
in
the
chat
here.
A
B
Kick
it
off
so
for
those
of
you
who
haven't
seen
any
of
the
announcements
we
shipped
1.2,
so
we're
pretty
excited
about
that
got
that
out
on
Thursday,
so
through
the
link
in
there
at
the
top
of
the
agenda.
So
you
can
get
the
latest
stocks
and
all
the
release
notes
and
the
upgrade
instructions.
So
please
check
it
out
and
send
any
feedback.
You
have
our
way
so
that
definitely
took
up
a
you
know.
B
You
know
we
want
to
we're
hoping
that
we
can
think
about
the
next
few
releases
and
kind
of
think
about
how
to
line
up
some
of
the
bigger
features
and
bigger
work
that
we've
been
talking
about.
I
will
say
that
you
know,
since
we're
approaching
the
end
of
the
year,
got
cute
calm
next
week
and
then
a
couple
of
holidays
I
think
that
you
know
the
rest
of
calendar
year.
2019
will
be
a
little
bit
slower
than
usual,
and
you
know
so.
Overall,
our
1.3
cycle
will
probably
be
a
little
bit
longer
than
usual.
B
C
D
C
C
The
last
couple
weeks
so
plan
to
respond
to
some
of
the
home
chart
PRS
today
and
tomorrow,
I,
don't
I'm,
not
gonna,
make
any
promises
about
them
getting
murdered
and
then
I'm
on
community
support
next
week
for
cube
con
I
think
basically
the
whole
team's
gonna
be
there.
So
community
support
response
may
be
slow,
but
I'm
gonna
try
my
best
to
respond
and
I
will
also
be
out
to
the
following
week.
A
C
C
Dillon
from
Red
Hat
is
doing
a
Valero
talk,
I
think
it's
related
to
their
fuse
or
effort
that
wraps
Valero
I
think
it's
like
the
openshift
three
to
four
migration
stuff.
But
there
is
a
another
Valero
talk.
That's
happening,
I,
think
that
is
on
Wednesday
the
day.
It's
it's
either
earlier
on
Wednesday
or
the
day
before,
I
can't
remember
exactly
when
it's.
A
F
I
had
and
I
dedicated
a
couple
days
to
it
last
week
in
there
is
also
blog
post
that
was
written
should
be
coming
out
at
some
point
soon.
I
do
need
to
update
the
helm
PR.
There
are
some
new.
There
were
some
questions
that
we
weren't
clear
and
I'm
not
going
to
be
at
the
community
lunch
lunch,
because
I
I
had
committed
to
attend
to
lead
one
of
the
tables
at
the
diversity,
lunch
and
hack
session
boomer,
and
that's
it
for
me.
A
A
D
Awesome,
alright,
odd
nom,
cool,
yes
mercy
just
been
keep
on
talking,
demo
prep,
oh
I,
guess
not
demo!
That
was
nice.
You
know
yeah
kind
of
finishing
up
the
slides
yesterday,
so
we
could
get
that
uploaded
and
I
also
need
to
kind
of
spend
a
bit
more
time
on
the
cig.
Apps
intro
slides
later
this
week
as
well,
so
see
I
kind
of
pretty
pretty
much
heads
down
this
week
on
on
just
that.
A
A
He
actually
found
a
bunch
of
really
nice,
workarounds
and
and
the
actual
way
to
do
this,
so
we're
gonna,
move
to
wards
Jekyll
for
for
all
the
websites,
so
blurred
on
Io
will
be
on
Jekyll
for
most
like
mid
December
or
something
like
that,
not
before
Q
Khan,
because
that's
just
nuts
but
yeah
mid-december.
Something
will
have
all
the
sites
moved
over
to
Jekyll
4,
which
should
make
build
times
a
lot
faster
and
things
like
that.
A
Another
thing,
so
the
we've
been
working
on
is,
of
course,
the
community
lunch
at
cube
con.
So
if
you
will
be
at
Q
Khan
in
San
Diego
next
week,
please
join
us
for
the
community
lunch
at
12:30
and
come
by
the
VMO
booth
or
look
for
the
Bolero
sign
in
the
lunch
area
at
the
convention
center,
where
we're
going
to
have
fines
on
a
few
different
tables.
We'll
have
maintained
errs
on
those
things.
A
E
A
C
But
people
were
asking
about
it
and
if
we
want
to
make
more
integrated
support,
then
it
might
be
a
little
delayed
simply
because
keep
gone
at
the
moment.
But
I
just
wanted
to
communicate
that
that
I
need
to
follow
up
on
the
channels
that
people
have
been
asking
on,
but
yeah
it's
it's
it's
on
a
radar,
but
I,
don't
know
that
it's
going
to
be
this
week
or
or
the
next
two
at
this
point,
I
think
I
mean.
C
B
Yeah
I
guess
the
one
other
thing
I'll
mention
is
that
if
you
are
using
the
helped
write,
you
can
always
still
use
the
Bolero
plug-in
add
command
from
the
CLI.
So
you
do
have
the
option
to
install
with
helm
without
any
plugins
and
then
after
the
fact
you
of
velaro
plug-in.
Add
it's
not.
You
know
not
the
ideal
UX,
but
it's.
It
is
another
way
that
you
can
approach
it
for
now.
A
G
Yeah
sure
so
the
talk
isn't
gonna.
Do
a
super
deep
dive
into
Valera.
We're
gonna
talk
about
a
couple
things.
So
basically,
a
net
is
part
of
the
oven
shift,
container
storage
team
here
at
Red
Hat,
so
she's
gonna
do
a
little
bit
of
talking
about
Brooke,
which
the
Rockhopper
ATAR,
which
allows
you
to
dynamically
provisioned,
set
volumes
on
a
Cabrini's
cluster,
and
so
we're
gonna
talk
basically
of
a
sample.
App
that
we'll
be
using
some
set
volumes
and
nuba
will
be
used
as
the
backing
store
so
with
an
s3
endpoint
from
nuba.
G
And
then
my
part
of
the
tork
talk
is
just
be
discussing.
How
well
arrow
can
be
used
to
backup
and
restore
the
set
volume
using
rustic,
and
then
we're
think
we
might
talk
a
little
about
the
CSI
snapshot
plugin
and
how
we
got
that
working
with
Seth
we're
not
totally
exactly
sure
yet
so
kind
of
working
on
the
presentation
but
yeah.
G
So
so,
primarily
I'm
gonna
be
talking
about
using
rustic
our
experience
developing
plugins
on
top
of
falero
we're
really
not
going
to
talk
much
about
openshift,
but
I
am
going
to
talk
about
how
easy
it
was
to
develop
plugins
on
top
and
yeah.
Do
a
little
demo
of
backup
to
restore
of
a
WordPress
app
with
the
set
volume
using
nuba
as
the
backing
store.
So
looking
forward
to
meeting
you
guys,
there
will
definitely
be
at
the
lunch
on
Wednesday
and
I'm
sure
we'll
be
attending
your
guys's
talks
too.
C
Cool
and
and
I'll
actually
throw
one
more
thing
out
there
I
found
out
today
that
the
CSI
volume
snapshot
API
is
the
the
CRTs
have
gone
to
the
one
beta
I
think
the
there.
The
biggest
change
that's
relevant
to
us
is
there's
an
extra
field,
but
I
need
to
read
through
that
and
confirm,
but
I'll
also
be
updating.
My
talk
to
me
to
mention
the
difference
between
that
the
Alpha
and
the
beta
and
all
I'll
have
to
look
through
that
to
see
if
that
affects
the
valero
plugin.
G
Well,
yeah,
actually
in
a
little
bit
issue
with
the
CSI
snapshot
stuff,
but
I
also
didn't
have
much
exposure
to
in
the
past
so
a
net.
My
my
co-presenter
has
worked
a
lot
of
CSI
in
the
past,
so
she
was
really
interested
in
Valera
support
for
it.
So
she's
kind
of
been
digging
down
that
in
the
last
week
or
so
yeah.
C
And
and
definitely
willing
to
talk
about
that
I
know
we
want
to
talk
about
it
more
and
we
thought
we've
also
kind
of
kicked
around
some
ideas.
So
right
now,
our
our
prototype
plugin
is
in
is
a
an
item,
action
plugin
and
that
kind
of
came
from
feedback
from
Anthony's
team
at
Delhi
MC
because
they
tried
implementing
it
as
a
volume
snapshot,
plugin
and
based
on
the
data
you
need
for
the
volume
snapshot
API
with
CSI.
C
We
didn't
quite
pass
down
the
right
stuff,
so
it
was
easier
to
do
the
volume
to
do
a
backup
item.
Action
plugin,
but
Steve
and
I
were
talking
at
one
point
and
wondering
if
we
could
fill
it
to
be
a
snapshot.
A
volume
snapshot,
plugin
I
need
to
reconfirm
with
Antony
on
like
what
what
exactly
we
were
missing,
because
it
would
be
really
nice
to
conform.
Everything
into
that
volume
snapshot
API
from
our
perspective,
but
that's
not
set
in
stone,
but
we
would
personally
I
would
like
to
move
the
CSI
volume
snapshot
stuff
forward.
C
They
want
to
move
that
forward.
I
know
they're
very
interested
in
hearing
user
feedback,
so
they
won't
hear
from
us
as
Valerio
users
and
I'm
sure,
from
open
check
users,
people
who
have
building
on
top
of
it
so
yeah.
It's
definitely
something
we
can
talk
about
next
week,
like
I'd
love,
to
hear
your
thoughts
on
using
the
your
Internet's
thoughts
on
using
our
prototype,
we're
using
even
using
CSI
volumes
snapshot
directly.
If,
if
that's
something
you
have
experience
with
yeah.
G
H
Yeah
and
then
this
actually
Anthony
so
I
played
around
with
that
prototype
I
think
a
couple
of
months
ago,
it's
been
awhile
I
need
to
refresh
my
memory,
but
if
I
remember
well,
I
think
that
the
reason
why
we
provide
your
approach,
where
you're
using
the
backup
and
restore
time
actions
was
so
we
tried
to
do
a
volume
snapshot
as
the
CSI
for
the
CSI
and
I.
Think.
The
issue
is
that
the
interface
itself,
the
volumes
natural
interface,
didn't
pass
down.
H
C
I
remember
it
required,
like
it
required
doing
a
bunch
of
calls
to
the
kubernetes
api
to
kind
of
gather.
All
the
information
you
needed
to
then
create
the
volume
snapshot
api,
the
CRD,
whereas
it
was
a
lot
quicker
to
just
say:
hey,
I
have
been
a
persistent
volume.
Here's
we've
got
the
handful
of
things
we
need
from
it
here.
F
D
C
By
the
just
for
some
context,
the
idea
for
it
came
from
I
think
Steve.
You
were
messing
around
with
putting
rustic
into
a
volume,
snapshot
or
plugin,
and
our
perspective
is
velaro
maintain
errs.
It
was
kind
of
like.
Wouldn't
it
be
great
if
everything
just
went
down
that
one
path,
so
we
didn't
have
a
path
for
volume,
stop
shutter
plugins
in
a
path
for
rustic
in
a
path
for
CSI.
C
So
that's
kind
of
the
reason
why
we're
thinking
about
exploring
that
I'm
not
saying
that
we're
going
to
change
it
to
that
necessarily
we
may
take
some
time
to
try
it
and
see
if
I
and
see
exactly
what
exactly
happens.
If
we
do
that,
but
I
I
know
your
team
tried
to
Antonia
I
know
the
the
biggest
hurdle
was
like
he
said
there.
Wasn't
it
wasn't
as
all
the
metadata
you
needed?
Wasn't
there
like?
It
was
with
the
backup
item,
action
version
right
and.
H
Also
I
guess
some
of
the
fields
like
I
think
availability
zones.
You
know
we
found
that
was
just
from
the
CSI
and
I.
Wasn't
that
relevant
in
some
cases,
so
I
had
a
question
so
I
think
a
few
months
ago,
I
might
have
had
that
within
the
cocky,
but
it
is
actually
moving.
The
entry
plugins
to
CSI
is
that
that's
something.
C
C
So
there's
work
to
provide
migration
paths
for
all
the
entry
plugins
to
out
of
tree
plugins
out
of
free
storage,
plugins
and
my
understanding
is
most
of
those
will
be
the
CSI
ones,
but
I
don't
know
that
for
sure
I
know,
I'm
90%,
confident
that
the
VMware
one
will
be
a
CSI
plugin
I
need
to
find
that
document,
because
I
know
they
have
a
deadline
on
it.
But
I,
don't
know
what
that
is.
H
C
Yeah
well
yeah
I
need
to
revisit
that
document,
because
I
know
one
concern
I've
had
is
part
of
the
experience.
Is
that
getting
CSI
set
up?
Is
it's
not
a
great
experience
at
the
moment
right
with
with
entry
plugins?
If,
if
you're
on
on
a
platform
that
has
entry
plugins
for
storage,
it's
a
lot
easier,
you
deploy
kubernetes
and
as
long
as
you
have
the
storage
settings
set
up
like
the
field
set
right.
It's
it
just
works
with
CSI.
You
have
to
do
a
lot
more.
You
have
to
deploy
more
manifests.
C
A
B
Folks
have
made,
but
you
know
definitely
appreciate
each
of
the
contributions.
I
know
I
think
we've
gone
through
the
kind
of
the
detailed
contributions
from
each
of
these
folks
in
in
past
meetings.
I
do
know.
One
of
the
ones
we
didn't
get
to
cover
recently
was
was
from
Antony
about
a
one
here
listed
on
the
top.
He
caught
a
bug
that
we
add
around
eing
plugin
subprocesses,
so
we
were
essentially
spinning
up
sub
processes
and
then
never
actually
terminating
them,
so
they
would
just
accumulate
within
the
the
valar
container.
B
So
he
caught
this
during
the
of
e
1.2
beta
testing
and
was
able
to
submit
a
patch
for
it.
So
that
definitely
was
a
goodness
issue
to
catch
I.
Think
it's
been
around
for
a
couple
of
versions:
I'm
not
exactly
sure
when
it
was
introduced,
but
it
was
really
great
to
get
that
fixed
for
1.2.
So
thanks
a
lot
for
that,
Antony.
H
H
A
E
I
was
gonna,
look
back
to
with
Nolan
and
initially
sure
on
CSI.
Are
we
planning
to
also
invest
some
time
and
maybe
test
out
what
windows
will
look
like
with
velaro
backups,
so
I
think
that,
probably,
as
the
community
is
maturing
a
little
bit
more
and
more
on
Windows
might
be
worthwhile
for
us
to
try
out.
C
F
E
E
B
Could
be
could
be
a
good
opportunity
also
to
you
know,
maybe
present
velaro
to
the
kubernetes
windows
community
and
see
if
we
can
find
some
folks
who
are
interested
in
it
and
you
know,
can
help
us
out
with
testing
because,
to
be
honest,
I,
don't
think.
We've
we've
heard
much
from
users
who
are
interested
in
using
it
on
Windows.
But
that
may
just
be
that
we're.
You
know
we're
not
in
the
right
forums.
If.
E
You
can
give
it
a
shot
on
our
end.
The
guys
are
working
on
CSI
for
Windows.
Is
there
folks
from
docker
and
right
now,
they're
pretty
busy
trying
to
get
the
CSI
stuff?
You
know,
keep
them
moving
and
creating
more
providers.
I
think
they're
doing
well,
I
think
another
doing
Azure
and
UTP
right
now
providers,
but
if
we
can
put
in
our
docket
and
try
it
out
there
now
I'll
bring
us
all
together
and
see
if
we
can
get
them
to
help
us
fix.
Whatever
is
necessary.