►
Description
Join us for Kubernetes Forums Seoul, Sydney, Bengaluru and Delhi - learn more at kubecon.io
Don't miss KubeCon + CloudNativeCon 2020 events in Amsterdam March 30 - April 2, Shanghai July 28-30 and Boston November 17-20! Learn more at kubecon.io. The conference features presentations from developers and end users of Kubernetes, Prometheus, Envoy, and all of the other CNCF-hosted projects
Deep Dive: Kubespray (a sig-cluster-lifecycle's project) - Rong Zhang, Suning.com
Kubespray is one of the most versatile kubernetes-cluster manager and benefits of an extensive and active community. Kubespray main focus is to deploy and manage production kubernetes clusters on bare metal and on all major cloud platforms. The deep-dive session will focus on making decisions with the community and opening discussions for the future of the project, the roadmap, the organisation, and how to contribute and help contributors more effectively.
https://sched.co/Nrr5
A
B
A
B
B
B
A
B
B
A
A
B
A
Supports
across
a
platform
and
across
architecture
deployment
and
our
community
was
established
in
2015
and
we
became
a
base
of
a
cube,
a
DM
since
2080.
You
just
need
to
bring
your
own
machine
to
run
and
support
and
install
ansible.
Then
you
can
deploy
Cooper
spray
and
we
are
certified
installer
of
bias
in
CF.
So
this
is
coop
spray
at
a
glance.
B
A
Coop
spray
in
the
past
need
to
be
chosen.
Now
you
can
try,
you
can
automatically
check
the
oil
as
a
system.
You
don't
need
to
choose
so
this
is
first
step.
Boot
spread
a
boot
strap
OS.
Second,
is
a
pre
install.
You
need
to
do
some
checks,
for
example,
D&S
hostname
check
on
network
plugin
check.
If
your
network
plug-in
is
not
within
coop
spray,
it
were.
B
A
B
A
A
B
B
A
It
will
generate
certificate
in
the
first
node.
It
would
generate
all
certificates
in
the
first
node
and
then
check
whether
a
certificate
is
present
on
the
second
node.
If
not,
it
will
copy
to
the
second
node,
so
it
will
copy
certificate
to
master.
To
all
the
knows,
it
is
support
member
client
and
at
me,
certificates.
B
B
A
Etcd,
if
you
have
a
three
etcd
and
if
the
two
fails,
then
the
entire
will
fail
and
then
the
api
server
high-availability
you
to
support
external
Alby,
for
example,
cloud
Alby
or
a
5lb.
Of
course,
it
support
other
herpes
I
will
not
go
into
details
and
copper
spray.
Also
support.
Look
why
be
based
on
and
GX
are
proxy.
B
A
Is
the
benefit
of
a
local
Albee,
it's
more
practical
and
has
more
economic
value,
whether
you
use
f5
or
cloud
Albee,
you
need
to
pay,
but
if
you
use
local
Albee,
if
your
usage
at
a
physical
machine
or
cloud
it's
free
of
charge,
so
it
has
a
better
value
for
money.
So
Cooper
spray
support
local
air,
be
let's.
B
A
Provider
I'm
not
going
to
detail
so
as
we
have
essentials,
souza,
debian,
container,
Linux,
Red,
Hat,
red
aura
atomic
Ubuntu,
Network
plugins.
It
depends
on
your
underlying
architecture.
For
example,
flannel
you
can
support
hosting
getaway.
You
can
put
in
them
use
a
hoss
get
way
to
separate
a
control
plan
from
other
plans
container
Angie.
We
support
ocular
cereal,
which
only
support
crl
under
redhead.
B
A
If
your
note,
if
the
number
of
notes
is
more
than
2000,
your
etcd
cluster
will
have
a
latency
problem,
you
can
divide
it
into
two
cluster.
One
is
to
store
metadata,
one
is
to
store
EDC
the
event,
a
turn.
In
this
way
it
will
improve
the
performance
of
etcd
h/a
mode.
I've
already
talked
that
cross-platform
deployment.
B
B
B
B
A
B
A
A
B
A
Of
which
actually
join
us
over
the
past
one
year,
that
is
of
the
past
one
year,
the
number
of
contributors
grew
by
over
200.
Now,
let
me
talk
about
the
close
of
contact
for
coops
for
a
community.
We
want
to
provide
production,
ready
clusters,
so
all
components
need
to
be
tested
and
all
components
need
to
be
safe
for
upgrades
all
components,
RLJ
ready
and
scalable
and
minimal
comprehensive
set
of
applications.
At
the
same
time,
curb
spray
is
inclusive,
all
components
run
on
all
supported,
operating
systems,
container,
runtime
and.
B
A
A
B
B
B
B
B
A
Currently
we
support
six
operating
systems
and
seven
natural
plugins
and
we
support
on-premise
and
cloud
deployment
and
our
deployment
strategies
include
oil.
One
separates
hose,
that
is
a
separate
Cooper,
master,
etcd
and
coop.
Note
cook,
master
and
etcd
are
deployed
on
one
note
or
they
can
be
separated
and
next
is
a
edge.
A
deployment.
B
B
Can
be
set
in
the
cube
net
kubernetes
no
default,
it
can
be
set
set
there
and
for
the
long
term
contribution.
For
example,
we
have
added
new
OSS
and
also
new
network
plug-in
or
new
storage
clogging.
If,
if
it's
a
plugin
that
only
work
on
a
single
OS,
we
don't
accept
that
and
also
a
helm
charts.
We
don't
accept
that
the
deployment
of
helm
charts,
for
example,
if
you
want
to
install
any
FJ
or
other
software
on
the
production
cluster,
you
cannot
just
use
one
step
to
satisfy
all
of
the
requirements.
It's
impossible.
B
A
B
2.0
in
the
past,
we
only
have
the
master
branch.
We
only
maintain
the
master
branch
for
the
other
branches.
Well,
they
are
not
that
stable.
For
example,
two
point:
seven
two
point:
eight
and
later
we'll
support
the
two
point.
Eight
point,
one
or
two
point
eight
point:
two:
as
long
as
we,
if
we
have
drugs,
we
can
fix
them
ourselves
and
also
we
can
use.
A
B
B
B
As
for
road
map,
this
is
the
road
map
now
for
all
of
the
options.
They're,
not
that
transparent
and
documentation
is
still
a
weak
link.
We
hope
that
the
configuration
is
more
transparent
or
we
can
make
them
into
parameters,
and
we
want
to
support,
is
sick
and
the
upstream
community,
for
example,
the
cube
ATM
component
config
and
a
etcd
admin.
B
So
for
a
podium
or
it
is
CD
admin.
You
know
that
for
component
config,
it's
for
the
configuration
of
each
component.
This
is
not
fully
released
into
the
community.
Yet
it's
just
that.
Somebody
in
the
community
mentioned
this
direction
for
a
cube,
a
diem.
If
we
want
to
deploy,
you
need
to
have
this
class
in
class
configure
and
prop
Rossi
and
cube
net.
You
can
figure
those
that
is
recorded
in
a
confetti
configure
map,
and
if
you
manage
a
cluster,
then
class
config
there
are
200
or
300
lines
of
config
files,
it's
hard
to
revise.
B
A
B
A
B
As
long
as
we
add
the
note,
then
we
run
it,
we
run
the
ansible.
Then
there
is
a
new
machine.
Yes
and
it's
automatic
right.
Yes,
really
yeah
yeah!
You
can
try
it.
Okay!
Thank
you,
I'm
not
done
yet.
Actually
so
an
interesting
topic,
do
you
think
Cuban
it
takes
to
be
Nazis.
Deployment
is
slow.
Yes,
it's
slow,
but
acceptable.
Three,
thirty
minutes
how
many
nodes
ten
ten
knows
30
minutes,
because
when
you're
downloaded,
you
need
a
lot
of
time.
B
B
B
A
B
It's
supposed
so
it's
a
command-line
tool,
there's
no
API
right,
no
API!
You
can
implement
API.
So
all
of
the
operation
is
based
on
command
line
with
parameter
to
upgrade
or
scale-up
right.
Yes,
okay,
I'm,
a
user
I'll
have
three
questions.
First,
tube
spray:
can
it
pack
binary?
You
can
download
binary
packet
and
other
things,
but
in
you
cRIO
what
in
China
you
cannot
access
to
it.
So
how
do
you
do
with
it?
Yeah?
You
can
listen
to
my
next
topic.
How
do
we
yeah?
B
Actually
there
is
this
operation,
for
example,
we
can
download
it
to
OneNote
yeah
yeah.
We
know
it
so
there's
the
delegate
option,
but
there's
a
problem:
the
delegate
node
when
it's
downloaded,
it's
/
temple
temporary.
So
one
day
later
the
temp
folder
is
claimed.
So
we
need
to
redownload
it,
but
we
know
the
volume
is
quite
big,
so
there
is
no
valid
option
to
put
it
in
a
very
good
folder.
That
is.
The
problem
now
speak
into
the
microphone.
So.
B
The
binary
files
and
image
do
we
have
the
repository
in
China,
yeah
yeah,
maybe
under
row.
There
is
a
folder
all
of
the
links
you
can
just
change
it
to
the
accessible
in
in
China,
but
we
haven't
found
it.
But
maybe
you
can
talk
about
that
later
and
also,
if
a
note
is
lost,
for
example,
it's
down
or
it's
deleted
by
fault,
then
in
the
infantry.
If
we
want
to
change
the
infantry,
will
the
script
this
script
will
fail
in
running.
So
what
do
you
do?
So?
B
We
can
just
try
to
use
remote
load
and
then
we
can
reinstall
it,
and
we
do
that,
so
you
uninstall
first
and
then
you
reinstall
the
cast
ADM,
because
in
it
there
are
some
files
that
can
cost
the
installation
itself.
For
example,
there
F
one
note
is
missing,
then
in
infantry
we
just
remove
the
line
of
that
and
then
we
run
the
movement,
the
run
the
road
again
right,
okay,
I
start
sit
class,
sir
life
cycle:
do
you
have
anyone
doing
the
alley
cloud
and
integrate
had
she
proposed,
but
in
the
in
the
system?
A
B
When
we
upgrade
with
it,
we
have
some
pain
points.
We
had
four
clusters
being
upgraded
by
that
and
the
other
key
cluster
is
very
important.
So
we
are
not
that
trusting
it
so,
for
example,
v4
the
grip
before
upgrade
for
when
it's
downloading
the
image
or
young
can
hang
there.
If
the
time
is
short,
it
may
just
jump
ahead.
I.
Remember
that
there
was
this
a
15
plus
3
master
node
we
upgraded
for
4
to
5
hours.
We
have
to
wait
there.
We
cannot
do
anything
else.
B
B
Because
if
you
switch
it
to
remove,
then
it
can
evict
some
parts
and
then
it
takes
some
time
yeah.
It
can
have
some
ends.
So
I
have
this
experience.
When
we
upgrade,
sometimes
the
image
or
yam
pack
can
be
downloaded
with
estable
with
some
node,
then
we
can
use
cube
spray
to
upgrade
the
cube
cluster.
Then
the
time
can
be
saved
for
downloading,
but
entering
the
maintenance
mode
is
really
time-consuming.
Thank
you.
One
minute
left
anything
else.
B
B
A
B
B
B
A
B
B
B
A
Why
etcd
Ferguson?
No,
it
seems
he
cannot
improve
their
performance
of
one
note.
Is
you
run
on
the
same
model?
Yes,
but
if
you
have
two
thousand
nodes
in
the
cluster,
you
can
have
why
etcd
for
metadata
and
the
other
is
for
event.
It
will
improve
the
read
and
write
and
the
speed
not
improve
your
performance.
Can
we
put
it
into
nose
if
I
put
it
into
a
nose,
you
can
improve
even
if
what
we
separate
them.
We
still
put
it
in
one
note
now.
Currently
we
can
only
put
it
in
one
load.