►
From YouTube: Velero Community Meeting/Open Discussion - March 3, 2020
Description
March 03, 2020
Agenda
Status Updates
[steve]
shipped v1.3.0!
GitHub release
blog post
getting go 1.14/modules PR ready (https://github.com/vmware-tanzu/velero/pull/2214)
working on moving code out of controllers, into packages – starting with backup & restore
https://github.com/vmware-tanzu/velero/pull/2299
more PRs to come
will finalize refactoring of restore/discovery code (https://github.com/vmware-tanzu/velero/pull/2248)
[nolan]
On community support this week
Worked through my review backlog yesterday
Reviewed Xing Yang’s ExecutionHook KEP upstream
Going to pick up some CSI work this afternoon - namely syncing VolumeSnapshots & VolumeSnapshotContents into object storage so we can use it in client commands and syncing them into clusters
Will be on Bret Fisher’s YouTube show talking about Velero on 4/30.
[ashish]
Was on community support last week.
Discovered a missing feature of importing snapshots with the csi-hostpath-plugin and implemented that in this PR Discover existing snapshots on disk int he csi-hostpathplugin repo.
Making progress on the CSI integration work.
CSI blogpost that I wanted to put together already exists at here
(carlisia’s on PTO)
[jonas] Finishing up the plugins page
Discussion Topics
[nolan] Backlog grooming
Michael has asked us to review our “Unprioritized backlog” in ZenHub and form a rough roadmap for 1.5 and beyond
Tag @skriss and/or Michael for any issues you’d like us to take a look at in the next 6-9 months
Contributor Shoutouts
Helm/Velero chart: https://github.com/vmware-tanzu/helm-charts/pulls?q=is%3Apr+is%3Aclosed
A
Hi
everyone
and
welcome
to
another
episode
of
the
Balearic
community
meeting,
slash
open
discussion
here
on
zoom
or
on
youtube,
if
you're
watching
after
the
fact
so
today
we're
going
to
talk
about
the
13.0
release,
we're
going
to
talk
about
a
bunch
of
stuff
regarding
refactoring
and
and
moving
a
code
out
of
controllers,
and
things
like
that,
so
yeah
strap
in
and
we
get
started.
So
it's
Dennis
updates,
Steve
and
I'll
share
the
agenda
as
well.
B
Yeah,
so
yesterday
we
shipped
1.3-
she
had
been
tuning
in
you
knew
that
was
coming,
so
we're
happy
to
get
it
out,
so
I
put
links
to
the
github
release
and
the
blog
post
here,
where
you
can
get
all
the
details
on
the
changes,
a
lot
of
bug,
fixes
and
and
smaller
fixes
in
this
release.
But
the
I'd
say
you
know,
one
of
the
biggest
items
was
the
improvements
and
bug
fixes.
B
We
made
around
custom
resource
definition,
backup
and
restores,
and
all
and
really
kind
of,
spearheaded
this
and
waited
through
a
whole
bunch
of
related
issues
here,
partly
related
to
the
API
group
version,
change
and
and
also
related
to
some
other
things.
So
the
blog
post
goes
into
some
excruciating
detail
on
on
what
those
fixes
were,
but
we
hope
that
this
is
this
should
be
in
a
in
a
better
State.
B
Now,
as
of
this
release,
and
then
the
you
know,
the
other
big
item
that
we
wanted
to
call
out
is
that
we're
we're
now
shipping
multi
Arch
docker
images
for
the
main
velaro
image.
So
we
have
not
just
the
linux
amd64
image
anymore,
but
also
arm
images,
arm-in-arm,
64
and
and
PowerPC
as
well,
so
usually
using
docker
manifests.
So
as
a
user,
you
don't
have
to
change
anything.
B
You
just
specify
the
image
the
same
way,
you've
always
done
with
Valero
/
Valero
and
the
version
tag
and
you'll
get
the
correct
architecture
fold
for
your
hosts
and
then
beyond
that
you
know
we
have
a
bunch
of
other
bug
fixes.
So
there
are
a
few
called
out
here
in
the
blog
post
that
you
can
look
at.
If
you
look
at
the
changelog
in
the
they
get
home
release
itself,
you'll
see
a
full
list
there,
so
definitely
wanted
to
you
know
say
thank
you
to
everyone
who
contributed
to
this
release.
B
C
B
And
then
yeah,
looking
at
probably
the
rest
of
the
week
for
me,
so
I
got
a
few
things
that
I'm
working
on
here.
So
we
have
a
PR
up
for
upgrading
to
go
1.14
and
also
moving
to
using
go
models
so
that
PR
is
pretty
close
to
ready
and
it's
more
testing
yesterday
and
so
I
think
this
is.
This
is
ready
for
a
kind
of
final
review
and
hope
to
get
that
merged
relatively
early
in
the
cycle.
So
folks
can
test
it
out
in
their
development
workflows
and
make
sure
everything's
working
as
expected.
B
B
So
I'm,
looking
at
moving
that
code
out
of
the
controllers
and
into
the
packages
which
I
think
cleans
things
up
from
a
development
perspective,
keeps
the
the
controller
strokes,
pretty
simple,
so
I'm
starting
to
work
through
that
and
the
the
first
PR
I
have
up
is
a
basically
a
change
to
how
we
kind
of
manage
the
informer
caches
and
basically
waiting
for
those
caches
to
be
populated
upfront
in
the
server
struct.
Rather
than
doing
it
in
each
individual
controller
that
uses
them,
which,
which
simplifies
some
other
parts
of
the
refactoring.
B
So
I
haven't
looked
at
this
in
a
couple
of
weeks,
but
it
I
had
written
the
code
and
done
some
testing
and
things
that
generally
looked
good,
but
I'm
gonna
pick
this
back
up
and
do
a
little
bit
more
testing
and
make
sure
everything's
good
to
go
and
hope
to
get
this
merged
before
too
long
as
well.
So
Nolan.
If
you
have
a
chance
over
the
next
few
days,
I
know
you've
looked
at
it
once
already
so
I
guess
I'll
update
here
once
I've
done.
D
Absolutely
so
so,
Steve
myself
and
a
few
others
have
kind
of
looked
at
some
of
their
tickets
that
you
had
in
the
backlog
flow
Valero
and
we
started
putting
some
priorities
on
some
of
them
kind
of
looking
through
some
of
the
bigger
investments
that
we
want
to
make
in
the
next
six
months,
nine
months
to
Alero
and
and
putting
them
into
the
different
stream
line.
Since
then,
how
I've
ever
anybody
has
any
feedback
on
some
of
those
items
or
big
concerns.
Please
reach
out
to
Steve
or
myself.
D
But
more
importantly,
if
there
is
one
or
two
features
that
you
have
not
parrot
eyes
and
you
feel
very
important
to
have
all
of
you
guys
know,
I
mean
our
backlog
is
big
and
you
know
we
don't
look
at
it
every
day.
But
if
there's
one
or
two
items
that
you
want
to
see
fix
in
Valero
in
the
six
months
nine
months,
please
find
those
tickets
comment
on
them:
tag,
Steve
and
I,
and
take
a
look
at
them
again.
B
Yeah
and
I'd
just
add
on
there.
You
know
we're
we're
continuing
to
work
on
CSI
snapshot
support.
That's
that's!
Definitely
a
big
priority
item.
Karlie's
has
also
been
been
working
hard
on
the
design
for
some
revisions
to
the
install
and
configure
x
to
hopefully
make
that
a
little
bit
cleaner,
and
so
those
those
big
priorities
haven't
changed,
but
we
are
really
trying
to
kind
of
wrap
our
arms
around
all
the
other
things
that
we
have
in
the
backlog.
B
C
Yeah
so
I
a
monkey
sport.
This
week,
I
was
going
through
my
orgy
backlog
yesterday,
so
I
didn't
get
to
all
the
review
items
that
I
had,
but
I
went
through
a
lot
of
them.
Not
everything
was
a
approval,
but
I
did
have
some
questions
like
I
was
going
through
the
go
114
and
modules
PR
that
Chris
had
and
testing
it
and
asking
some
questions.
C
The
first
thing
I
want
to
do
is
sink
the
volume
snapshot
since
not
following
stuff
that
contents
into
object,
storage,
kind
of
at
the
same
level
that
we
do
the
Valero
native
snapshot
Struck's,
so
that
we
can
do
some
client
command
descriptions,
and
we
can
also
look
at
syncing
them
into
clusters
and
I
need
to
do
some
evaluation
of
what
issues
that
needs
to
look
like
and
then.
Finally,
just
people
are
we're
on
April
30th
I
will
be
on
Brett
Fisher's
YouTube
show
talking
about
Valero
I.
C
A
Awesome
yeah
and
we
also
have
steep
slope
cut
from
the
contour
team,
who
will
be
doing
a
presentation
on
cool
yep,
so
yeah.
This
is
really
cool.
Brett
Fisher.
Does
this
live
YouTube
screen?
It's
I
believe
it
starts
around
noon.
Ish,
Eastern,
Time,
we'll
get
a
link
and
everything
in
here
before
this
happens,
so
people
can
join
in
and
watch
it
live
and
yeah
I
think
it's
gonna
be
super
super
fun
and.
E
For
me,
I
was
a
community
support
last
week,
kind
of
reviewing
PRS
and
getting
things
moving
along
and
kind
of
backfill.
My
time
looking
at
the
CSI
stuff,
while
doing
that
I
discovered
like
a
missing
feature
on
the
CSI,
was
the
first
plug-in
where,
like
it
wasn't
supporting
importing
on
disk
snapshots,
which
was
kind
of
critical
for
our
cluster
migrations
in
audio
testing.
E
So
I
I
have
a
PR
up
for
that
that
should
hopefully
get
merged
sometime
this
week,
I
already
have
that
fix
in
my
code,
so
I
am
I'm,
not
blocked
with
that
being
merged.
So
I'm
like
making
more
progress
on
the
CSI
integration
work,
and
we
spoke
about
having
us
blog
post
about
CA
setting
up
CSI
put
together
but
looks
like
something
like
that
already
exists,
so
I'm
not
going
to
spend
more
time
looking
at
it
or
working
on
it.
So
that's
that's
just
an
update
for
folks.
A
E
A
I
think
we
can
so
we
do
have
a
bunch
of
plugins
right
now
and
I
think
it
would
be
worthwhile
to
have
kind
of
a
how-to
here's
how
to
use
the
lair
with
different
plugins
and
then
once
we
have
the
CSI,
plug-in
and
I
think
most
likely.
This
will
be
after
the
ones
out
for
release.
If
I'm
correct,
we
can
have
a
specific
one
around
the
CSI
plugin.
Does
that
make
sense
yeah
that's
kind
of
what
I
was
thinking.
E
A
F
C
F
C
Really
briefly,
volume
snapshot
lives
inside
of
a
namespace,
and
is
you
can
kind
of
think
of
it
like
a
relationship
between
a
PVC
and
a
PV?
The
volume
snapshot
contents
is
the
representation
of
the
actual
snapshot
on
disk
and
that
lives
at
a
cluster
level.
It's
a
cluster
scoped
and
that
actually
points
to
what's
called
a
snapshot
handle,
which
is
the
snapshot
ID
that
lives
on
disk.
C
C
So
it's
a
chain
of
things,
I
think
I've
ever
seen
a
little
ppm,
PVC
yep,
it's
almost.
They
modeled
it
almost
exactly
like
that.
So
Valero
Valero
backs
up
both
those
objects.
In
order
to
get
you
the
full
thing
and
though
the
thing
I'm
looking
at
is
to
basically
take
those
and
put
them
directly
in
object,
storage,
so
Valero
can
access
them
for
our
things,
like
Valero
backup
described
in
the
client,
so
it
can,
it
can
display
them
more
easily.
A
D
D
B
D
A
Then
just
a
quick
update,
so
we
talked
last
week
regarding
our
attendance
over
at
cube
con,
so
you
could
meet
with
me
and
Nolan.
We're
not
gonna,
be
attending
cube
con
anymore.
So
but
we'll
be
here.
So
if
you
want
to
meet
us
and
chat
with
us,
we're
always
here,
I'm
slack
or
in
the
community
meetings.
A
C
So
actually
that
was
zone
that
was
mostly
what
Michael
and
Steve
we're
talking
about.
Where
Michael
and
Steve
had
asked
the
maintainer
zaheer
to
go
through
the
the
backlog
in
and
identify
stuff
in
the
unpro
unpriced
backlog,
they
had
asked
us
really.
You
know
about
three
items
to
start
building
up
a
roadmap
so
that
we
can
have
like
a
longer
term
vision,
so
I
just
wanted
to
like.
C
That's
mostly
what
I
wanted
to
bring
out.
There's
the
Zen
hub
view.
You,
if
you
don't
want
to
install
the
extension
on
your
browser.
I
will
update
the
notes,
but
there's
a
web
view
through
Zen
hub
that
decorates,
the
Zen
hub
decorates
github,
with
some
metadata
for
swim
lanes
and
things
like
that,
and
we
use
that
for
organizing,
but
our
own
priorities.
Backlog
is
pretty
big
and
I
think
it
would
be
helpful
both
for
us
and
for
the
community
to
have
a
slightly
longer
view
of
what
we're
planning
for
the
future.
E
Was
just
scrolling
through
some
of
the
stuff
in
the
unquiet
ice
backlog,
there
are
142
issues
in
the
n
prioritized
backlog
just
in
the
core
Valero
repo.
So
we
might
just
want
to
just
bite
the
bullet
and
scroll
through
that
and
ruthlessly
close
issues
that
have
kind
of
stuck
around
for
too
long,
and
some
of
them
may
not
be
relevant
anymore.
So,
maybe
now.
B
E
B
Know
we
could
we
can
get
more
aggressive
and
close
out
things
that
you
know,
while
they
may
be
valid
feature
requests,
aren't
things
that
we
see
ourselves
working
on
anytime
soon?
No,
but
it
is
also
nice
to
you
know,
leave
some
of
those
open
for
external
contributors,
who
you
know
community
contributors
who
may
have
particular
items
they
want
to
work
on.
So
we
can
talk
about
it.
Definitely
you
know
we
should
close
things
that
are
not
relevant.
I,
agree,
yeah.
E
B
A
B
So
big
thanks
there
and
then
I
know
we
had
we
continued
to
have
kind
of
a
stream
of
small
pr's
to
the
helm.
Chart
I'm,
not
I,
haven't
had
a
chance
to
review
them
in
details
that
we
have
a
link
here
to
the
commits.
But
you
know
definitely
a
thank
you
to
anyone.
Who's
helped
to
kind
of
get
that
up
the
up
to
snuff.
A
A
All
right,
then
we're
going
to
close
it
out.
Thank
you
all
for
joining
today
and
yeah
per
usual.
This
will
be
up
on
YouTube
if
you
missed
anything
or
if
you
want
to
revisit
something
it'll
be
up
on
YouTube
later
today
have
a
fantastic
week.
Everyone
and
we'll
see
you
soon
again
have
a
great
one.
Bye
night,
bye.