►
From YouTube: Mesos Developer Community Meeting (Feb 9, 2017)
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
C
D
A
D
D
To
be
q1000
people's
for
parameter
premises.
This
way
we
can
be
a
little
more
specific
about.
What's
being
that,
what
I'm,
like
so
bricks,
April,
we
have
all
these
things.
D
Listed
some
of
them
I
imagine
are
done.
Some
of
my
body
are
being
pushed
back,
Orson
or
death
years
old.
Let's
so
it's
kind
of
go
around
room
and
see
which
they
may,
as
you
can
address
so
raisin,
pretty
big
ticket
Joseph.
What
do
you
want
to
do
with
his
ticket?
You
want
to
see
your
business,
who
have
a
milestone
for
which
you
read
declared
estate.
G
H
It
could
be
if
you
need
it
up.
A
little
bit
question
is:
is
this
for
people
who
want
to
know
what
we're
focused
on
or
for
us
to
figure
out
like
which
particular
types
14,
because
I
think
generally
people
would
line
all
that
were
interested
that
we're
working
on
natives
on
windows
right?
They
might
not
all
right,
yeah.
D
D
G
I
G
G
G
I
I
personally
will
disagree,
including
the
first
pretty
pretty,
seizing
up,
2007
and
and
the
implementation
might
not
be.
Chocolate
might
not
be
easy
because
the
eternal
change
many
finally
recognize
it,
but
the
precise
almost
done.
You
still
have
a
couple
of
issues,
but
hopefully
we
can
have
got
some
conclusion
g
next
week,
so
that
I
said
those
time.
Change
from
that
and
sky
yoga
has
patches,
remember
rd
gen.
I
will
delete
something
useful
for
those
patches,
so
really
can
length
I,
don't
paper.
Okay,.
G
G
E
A
May
happen
well
within
a
few
months:
that's
not
right!
Okay,.
J
J
Ticket
Australian
yeah,
that's
a
separate
one
from
there.
It
wouldn't
be
too
much
harder
to
get
the
modules
exposed
in
the
same
way,
but
they
can
pull
it
back
and
me.
Could
you
try
to
do
that?
Yeah
yeah?
We
can
take
it
out
of
the
current
goals
until
we're
ready
to
start
working
on
emily.
Is
it
pick
either
an
epic
it
cook
for
that
one,
the
public's?
He
is
jessica,
kicker
I.
Don't
they
look?
Okay?
J
What's
the
one
thing
that
couple
were
very
much
so
Mesa
sphere
has
a
jinkins
job
that
runs
on
mesas
master
and
can
be
configured
to
run
on
any
review
board
request,
I
grabbed
with
a
white
listed
list
of
contributors
enough
so
and
with
that
we
can
get
a
win.
Those
bills
in
our
review.
Our
extra
review
bot,
as
well
as
root
tests.
J
Guide
to
the
public
is
losing
or
creating
an
open
instance
that
does
all
the
does
its
job
openly
and
a
lot.
So
people
can,
you
know,
see
a
review
and
see
that
oh,
this
failed
to
build
on
windows.
Click
through
see
the
logs.
Why
try
again?
What
does
that
have
to
do
with
the
opposing
the
superstructure?
Probando
say
I
assume
that
that
would
also
be
a
Jenkins
job.
That
I
think
this
is
not
the
King
controller.
D
D
K
G
D
G
A
G
G
Santos
they
should
know
that
out
improvement.
So
I
guess
we
are
making
one
increment.
Let's
skip
that
one.
We
are
adding
multiple
custard
support
right
now,
just
landed,
actually
we
might
be
making
some
more
slit
should
keep
it
away
from
us,
and
maybe
we
should
change
it
to
accept
children
in
progress.
I
guess.
G
Investigating
the
wrist
restaurant,
a
signature
machine
reboot,
so
this
is
what
Apple
guys
are
trying
to
work
on.
You
know
you
anthem,
apple
online,
so
there
are
a
lot
of
discussions
aren't
even
until
last
week,
so
there,
let's
keep
it
there
operator
a
PIV
one
skater,
a
baby,
one,
those,
I
would
say
actually
pull
it
down
into
q2
I.
Think
we'll
pull
up
this
back
up
depending
on
what
we
decide
for
our
backlog
for
AP
a--,
so
600
7202,
a
6-4
6-0,
think
there's
a
Mac
bugs.
G
J
Continued
existence,
so
we
didn't
yeah,
we
can.
We
can
close
that
one
and
then
have
an
improvements
one
because
we
haven't
done.
We
haven't
actually
done
all
the
CLI
work
for
the
Mesa
COI,
as
well
as
a
few
other
things
that
are
not
trying
to
call
it
out
change
man.
Yes,
we're.
Definitely
calling
that
a
feature
out
in
the
changelog
close
this
one,
Sunday
area
improvement,
one
yeah
I
do.
G
D
G
D
G
G
G
I
G
G
We
suffer
with
tcp
and
a
shitty
fee
and
command
health
checks
on
the
commanding
sector.
We
do
not
support
command
checks
on
the
default
sector,
which
we
are
fixing
now
and
they're.
Adding
a
new
type
of
check
called
just
check,
not
he'll
check
which
lets
you
not
kill
stuff,
but
actually
just
get
data
from
a
threat
sector
yeah
for
for
being
stuff.
That's
that
schedulers
want
to
do,
but
I
think
Nessa's
would
ever
act
on
so
I'm
calling
is
general
on
interpreting
check.
I
G
M
J
Release
yeah
I
can
talk
about
the
1.2
release,
so
we
are
at
we've
got
all
our
water
features
done.
I
am
about
ready
to
cut
a
branch
and
a
12
X
branch
and
bump
the
version
2
13
and
master,
and
then
I
will
update
the
change
log
in
the
docs
in
the
branch
and
we
can
cherry-pick
over
any
other
bloggers
or
criticals
that
land
next
couple
of
days
and
after
after
I
have
a
changelog
and
everything
ready,
we
can
cut
an
RC
1
code
on
it,
but
I'm
going
to
go
ahead
and
cut
the
branch.
J
J
I'm
time
to
reach
anything,
you
can
always
check
the
version
number
and
configure
a
C&C
make
lists,
and
that
will
be
what
version
it's
actually
going
into
unless
you
bothered
and
cherry-pick
it
into
a
release
branch.
But
for
the
time
being,
please
just
notify
me
of
any
patches
that
you
want
cherry
pick
so
that
I
can
keep
an
eye
on
them
and
I
don't
get
surprised
by
a
broken,
build
or
something
can
you
send
them
on
the
guys.
J
J
Jira
as
it
only
small
team
now,
so
we
had
we're
done
with
all
the
the
future
blockers,
though
I
think
there's
one
more
bloggers,
bug
that
we
can
share
it
again,
but
I
just
want
to
get
the
branch
cut
early
so
that
we
don't
introduce
new
blockers.
I,
don't
see
any
blockage.
Oh
yeah
I
think
we
there
was
one
that
was
lingering
for
a
little
bit,
but
we're
back
down
to
know
blockers
only
three
critical,
so
we're
doing
good.
Oh
yeah,
sharpen
those.
J
We
can
cherry
pick
them
if
they
land
before
you
know,
I
have
the
RC
and
start
the
vote
after
we
started
the
vote.
You
need
to
make
an
argument
that
it
is
a
release,
blogger
and
it
would
be
a
reason
to
cut
an
RC
2
I
suspect
that
there
will
be
some
things
and
there
will
be
an
RC
2.
But
you
know
we
can
those
decisions
when
the
bugs
show
up.
J
So
that
some
something
I
I
am
in
favor
of
it's
essentially
kind
of
what
this
branching
before
the
RC
one
is
yeah.
As
soon
as
you
branch
you
bump
the
version,
you
have
essentially
done
a
code
freeze
on
that
release
and
then
anything
further
past
that
has
to
be
cherry
picks
that
are,
you
know,
argued
/
shock
right.
J
D
You
feel
like
cutting
it
when
we
want
to
cut
when
we
want
to
release
is
early
enough
or
they
want
to
do
it
like
if
we
cut
a
release
every
two
months
and
in
history
get
some
bacon
it
like,
as
it
might
be,
like
you
know,
the
cut
of
range
I
don't
know
a
week
before
we
want
to
release
or
two
weeks
your
release
or
something
like
that.
That's
not
venues
terrific,
so
that
we
can
meet
the
release
date.
So.
J
Instead
of
like
yeah,
I
like
I
like
that,
but
you
know:
if
we
have
any
blocker
feature
work,
we
basically
need
to
wait
until
the
future
work
is
complete,
because
cherry-picking
like
one
or
two
patches
for
a
bug
fix
is
easy.
Cherry
picking
what
a
chain
of
12
that's
happened
after
some
refactoring
is
impossible
yeah.
So
if
the
it
once
we're
down
to
terms
that
look
like
they
will
be,
you
know
bug
fix
patches
that
we
cherry
pick.
We
can
cut
the
branch
as
soon
as
that
happens.
J
Podracing
ifta:
well,
it's
time
based
and
then,
as
soon
as
the
time
happens,
we
have
to
resolve
all
the
bloggers.
Ok,
but
a
cert
longer.
If
we
have
a
rolling
window
of
more
blockers
showing
up,
then
you
can't
cut
the
release
good
at
the
moment.
If
I
cut
the
release
and
it's
time
base
and
I
vote,
we
set
the
boats
for
RC
1
and
then
we
say:
there's
10
blockers,
great,
stop
the
both
you
know,
cut
the
you
know
fix
these
10
blockers,
which
could
take
a
month
and
then
kind
of
RC
tube.
D
G
Your
honor
I
mean
it
is
click
it's
not
strictly
time-based,
but
we're
still
trying
to
not
add
new
features
of
the
last
minute,
but
I
think
one
thing
we
could
improve
on
is
that
right
now
we
start
to
his
burn
down.
Failing
the
charger
thing
after
the
two-month
deadline,
she
probably
started
sooner
than
that.
G
Maybe
two
weeks
before
that
start,
the
dashboard
start
telling
people
to
not
check
out
my
new
features
at
this
time,
actually
just
focus
on
fixing
bugs,
and
ideally
we
should
agree
on
features
for
the
release
even
much
before
job
and
not
add
new
features
into
that
urn
you're
not
going
to
make
it.
If
you
add
some
feature
working
on
it,
it's
fine,
but
that
shouldn't
be
a
lease
blocker
and
me
to
get
agreement
on
that
much
before.
That's
not
for
something
we
have
been
doing
a
good
job
of
you
know
something
we
could
improve.
G
This
race,
because
it
took
like
a
month
long,
which
is
upgrade
and
what
we
anticipated
but
and
has
that
a
mostly
white
pixels.
No,
we
will
o'clock
and
two
or
three
features
some
features
kind
of
you
wanted
to
really
get
in
with
the
Allegiant
plan
for
it
until
the
very
end
and
I
think
just
can
unfortunate.
Maybe
like
a
part
of
this
is
we
would
say
two
months
is
not
too
quick
enough
for
bring
you
later
this
year.
Maybe
it
has
to
be
33
months,
something
and
then
I
fished.
J
The
other
other
option
is
released
rains
where
everything
goes
behind
a
disabled
flag
and
if
it's,
we
have
a
lot
of
things
that
are
for
changes
to
think
silly.
Libraries
like
stout,
but
a
lot
of
the
features
could
be.
You
know
you
put
in
kind
of
a
core
infrastructure
and
then
everything
is
disabled
by
default
and
then,
if
you
aren't
ready
to
enable
it
by
the
time
we
cut
the
release
that
feature,
is
you
know
disabled
for
that
release?
We
don't
even
announce
it.
It's
still
in
development.
J
G
I
think
the
problem
has
not
is
way
that
people
wanted
that
to
be
enabled
I
mean
if
they
were
looking
disabled.
They
might
have
said,
like
this
kind
of
Locker,
there's
parts
of
intention
there
and
we
won't
announce
it.
So
no
one
can
use
it
and
they
will
use
it
in
next
release
by
the
thing
that
happened.
First,
we'll
all
go
back
and
pick
an
example
really
wanted
it
in
one
or
two:
it's
not
like
you
said:
okay,
it's
okay.
If
it
miss
this
release
gets
into
hundreds,
okay.
G
M
C
Shorter
time
periods
for
time-based
releases
makes
it
easier,
I
think
to
pass
on
including
something
any
given
release.
Is
there
we
go
Switzerland
yeah,
it's
no
big
deal
it's
just
a
month
or
six
weeks
until
the
next
point,
and
you
like
confident
it
will
definitely
happen
in
that
time
period
and
not
be
delayed.
C
G
I
we
used
to
do
one
once
a
month,
I
think
back.
Then
the
feedback
was
too
often,
mainly
because
we
can't
get
only
one
version
compatible.
Things
are
efficient,
that,
like
10
releases,
come
three
months
and
what
the
hell
I
got
to
go
through
all
these
leaders
to
get
the
new
feature
additional
chunk
and
not
only
allowed
people
to
jump.
Maybe
that's
that's
a
finisher
for
them.
It's
probably
easier
for
us
to
do
like
a
monkey.
Ladies
I,
don't
know
if
you
can
automate
a
lot
of
this
release
which
well
the
other
thing
already.
G
No
wonder
one
month
is
also
because
it's
out
of
commitment
from
the
Phillies
manager
to
do
it
takes
quite
a
bit
of
time.
It's
not
as
simple
as
a
single
good
thing
that
it
looks
in
with
this
goes
honestly,
especially
back
forty
four
qualities
and
the
kind
of
shepherding
all
the
tickets
like
cat
herding,
and
all
that
now,
let's
reset
one
mother
is
probably
like
two
to
three
spent
two
months
is
ok,
but
I
think
you
get
to
a
place
where
it
would
should
be.