►
From YouTube: Velero Community Meeting - August 17, 2021
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
Hello,
everyone
and
welcome
to
the
belaru
community
meeting,
slash
open
discussion
today
is
august
17th,
2021
and
per
usual
we're
going
to
go
through
some
status
updates
and
then
discussion
topics.
So
first
up
we
have
bridget.
B
Hi
everyone,
so
it's
just
announcement
to
say
that
finally
got
valero163
released
so
huge
thanks
to
everyone
that
was
involved
in
in
doing
reviews
and
and
getting
that
out.
So
this
release
introduces
compatibility
for
kubernetes
122,
specifically
removing
the
use
of
apis
which
have
been
deprecated
and
then
removed
in
122..
So
if
you
are
running
once
more
and
bolero
on
kubernetes
122,
you
need
to
use
at
least
this
version.
B
Yep,
so
that's
that
part
for
the
rest.
This
week,
I'm
gonna
be
spending
all
the
time
trying
to
catch
up
on
pr
reviews.
Backlog
is
is
still
pretty
big
and
we
need
to
like
make
some
progress
there.
So
I'm
going
to
be
spending
some
time
on
that.
Also,
I
owe
a
feature
branch
on
the
repo
for
fun
to
put
the
plug-in
versioning
work
in.
B
A
Any
questions
for
brigade
any
questions
on
163.
A
B
A
Thank
you
all
right
dave.
I
think
this
is
both
a
status
and
a
discussion
topic.
C
Yeah,
well,
I
have
a
demo.
It's
very
short,
don't
expect
much,
but
it's
upload
progress,
monitoring,
I'm
just
still
plugging
away
on
it
and
I'm
gonna
start
pushing
in
more
of
the
code
reviews
because
it
kind
of
it's
a
seems
like
a
small
feature,
but
it
gets
its
fingers
into
a
lot
of
different
stuff.
So
but
yeah,
that's
that's
pretty
much
free
and
just
you
know
other
things
like
we
start
looking
towards
what
one
eight
would
look
like
very
early
discussion
on
that.
A
So
do
you
wanna
dive.
C
C
Okay,
so
what
this
enables
us
to
do
at
this
point,
so
I
have
a
the
new
item:
snapshotter
plugin
and
I've
done
a
kind
of
a
hack
in
the
csi
plugin
to
bring
it
under
this
new
api,
and
this
at
this
point
we'll
always
wait
two
minutes
before
it
says
that
it's
finished
with
that's
finished
uploading
into
a
into
a
backend.
So
here
we
are,
we've
got
the
backup,
is
running
it's
in
progress
right
now
and
as
we
go
through.
D
C
Back
everything
up,
let's
see
if
I
can
just
scroll
forward
a
little
bit.
C
So
now
I
went
ahead
and
started
and
kicked
off
a
second
backup
just
to
let
it
be
there
and
our
first
backup
is
still
in
progress.
At
this
point,
and
now
we
actually
see
this
one's
in
progress,
because
in
the
few
seconds
between
after
we
checked
the
last
one,
it
actually
moved
over
to
being
into
the
uploading
phase.
C
C
And
we
can
show
so.
The
other
thing
that's
happening
in
here
is
the
item.
Snapshots
are
part
of
the
backup,
so
they're
being
shown
properly,
so
this
will
fix
up
our
issue
with,
like
the
vsphere
plug-in
right
now,
with
the
vsphere
plug-ins,
don't
become
separate
things
in
the
backups
as
we
move
it
to
item
snapshot.
C
It'll
continue
to
be
here
and
we'll
fix
up
the
the
display.
C
And
then
here's
the
the
second
backup
is
continuing
at
this
point.
So
it's
in
the
uploading
phase
and
eventually
it
completes
as
well,
and
it
has
its
own
set
of
snapshots.
So
that's
it
for
demo
day.
B
No
no
like
because
I
was
I
was
giving
a
demo
to
the
the
folks.
I
was
doing
a
deep
dive
on
the
aws
plug-in
for
the
new
team
members
in
beijing,
and
that
was
one
of
the
things
that
had
come
up.
Was
the
discussion
around
this
because
we
were
looking
at
ebs,
snapshots
and
hi.
The
blur
backup
says
yep,
I'm
done,
but
you
go
to
the
aws
console
and
you
see
the
snapshots
are
still
being
created.
B
So
this
is
one
of
the
things
I
pointed
out
that
we
would
be
able
to
create
an
item
snapshotter
plugin,
essentially
under
the
aws
one,
which
would
be
able
to
show
that
progress.
So
this
is
exciting
to
see
it
move
forward.
So
thanks
for
the
demo.
C
Yeah
thanks
yeah,
so
that'll
clean
up
some
of
that
stuff
and
then
that
poises
us
towards
moving
towards
the
astrolabe
apis
internally
and
also
getting
the
data
movement
stuff
pulled
up
into
valero
so
that
we
can
move
things
out
of
snapshots.
Yeah.
D
So
the
so
that
issue
that
was
just
mentioned
with
the
snapshots,
not
you
know,
is
some
snapshot
saying
it's
done,
but
it's
not
really
done
yet.
I
know
that
was
an
issue
came
up
in
discussion
at
least
a
year
or
so
ago.
Now
I
know
we
ended
up
putting
something
in
place
in
our
fork
on
the
conveyor
side
and
then,
but
there
were
some
issues
with
the
way
it
was
implemented.
C
Different
that
you
need
in
here,
you
know
yeah,
kick
it
in
I'll,
put
the
item,
snapshot
or
apis
and
code
in
today
or
tomorrow
for
review.
C
C
Yeah
and
then
may
not
make
one
seven,
there's
gonna
be
a
lot
of
reviews
and
just
a
lot
of
cleanup
to
get
this
done,
but
it'll
be
in
if
it
doesn't
hit
one
seven
it'll
be
an
early
one,
eight
and
then
we've
got
to
go
through
and
start
converting
plugins
over.
C
So
the
existing
plugins
continue
to
work,
but
my
goal
is
to
get
all
as
all
of
our
existing
plugins
moved
under
this
item
snapshotter
as
soon
as
we
can,
and
then
we
can
get
rid
of
these
issues
like
aws
vsphere,
I'm
sure
there's
other
places
open.
Ebs
probably
has
this
problem.
D
Yeah
I
know
when
I
ran
into
this.
It
was
definitely
an
issue
for
aws
and
vcr
and
I
think
gcp
as
well.
Those
are
the
three
plugins
we
work
with
and
I
ended
up
putting
an
nrf
similar
code
in
place
to
sort
of
wait
for
that
to
be
done,
but
it
was
on
the
backup
side,
and
so
this
is.
This
is
a
cleaner
approach.
Once
that's
in
place,
we
should
be
able
to
basically
abandon
our
local
commits
there
to
do
that.
It
sounds
like.
C
Yeah
that'll
be
cool,
and
then
we
can
get
this
all
unified
together
and
eventually,
we'll
have
so
right
now.
The
the
uploading
is
is
strictly
internal
to
the
plugins,
but
there's
no
reason
why
it
has
to
be
so.
As
we
move
forward
with
like
putting
data
movement
into
valero,
then
you
know
it's
pretty
much
the
same
thing.
We
took
a
snapshot.
C
The
snapshot
uploading
is
in
progress.
You
know
it
may
be
because
plugin's
doing
it,
maybe
because
valeria's
controlling
it
whatever
it
is.
That
can
all
happen
after
the
the
backup
proper
has
finished.
We
can
move
on
to
take
it
to
another
one,
I'm
considering
if
we
should
get
some
limits
on
the
number
of
backups
that
we
can
have
outstanding
any
any
thoughts
on
that,
because
I'm
worried
I'm
gonna
wind
up
with
like
30
backups
on
the
uploading
phase.
B
Yeah,
I've
always
wondered
about
that,
because
we
do
with
this
like,
for
example,
the
schedules
that,
if
you
create
them
such
that
there's
they
happen
more
frequently.
But
the
time
between
each
backup
being
triggered
is
is
less
than
the
amount
of
time
it
takes
to
do
a
backup.
You
can
easily
end
up
with
like
a
backlog
that
continues
to
grow,
that
valero,
like
won't,
be
able
to
catch
up
with.
C
Yeah
yeah,
so
that's
that's
something
we
should
consider
here,
it's
kind
of
hard
to
say,
because
something
like
ebs,
you
know
you've
got
a
huge
amount
of
infrastructure
behind
ebs,
so
even
if
you're
running
ahead
of
ebs,
it'll
catch
up
sooner
or
later,
and
it
pretty
much
guarantees
things
the
vsphere
plug-in.
C
You
know
that
each
each
additional
upload
is
taking
system
resources
in
the
background
and
it's
eating
out
of
a
finite
pool,
and
you
also
got
into
issues
like
the
total
number
like
we
don't
put
a
hard
limit,
but
the
more
these
fear
snapshots
we
have
outstanding,
it
slows
things
down
and
also
the
longer
a
snapshot
is
outstanding.
The
longer
it
takes
to
consolidate
and
merge
it
down
when
you
delete
it,
because
all
the
changes
that
happened
after
that
snapshot
have
to
get
merged
into
the
base
underneath
of
it.
C
C
A
Thank
you
dave.
Does
anyone
have
anything
they
want
to
discuss
today?
Anything
they
want
to
bring
up.
A
All
right,
then,
I'm
gonna
dive
into
some
contributor.
Shout
outs,
see
here
all
right,
so
we
have
three
of
these.
A
The
first
one
is
from
tiger
cow,
villainy
software
engineer
at
openshift
who
updated
the
ibm
config.md,
so
documentation
changes
there,
making
sure
that
we
properly
handle
ibm
storage
super
nice.
Thank
you
for
that.
B
Yeah
I
can
chat
about
this,
so
lars
has
actually
gone
through
and
made
quite
a
few
of
these
changes
where
there
were
a
few
like
situations
in
blur,
where
we
weren't
checking
the
errors
that
were
being
returned
from
various
function.
Calls
one
of
these
actually
helped
us
uncover
an
issue
with
using
one
of
the
deprecated
apis.
So
I
think
if
that
hadn't
been
introduced,
it
allowed
us
to
to
find
another
bug
with
them.
B
A
B
Yes,
so
this
one
I
didn't
review
it,
but
when
rustic,
when
we're
performing
rustic
backups,
if
the
pod
has
been
is
like
completed
or
stopped
or
it's
just
currently
not
in
a
running
state
rustic,
our
backup
would
continue
to
try
and
back
up
those
volumes,
and
that
would
cause
like
an
error
because
the
volume
wasn't
accessible.
B
So
I
think
this
check
or
sorry
this
change
introduces
the
check
to
make
sure
that
the
the
pod
is
actually
running
before
proceeding
with
the
backup,
so
so
that
should
fix
that
error.
So
thank
you
very
much.
A
Yeah,
thank
you
marcel
yeah
awesome.
So
with
that,
thank
you.
Everyone
for
joining
have
a
fantastic
rest
of
the
week.
Oh
one
small
thing
so
next
week
we're
we're
doing
a
a
community
engagement
health
check
for
valero.
So
I'll
present
some
of
those
results
on
the
community
meeting
next
week
as
well.
C
Before
we
go,
turnout
seems
extremely
small.
I
mean
even
for
normal
here.
So
did
something
go
something
happen
with
the
invite
or
something.
C
Yeah,
okay,
yeah
elvenar,
usually
shows
so
I'll
ping
her
and
see
what
what
happened.
I
think.
B
She
she
that
message
saying
that
she
had
another
meeting
that
she
had
to
attend
and
had
to
skip
this
one.
Unfortunately,
but
yes,
she
didn't
know
about
it.
So,
okay.