►
From YouTube: 2021-04-06 Rook Community Meeting
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
All
right
the
recording
has
started-
and
this
is
the
april
6
2021,
brook
community
meeting.
We
will
start
as
as
always
with
our
milestones
and
what's
going
on
in
terms
of
releases,
so
the
the
pigs
coming
up,
both
the
patch
release
for
1.5
and
then
the
1.6
release
as
well
too.
We'll
start
with
patch
releases.
Here
we
had
yeah.
There
was
this
issue.
Here
was
a
good
one
to
bring
up
travis.
I
believe
you
brought
that
up.
A
So
so,
let's
see
so
we
have
like
about
a
week
or
so
that
we're
planning
to
do
a
patch
release
kind
of
around
the
same
time
as
the
1.6
release,
and
then
this
particular
issue
here
got
did
not
make
it
into
the
last
patch
release.
B
A
C
A
6916
endpoint
is
not
available
in
object,
store
yeah.
We
kind
of
mentioned
this
briefly
at
the
at
the
last
community
meeting,
travis
that
you
know
it
was
we,
it
didn't
make
it
into
the
last
patch
release,
but
it
certainly
is
a
good
candidate
to
be
included
in
the
next.
Oh
right.
B
B
C
A
B
A
Yeah
any
any
other
items
mentioned
that
would
be
big
candidates
for
the
next
1.5
patch.
B
Well,
I
guess
the
the
second
one
here
in
the
step,
the
step
dashboard
credentials
are
failing
to
be
set,
and
the
latest
step
releases
that
one
we
need
to
get
back
in
1.5
as
well.
So
superbomb
has
a
fix
for
that.
Just
doing
final
testing
and
the
first
one
on
this
list
is
one
we've
had
for
a
long
time.
I'm
going
to
move
it
out
of
the
blocking
release,
column
we're
we're
narrowing
in
on
the
fix.
We
may
have
it
in
time
and
tobacco
105,
but.
A
Is
that
on
the
1.6
board,
you
said,
or
does
this
not
affect
effect
of
the
like
1.6
and
builds
from
mainline.
B
B
A
A
So,
let's,
let's
bring
up
1.6
as
well
too.
So
tomorrow
is
the
proposed
feature,
freeze,
time
period
and
then
we'll
be
cruising
for
the
next
week
to
stabilize
and
get
the
release
out
on
at
the
15th.
A
Let's
take
a
look
at
the
1.6
board
here
and
we
have
looks
like
we
have
some
issues
with
our
ci
with
with
trying
to
integrate
in
with
the
new
south
pacific
release,.
B
Right
so
that's
the
main
issue,
we're
looking
into
right
now
and
given
other
distractions
and
priorities,
it's
been
hard
to
focus
on
it,
but
so
just
late
last
week
created
this
pr.
That
would
finally
run
tests
against
pacific
and,
of
course,
enable
pacific
as
a
supported
version,
but
the
ci
is
failing
in
some
scenarios
for
block
file
and
object
so
across
the
board
kind
of
there's
some
issues
to
look
into
there.
B
B
Right,
if
we're
going
to
say
we
support
pacific,
then
yeah.
We
need
the
test
to
pass
against
pacific.
A
B
Right,
I'd
still
like
to
try
and
branch
in
the
next
day
or
two,
even
if
that
issue
is
still
persistent
and
but
let's
I
want
to
see
how
far
we
can
get
the
next
day
or
two
and
maybe
it'll
just
be
resolved,
and
it
won't
be
an
issue.
A
Who's
who's,
focusing
on
that
right
now,.
A
All
right
any
other
like
to
do
column
here.
You
know
a
lot
of
these
will
just
you
know
they
won't
quite
make
the
cut
or
the
time
constraints
for
1.6
and
we'll
move
them
along
to
1.7,
probably
a
triage
to
see
you
know,
what's
still
still
relevant
and
still
high
value
to
include
in
the
plan
for
1.7.
B
Exactly
I
didn't
get
a
chance
this
morning
to
full
triage,
make
sure
things
I
want
to
make
sure
whatever
we
think
is
blocking
the
release
is
in
that
blocking
release
column,
even
if
it's
in
progress
or
in
review
yeah,
so
it
can
burn
down
exactly
and
everything
else
can
wait
for
a
patch
release.
Basically,
if
we
need
to
that's
that's
fair.
A
Okay,
any
other
notes
on
1.6
seems
like
the
priorities
are
clear
around
you
know,
getting
the
specific
support
in
getting
the
things
finished
off
in
the
block
and
release
column
like
the
upgrade
guide
as
well
to
make
sure
that
that's
that's
clean
and
good
for
the
1.6
release,
any
anything
else
on
1.6.
A
Cool
okay.
Well
then,
the
the
plan
is
clear:
there
we're
winding
down
towards
getting
a
release
out,
so
keep
focusing
on
that.
The
release
blog
image
did
I
I
I
pinged
matt
on
that.
Travis
did
you.
You
want
me
to
ping
him
again
too
and
make
sure
that's
still
on
his
radar.
C
A
Oh
yes,
that
also
yes,
okay,
cool,
so
plane
is
in
plans
in
place
for
1.6.
We
can
move
ahead
on
then
to
the
community
topics.
I
suppose
so
alexander
wanted
to
bring
up
a
particular
issue
here
about
a
sort
of
I
guess:
functionality,
regression
or
a
breaking
change
in
terms
of
the
way
metrics
are
defaults
to.
Is
there
a
link
to
to
the
issue
that
was
associated
with
this
or
any
more
context
on
this
one?
Maybe
travis
or
seb
or
blaine.
B
C
B
So
the
so
grpc
metrics
are
something
so
this
csi
reports,
these
metrics
and
it's
it
was
added.
They
were
added
a
long
time
ago
like
a
year
ago.
I
can't
remember,
but
really
they
they're
advanced
metrics,
that
don't
you
know
I
just
don't
need
very
often,
so
they
really
should
have
been
disabled
by
default
from
the
start
and
they're
also
using
a
port
that
can
potentially
conflict
with
other
things
running
in
the
kubernetes
cluster,
because
it's
running
on
the
host
network
and
so
the
yeah.
B
We
were
finding
some
downstream
examples,
at
least
where
this
port
was
a
problem.
But
upstream
people
are
constantly
asking.
How
do
I
record
because
it's
conflicting
with
something
else
too?
So
it's
not
just
downstream
it's
it's
upstream.
So
since
these
advanced
metrics,
we
didn't
see
it
as
a
breaking
change.
B
But
for
those
who
want
to
re-enable
it,
there
is
a
setting
to
re-enable
the
metrics.
So
that's
not
the
problem
so
this.
So
whoever
reported
that
I
can
remember
just
said
that
they
they
were
surprised
by
the
metrics
going
away
and
they
had
to.
But
then
I
think
they
were
fine
once
they
re-enabled
them.
B
Yeah,
that's
the
summary.
I
think
we
did
the
right
thing,
but
and
if
only
one
person
reported
they
were
confused.
I'd
say
it's
a
successful
small
breaking
change,
but
yeah
breaking
changes
should
always
be
carefully
considered,
and
this
is
one
that
just
didn't
seem
like
it
would
impact.
Hardly
anyone.
C
A
Fair
yeah,
I
think
in
general
yeah.
I
totally
agree
the
consideration
of
their
patch
releases,
not
not
containing
breaking
changes
and
something
I
would
completely
completely
agree
with.
So
like.
There
was
some
consideration
to
this
and
it
was
not
quite
a
typical
case.
So
to
speak,
but
and
and
then
at
least
at
least
the
your
users
aren't
like
entirely
blocked
that
they
can
and
re-enable
them
again
as
well
too,
so
that
that
is
at
least
a
good
thing.
B
A
And
let's
see
mike
added
some
a
note
on
the
seth
developer
summit,
which
is
tomorrow,
I
think
any
any
more
context
on
that,
one
that
doesn't
look
like
mike's
able
to
join
the
call
today.
B
B
The
topic
will
probably
mostly
be
around
for
dashboard
integration
and
using
the
saf
dashboard
to
drive
rook
crds
through
the
manager
module,
which
we
haven't
really
made
a
focus
until
now.
It's
from
rick's
perspective,
we're
really
driving
from
the
crs
and
through
kubernetes,
but
it's
always
nice
to
have
trying
to
have
a
ui
that
can
take
care
of
all
those
details
for
you
right.
So
that's
that's
kind
of
the
discussion,
cool.
A
Cool
is
quincy
a
typocephalopod
too.
A
Yeah,
it
is
yeah,
that's
where
it
comes
from
just
because.
C
A
All
right
sweet
awesome
so
that
discussion's
tomorrow
and
then
folks
can
add
topics
or
ideas
for
the
kind
of
brainstorming
planning
sort
of
sessions
that
will
be
going
on
tomorrow,
all
right
so
yeah,
I
just
pinked
matt
by
the
way
too
travis
and
he
said
that
he
could
get
that
within
within
a
day
or
two.
So
that
should
be
no
problem.
A
I
told
I
told
him
too,
that
was
like
the
release
is
going
to
be
running
next
week,
so
you
have
a
little
bit
of
breathing
room
he's
like
okay
I'll
still
get
it
done,
though,
so
we
should
I'll
say
that
over
to
you
as
soon
as
we
get
it,
then
thank
you
any
other
topics
or
things
that
are
on
people's
minds.
They
want
to
bring
up
here
at
the
end
of
the
topics
on
the
agenda.
B
Oh
actually
for
kubecon
one
thing
to
note
I'll
put
I'll
put
in
the
notes
here,
but
we
did
get
two
talks
submitted
for
rook.
So
there's
a
rook
lightning
talk
which
we
got
as
a
graduated
project.
So
I
submitted
that
one.
It's
just
like
seven
minutes,
long
or
whatever,
and
and
then
blaine
and
satoru
put
together
the
recording
for
the
rook
deep
dive
and
submitted
that
so
we're.
I
think,
we're
all
good
to
go
for
kubecon.
B
A
B
B
A
And
yes,
abdullah,
you
have
your
hand
up.
Did
you
wanted
to
add
something
to
it?.
D
Well,
hi,
I'm
new
to
luke,
so
I
just
joined.
I
need
some
help
in
getting
my
cluster
up
and
running.
Also,
one
of
the
questions
that
I
have
is
do
we
support
two
networks
for
the
self
pods
for
public
and
cluster
networks
and
if
you
guys
have
any
configuration,
I
can
use
that
that'll
be
great.
B
B
D
A
Yeah
travis
is
there
any
general
high
level
guidance
I'm
going
to
share
real
quick
just
about
you
know
using
two
separate
networks
for
the
you
know:
cluster
traffic
and
driving
traffic.
That's
that's
a
sure.
That's
supported
right.
Just
a
high
level
high
level
support
would
be
nice.
B
Right
right,
so
the
yeah
stuff
of
stuff.
Of
course
you
know
it
has
support
for
the
public
and
cluster
networks
and
the
cluster
network
is
for
the
the
back
end
or
the
osd
backfill,
and
that
repair,
where
the
replicas
also
go
across
that
network.
So
yeah
rook
does
have
support
for
specifying
that
separate
cluster
network
either
with
multis
sebastian
can
speak
more
to
that
or
there's
also
some
settings
in
our
advanced
settings
or
advanced
user
guide.
D
Do
you
have
to
use
multi
so
can
you
use
this
enterprise
cmi,
for
example,
is
multiple?
Can
I
need
it.
D
B
C
D
Did
I
try
the
configuration
with
host
networking
and
I've
got
run
into
so
many
problems
with
the
monitors
forming
a
quorum
right
so
with
the
host
network
configuration
that
doesn't
seem
to
work
and
it
seems
like
we
have
to
do
the.
I
wanted
to
also
try
the
motors
approach,
but
I
haven't
tried
it
yet.
I
just
need
to
get
more
details
on
how
to
configure
that,
because
you
had
to
set
up
a
couple
of
attachments
to
the
pods
and
stuff.
B
C
C
D
C
C
Yeah
just
open
a
ticket
here
with,
and
we
have
a
description
on
how
to
open
a
bug
so
that
we
have.
You
have
to
report
versions
and
provide
logs
and
everything.
C
A
All
right
cool
yeah
thanks
for
bringing
that
question
and
then
hopefully
we
can
follow
up
on
the
issue
and
I'll
slack
to
get
into
more
details
about
it.
But
you
know
it's
good
to
have
that
high
level
conversation
here
and
kind
of
get
some
direction.
So
thanks
thanks.
Everybody.
D
There
one
more
one
more
question:
for
example
like
for
to
use
volte
cni,
you
need
to
remove
the
cmi
that
comes
with
the
kubernetes
or
or
you
can
have
them
both
installed
on
the
cluster.
C
I
think
you
have
to
use
multis
and
then
used
a
cni
inside
motors,
because
multis
is
a
mere
meta
operator.
If
you
will
that
embeds
multiple
cni's.
C
A
All
right,
awesome,
okay,
so
that's
everything
on
the
agenda
here.
So
if
there's
not
any
anything
else
to
bring
up,
then
we
can
go
ahead
and
adjourn
for
the
week
and
then
we
will
work
on
getting
the
1.6
official
release
out
next
week,
along
with
the
next
1.5.10
patch
release,
2.