►
Description
We will create and troubleshoot a Kubernetes workload using Octant, and then open up the conversation for questions. Octant is an easy to use open source tool that lets you visualize a Kubernetes cluster using a Graphical User Interface. It allows you to see workloads with associated objects and has tools that allow you to understand if there are potential issues with your configuration.
Presenter:
Wayne Witzel III, Octant Maintainer @VMware
A
To
thank
everyone.
Who's
joining
us
today
welcome
to
today
CN
CF
webinar,
how
to
better
understand
kubernetes
workloads
using
octant.
My
name
is
Christian
Jones
I'm,
a
cloud
strategist
and
co-founder
at
arctic
foxes,
and
also
a
CN
CF
ambassador,
technically
I'll
be
moderating
today's
webinar
and
in
this
regard,
I
would
like
to
welcome
our
today's
presenter
Wayne
witzel,
the
third
who
is
not
a
maintainer
at
VMware.
A
If
you
must
given
items
before
we
get
started
during
the
webinar
you're,
not
able
to
talk
as
an
attendee,
there
is
a
Q&A
button
at
the
bottom
of
your
screen.
So
quite
technically
down
here,
please
feel
free
to
drop
all
your
questions
in
there
and
we
will
get
to
as
many
as
we
can.
At
the
end.
This
is
an
official
webinar
of
the
CN
CF
and
the
such
years
subject
to
the
CN
CF
code
of
conduct.
Please
do
not
add
anything
to
the
chat
or
questions
that
would
be
in
violation
of
that
code
of
conduct.
A
Basically,
please
be
respectful
and
nice
to
your
fellow
participants
and
also
be
presenters.
Please
also
note
that
the
recording
and
slides
will
be
posted
later
today
to
the
CN
CF
webinar
page
at
CNC,
f,
dot,
io
forward,
slash
webinars
and
with
that
I
will
hand
it
over
to
Wayne
to
kick
off
the
presentation.
B
Great,
thank
you
so
much
Christian
yeah
thanks
first,
let
me
again
say
thank
you
to
everyone
for
taking
time
out
of
your
day
to
come
and
learn
a
little
bit
more
about
octave.
So
with
that,
we
will
we'll
get
started
with
this,
with
this
webinar,
so
using
octants
platform
to
better
understand
a
workload
within
kubernetes
namespace
and
the
agenda
today.
B
So
I
just
want
to
say
that
like
I
am
the
thing
that
excites
me
about
octant
right.
It's
kind
of
this
statement
right
here
and
this
was
highly
extensible
platform
for
developers
to
better
understand
the
complexity
of
kubernetes
and
I.
Think
I
think
we're
building
something.
That's
pretty
exciting
and
pretty
awesome
for
developers
to
do
just
this.
But
this
question
is
or
this
this
this
statement
is,
is
a
little
kind
of
broad
and
so
I
want
to
just
touch
before
we
get
into
the
demo
kind
of
exactly
what
that
means.
A
B
A
B
You
very
okay,
great
so
yeah,
so
here's
the
quote
I
was
talking
about
this
is
directly
from
our
readme.
This
is
the
highly
extensible
platform
for
developers
to
better
understand
the
complexity
of
kubernetes
clusters
and,
more
specifically,
with
octant
to
really
get
a
high
fidelity
view
of
a
namespace
and
even
further
of
the
of
your
workload
as
developer
in
that
namespace.
B
So
we
do
this
by
octant
kind
of
ships
with
a
client
that
has
a
set
of
views
and
and
actions
that
you
can
do
within
a
cluster.
It's
kind
of
like
our
version
of
like
the
batteries
that
are
included
with
octant
and
all
of
those
views
that
were
that
we're
shipping
with
often
could
in
theory,
be
kind
of
developed
using
our
auctions.
Plugin
system,
which
will
get
in
we'll
touch
on
briefly
in
the
demo.
B
But
that's
where
that
extensible
platform
comes
from
is
the
component
based
design
of
octant
makes
it
very
easy
for
someone
to
take
of
take
some
custom
resource
or
take
some
third-party
tool,
or
some
other
API
that
they
want
to
visualize
within
or
as
part
of
or
alongside
of
their
cluster
and
their
workload.
They
can
do
that
with
with
plugins
and
octant,
and
so
that's
kind
of
what
that
statement
is
referring
to
is.
Is
that
the
the
extensible
platform
is
what
we're
providing
with
plugins
and
then,
where
our
goal
is
to
create
that
view.
B
That
really
makes
it
easy
to
understand
what's
happening.
So
some
of
the
things
that
I
personally
use
achtung
for
is
like
almost
every
day
is
the
just
a
UI
for
keeps
ETL.
That's
just
like
the
baseline,
like
there's
a
lot
of
things
in
cubes
ETL.
That
I
can
just
use
oktin
for
I
use
its
kind
of
better
understand
a
workload
that
I'm
not
familiar
with.
B
B
The
idea
for
this
demo
came
from
working
with
someone
who
didn't
have
a
whole
lot
of
kubernetes
experience,
but
they
had
a
lot
of
you
know:
config
management
and
ansible
and
and
other
experience
and
and
I
wanted
to
just
kind
of
watch
them
struggle
with
getting
a
work
load
running
up
in
kubernetes
and
using
octants
kind
of
help
that
provide
that
process
along.
So
that's
where
it's
kind
of
the
the
motivation
and
and
the
purpose
of
this
demo
came
from
so
with
that
I
am
going
to
move
right
in
to
the
demo.
B
All
right,
so
this
so
I've
downloaded,
octants
and
I've,
run
it
locally
and
currently
octant
will
look
for
your
cute
config
and
all
of
the
usual
places.
It'll
also
respect
that
you
config
environment
variable
and
it
runs
locally
as
you
as
the
permissions
that
that
cute
config
has
and
so
I
just
want
to
go
over
this
UI
quickly.
Kind
of
moving
from
you
know
top
to
bottom
right
to
left
kind
of
touch
on
each
item.
B
I'm
not
going
to
drill
down
into
each
item
right
now,
we'll
do
that
during
the
demo
and
then
people
can
ask
questions
and
and
things
during
that
process.
So,
while
I'm
demoing
I
encourage
folks
to
ask
questions
and
if
I
see
them
and
I'm
able
to
answer
them
kind
of
in
line
I'll
do
that.
But
we
have
Brian,
Sam
and
Millan
here,
also
able
to
field
and
help
with
the
Q&A
questions
as
they
come
up.
B
So
even
if
I'm
not
able
to
answer
them
live
during
the
demo,
they
may
be
able
to
you
and
then
at
the
end,
we're
having
a
dedicated
commit
to
you
and
a
block
of
time
as
well.
So
you
get
to
any
questions
that
may
have
been
missed
or
whether
we
didn't
have
time
for
while
going
through
this
demo,
so
first
I
just
want
to
so
we'll
start
up
here.
This
is
the
context.
Switcher
currently
I
only
have
a
single
context,
but
if
I
had
multiples,
this
would
kind
of,
let
me
switch
through
them.
B
Next
up
we
have
a
namespace
selector,
we'll.
Let
me
switch
through
the
different
namespaces
that
I
have
access
to.
We
have
the
label
filter
selection,
which
currently
we
have
no
filters
and
then
on
the
far
left
side.
Is
this
nav
menu?
It's
a
collapsible
nav
menu.
It
comes
collapsed
by
default
and
that's
actually
I'll
be
using
it
throughout
the
demo,
as
is
collapsed.
But
for
folks
who
want
to
see
more
of
the
context
and
and
see
more
of
the
labels,
they
can
slide
it
out
and
see
all
that
information.
B
Each
section
of
the
navigation
is
kind
of
divided
up
into
areas
of
responsibilities.
So
this
first
section
here
is
all
about
the
current
namespace
oh
and
includes
just
like
the
namespace
overview,
which
is
what
you're,
seeing
now
a
breakdown
of
workload,
information
and
discovery
and
load,
balancing,
config
and
storage
and
custom
resources,
there's
also
some
auerbach
stuff
and
events
and
we'll
get
into
some
of
these
throughout
the
demo.
B
The
the
cluster
overview
section
is
is
things
that
are
specific
to
the
cluster
as
a
whole
and
that's
stuff,
like
namespaces
and
your
custom
resources,
the
definitions,
the
CR,
DS
and
and
know
things
like
that,
and
then
finally,
there's
a
section
down
here,
that's
dedicated
to
plugins
plugins
that
get
added
to
octant
can
be
treated
as
a
module.
So
if
you
think
of
each
of
these
icons,
you're
seeing
are
a
module,
then
plugins
can
also
act
in
that
fashion.
B
So
the
current
namespace
is
pretty
boring.
There's
just
some
some
demo
things
here
that
I've
set
up.
So
what
I
want
to
do
is
I
want
to
create
a
namespace
that
we're
gonna
be
using
for
this
demo.
Specifically,
so
I
have
a
namespace
yamo
here
that
I'm
going
to
apply
and
that'll
get
us
our
CN
CF,
octant,
webinar,
namespace,
created
and
then
we'll
once
that's
there.
B
We
will
move
into
creating
the
deployments
and
the
configs
and
things
like
that
and
we'll
see
those
showing
up
so
applied
the
namespace,
and
now
we
can
see
that
it's
showing
up
here
in
the
namespace
list.
It
wasn't
there
before
so
we'll
select
it
and,
as
expected,
there's
not
much
here.
So
what
we've
done
is
we've
taken
this
little
JSON
server,
docker
file
and
we've
created
a
local
image
for
it,
because
this
is
what
we're
developing.
B
B
B
B
It
is
using
our
CN
CF
webinar
namespace,
it's
using
the
8080
port.
It's
got
our
image
that
we've
created
a
JSON
server
and
our
in
our
local
registry
and
all
of
that,
and
we're
also
going
to
be
using
this
service,
which
I
showed
before
same
thing:
8080
mapping
from
80
to
80
80
on
the
outside
the
selector
things
like
that.
So
there
there
are
some
errors
in
here
that
are
intentional
and
they
are
errors
that
were
made
when
doing
this
that
were
great
to
kind
of
fix
and
solve
with
Hawkins.
B
So
what
I'm
going
to
do
now
is
I'm
going
to
all
just
apply
the
the
yamo
that
I
showed
quickly
there
and
so
I.
Just
did
that
and
you
can
see
this
is
real.
This
is
real
time.
This
is
the
result
of
me
applying
that
yeah
mol
into
the
cluster,
so
the
yeah
mole
successfully
applied
in
my
with
my
cube,
CTL
command.
Everything
said
created
and
everything
went
perfect
right.
B
So
as
someone
new
to
kubernetes
right
I
would
think
oh
well,
if
it
applied,
it
must
work,
but
it
turns
out
that
is
not
always
the
case.
So
using
octant
I
can
see
that
there
are
some
issues
with
some
of
these
things,
we're
getting
there's
a
status
indicator
on
my
pods
and
my
deployment
that
are
showing
some
type
of
error
message,
and
so
I
I'm
gonna
use
octants
to
kind
of
try
and
determine
what
these
problems
are,
and
one
of
the
easiest
ways
to
do
that
with
octant
is
just
use
this
resource
viewer.
B
You
know
there's
some
areas
here,
but
there's
some
that
are
actually
pretty
clear,
which
is
that
this
container
image
is
not
present
and
if
we
look-
and
we
think
okay
well,
I
was
the
one
who
typed
in
that
container
image.
So
I
kind
of
know
know
what
it
should
be.
Maybe
right
so
we
can
go
look
and
we
can
decide.
Oh
okay!
Well,
let's
maybe
we
want
to
fix
that
so
the
involved,
so
we
can
click
on
the
event.
B
We
can
see
more
details
about
it
and
then
we
can
go
right
to
the
involved,
object
right
and
in
doing
so
we're
now
we
were
at
the
level
where
that
error
happened
right.
So
so,
with
a
couple
clicks
we're
now
putting
ourselves
into
the
context
of
where
this
error
message
is
originating
from,
and
we
know
that
this
image
is
what
is
what
is
wrong,
because
the
events
told
us
that
that's
what
is
wrong?
B
We're
also
seeing
the
current
state
right
here
so
I,
I'm
gonna,
plug
in
because
I
use
kind
and
docker
a
lot
that
I
that
I
had
installed.
When
my
when
my
friend
was
going
through
this
and
the
plug-in
will
show
me
my
my
local
images
right
and
one
other
thing:
that's
nice
is
that
you
get
the
tag,
but
I
also
get
the
images
that
are
in
my
my
kind
registry
here.
B
So
this
is
an
example
of
just
what
you
can
do
with
a
plugin.
You
know
it's
not
necessarily
related
to
kubernetes,
but
in
my
case
it's
related
to
kind
and
my
local
doctor
images,
and
what
I
want
to
do
here
so
by
here.
I
can
see
that
the
tag
that
I'm
supposed
to
be
using
is
actually
this
latest
tag
and
if
I
go
back,
I
can
just
use
the
back
button.
In
the
end,
the
browser
here,
if
I
go
back,
we'll
see
that
it's
currently
set
to
current.
B
So
that's
an
easy
correction
to
make,
and
this
is
exactly
so
exactly
what
happened
was
my
my
friend
went
and
they
said:
oh
well,
I'll
just
change
that
here
in
the
pot
right
just
its
latest
and
they
hit
update
and
you
know
it
worked
and
actually
it
the
the
it
got
applied
and
they
went
to
the
resource
viewer
and
everything
was
Green.
So,
like
Oh
perfect
now
now
I'm
set
everything's
green,
green,
green.
All
the
way
down.
That
means
it'll
just
work.
So
in
fact,
let's
well,
we
can
just
go.
B
We
can
just
go,
try
and
visit
this
thing.
We
can
see
what
it
does
says,
there's
no
healthy
upstream.
So
it's
it's
not
quite
working.
So
then
we
were
like
okay.
Well,
let's
go
back
into
the
pod,
which
we
know
is
is
what
is
the
thing
that
is
running
our
application
right
and
from
right
within
the
pod?
What
we
can
do
is
we
can
just
attempt
to
test
it
because
there's
this
ability
to
port
forward
directly
to
the
pod
in
our
pod
is
exposing
this
set
of
container
ports
on
8080,
and
we
can.
B
We
can
just
click
there
and
we
can.
We
can
test
it,
that's
working,
but
it's
not
working
so
okay,
so
that
makes
us
think
well
alright.
So
if
I
can't
connect
directly
to
my
pod
with
this
port
forward,
then
something
must
be
wrong
with
the
pot
itself.
So
the
next
step
was
to
go
into
the
logs
and
take
a
look
so
limit
to
the
logs,
and
we
saw
that
look.
This
is
running
right.
B
I
can
see
that
it's
running
I
can
see
that
that
the
the
the
server
has
started
up
and
it
appears
to
be
listening
for
requests.
So,
what's
so
what's
going
on
here
and
it
took,
it
actually
took
a
second
because
we're
both
staring
at
it
and
and
I
realized
that
what's
going
on,
is
that
this
is
here
it's
listening
on
port
80
right
and
if
we
go
back
and
look
oh
well,
we
set
it
to
8080
right.
B
So
simple
enough
right,
we
go
back
into
the
amyl
we
find
where
we
had
where
we,
where
this
thing
has
this
right,
and
we
just
added
here
right.
We
think
that's
gonna,
but
we
can't
end
it
there
right
it
and
we
get
this
semi
cryptic
message,
but
the
important
part
is
that
you
cannot
change
any
other
fields
of
a
pod,
except
for
the
the
allowed
editable
fields
right
and
container
port
is
not
one
of
those
things
that
you
can
change
on
a
pod.
B
Well,
how
do
we
change
it?
This
is
where
the
resource
viewer
comes
back
into
play
right.
We,
even
though
we
don't
have
a
really
deep
understanding
of
why
this
part
even
was
created
in
the
first
place.
We
just
know
that
it
happened
to
get
created.
We
can
deduce
that
the
deployment
that
we
defined
is
what's
actually
controlling
this
pod
right
so
and
granted.
You
might
know
that
if
you
read
enough
of
the
documentation
or
but
a
lot
of
people,
they
just
they're
copying
pasting.
They
just
want
to
get
up
and
running
right.
B
They
don't
want
to
spend
the
upfront
time
to
have
like
deep,
like
tacit
knowledge
of
kubernetes,
just
to
get
their
containerized
application
deployed.
So
we
move
back
into
the
deployment,
and
we
see
that
in
fact
here
in
the
deployment
our
our
container
ports
are
listed
as
8080.
So
the
other
thing
they
noticed
and
and
that
you'll
see
here-
is
that
the
current
image
is
still
current
and
remember
we
had
to
fix
it
previously.
B
B
B
B
Now
we've
applied
this
change.
We
have
a
new
pod
running,
you
see
it's
11,
12
seconds
old
and
if
again,
we
now
try
to
test
what
we
had
tested
before,
which
is
our
port
forward
right.
We
get
a
success
so
that
that's
definitely
progress
in
the
right
direction.
So
now
we
can,
we
can
say,
with
with
with
a
pretty
high
level
of
confidence,
that
this
deployment,
which
created
this
replica
set,
which
is
now
has
created
this
pod.
This
is
a
this
is
a
healthy
path
through
here.
B
So
with
that,
it's
we
can
probably
try
are
our
main
endpoint
again
right,
our
outside
service.
This
is
representing
coming
from
some
public
domain
and
into
the
cluster,
but
we're
still
getting
this
no
healthy
upstream,
which
is
not
a
I
mean
if
you're
familiar
with
you
know
other
serving
tools
and
nginx
or
H
a
proxy
views
in
the
past.
Where
you're
not
really,
you
kind
of
know,
maybe
what
that
means.
Maybe
you
google
for
it.
B
So
we
we
head
back
into
the
namespace
overview
to
kind
of
get
an
idea
of
what's
happening
because
that's
where
we
found
before
we
were
having
this
set
of
problems
right
and
and
while
we're
here
we'll
notice.
Some
of
these
old
replica
sets
still
hanging
out
because
of
the
image,
change
and
Achtung
gives
you
a
convenient
way
to
kind
of
just
remove
some
of
these
resources
and
things
that
you
might
not
need.
So
we'll
just
go
ahead
and
remove
those
old
replica
sets
while
we're
here.
B
So
as
we
scroll
through
this,
you
can
see,
we
have
our
ingress,
it
looks
fine
and
then
we
get
to
our
service.
Now
our
service
is
showing
some
type
of
a
status
warning
here
and
that's
the
service
has
no
endpoint
addresses.
So,
let's
just
let's
just
click
on
that
and
see
what's
going
on
here,
so
it
doesn't
seem.
B
There
are
any
end
points
that
seems
true,
and
if
we
take
a
look
at
the
resource
viewer
right,
we
can
see
that
the
ingress
looks
ok.
Alright,
we
can.
We
can
verify
that
and
come
in
and
it's
linking
to
this
JSON
server
service.
And
so
what
would
cause
it
to
not
have
any
end
points?
So
that's
probably
something
where
we
end
up
having
to
go
out
and
Google
and
that's
exactly
what
we
did.
B
We
googled
it
and
we
found
that
we
learned
that
services
right
they
their
endpoints,
can
be
done
the
selectors
right
and
if
you
have
the
right
selector
you
get
an
endpoint
created
automatically
for
you.
You
can,
and
in
our
case
we're
looking
at
our
selector
and
we're
noticing
a
typo
and
and-
and
that
is
something
that
we
can
correct-
begin
by
going
to
the
yeah
moledet
right.
But
what,
if
I,
don't
actually
know
right?
So
the
other
option
is
to
I
can
go
back
to
you.
The
well.
Is
it
the
deployment
selector?
B
Is
it
that
what
like?
What's
the
selector
I'm
supposed
to
use
here
and
octant
kind
of,
gives
you
a
convenient
way
to
navigate
through
and
try
them
right?
So
this
shows
you
this
your
list
of
potential
selectors
to
choose
from.
We
have
the
one
that's
currently
there
we
have
the
another
one
that
is
an
annotation
for
a
kubernetes
application,
and
then
we
have
this
other
one.
B
That
is
just
a
straight
app
:
JSON
server,
so
we're
going
to
try
this
one
and
submit
it
and
see
if
that
makes
a
difference,
and
when
we
did
that
we
immediately
got
this
endpoint
mapping
that
shows
up-
and
that's
also
now
reflected
here
in
the
resource
viewer
you
can
see
before
when
we
were
looking
at
this.
All
we
saw
was
this
deployment
replica
set
pod.
B
We
didn't
see
that
this
ingress
service
pod
right,
so
we
didn't
have
that
that
idea
that
those
things
were
connected
and
now
by
fixing
this
this
selector
and
having
it
select
the
the
correct
thing
and
what
we
presume
is
the
correct
thing.
We
can
now
clearly
see
this
this
chain
of
here's,
our
deployment
that
we
created
and
got
things,
and
then
here's
this
ingress
in
service
that
we
created
and
how
they
are
now
connected
to
this
pod.
B
So
that's
fixed.
Now
we
should
be
able
to
visit
our
workload
weight.
This
is
yet
another
error,
and
so
here
we
can.
The
the
benefit
that
we're
seeing
here
is
that
as
we're
working
through
these
errors,
the
air
messes
are
shifting
and
changing.
But
what
we
don't
need
to
do,
because
we've
already
done
this-
is
we
know
that
the
deployment
replica
set
pod
path
right?
We
know
that
this
works
right.
We
verified
that
our
pod,
through
this
port
forward,
is
doing
what
we
expect.
It's
running
the
workload
correctly.
B
It's
running
this
this
container
right,
so
we
we
can
remove
that
whole
line
of
like
troubleshooting
and
thinking
from
our
brains.
While
we
try
to
figure
out
what
this
means
and
going
back
to
the
resource
viewer,
we're
gonna
know
that
it
exists
somewhere
in
this
path.
Right.
We're
like
this
path
of
ingress
service
pod
is
where
this
problems
is
taking
place,
so
I'm
just
going
to
start
at
the
top
and
work
my
way
through
the
the
from
the
ingress.
B
B
So
if
I
click
through
to
here,
this
piece
seems
like
the
piece
that
is
there
something
wrong
with
it,
and
so
this
this
is
one
of
those
where
it
might
be
hard
to
find,
because
it's
it's
if
you're
staring
at
something
for
a
long
period
of
time,
it's
kind
of
easy
to
move,
eighths
and
zeros
around
and
not
notice
them,
but
it
one
of
the
things
here
is.
If
we
started
to
just
check
off
items
like
okay,
we
had
the
right
selector
its
cluster
IP,
the
ports.
B
Oh
I,
remember
we
had
to
fix
the
ports
before
and
if
we
go
take
a
look
at
what
we
were
like
back
at
the
deployment.
We
remember
that
when
we
were
looking
at
this
deployment
originally
this
was
set
to
8080,
which
makes
sense
that
the
service,
that's
controlling
that
deployment
or
that's
controlling
the
ingress
into
what
that
deployment
is,
is.
B
It's
just.
This
is
one
of
those
contextual
items
that
we're
hoping
you
know
we
can
these.
This
is
an
example
of
something
where
we're
hoping
long
term
acting,
can
kind
of
start
to
surface
and
drive
actionable
items
around
some
of
these
things.
You
know
where
it
could
have
some
intelligence
around
the
fact
that
your
deployment
is
on
on
80
and
all
of
your
your
pods
are
on
80
and
you
have
a
service
on
8080.
B
Maybe
it
should
be
80
right
if
there
was
a
little
thing
that
popped
up
here
and
just
said
hey,
this
doesn't
look
right
and
that's
what
we
mean
when
we
say
we
wanted
to
be
like
a
high
fidelity
view.
We
really
want
to
get
the
the
contextual
actionable
information
up
to
developers
so
that
they
can
do
something
with
it.
So
with
that
fix
on
my
service,
I
can
now
go
back
out
and
try
this
again
and
look
at
that.
So
we
have
a
successful.
B
Ingress
into
our
our
service,
like
into
our
pod
and
everything,
seems
to
be
functioning
and
working
exactly
as
we
would
expect
it,
and
now
our
resource
viewer
is
all
green,
our
namespace
overview
as
we
scroll
through
the
list
will
you
know
we
have
green
statuses
for
everything
and
we
no
longer
have
any
issues
and
things
appear
to
work
as
they
should.
One
of
the
things
that
I
wanted
to
mention
quickly
that
were
that
we
are
also
working
on
is
the
applications
view
which,
when
you
have
metrics
enabled
you
I
forgot.
That
kind
doesn't
support
metrics.
B
When
you
go
back
see
you
here,
you'll
see
that
there's
no
four
pods
and
we
can
see
the
pods
listed
here
as
well.
So
so
really
that's
that's
it,
and
so
now,
at
this
point
we
would
try
and
copy
and
paste
some
of
this
out
and
go,
make
our
yamo
edits
and
and
make
our
updates
to
ensure
that
we
are
preserving
all
of
those
changes
in
those
fixes
that
we
made
and.
B
A
Perfect,
thank
you.
So
much
thanks
for
the
great
presentation
is
Wayne
said
we
now
have
time
for
some
more
questions
feel
free
to
put
them
into
the
Q&A
box
at
the
bottom
of
the
screen.
If
you
still
have
some
I
think
we
have
quite
a
bit
of
time,
and
there
is
not
enough
question
for
as
much
time
as
we
have
so
far.
So
let
me
just
start
at
the
bottom
from
mic
Fowley,
which
is
kind
of
a
long
one.
Let
me
try
to
sum
that
up.
A
B
So
Mike
I
would
say
that
the
error
that
you're
getting
is
is
not
normal
and
that
the
chocolaty
install
has
been
known
to
to
work
well.
So
if,
if
you're
getting
an
unauthorised
issue
when
running
the
octant
executable,
it
is
likely
due
to
the
configuration
with
the
cube
config
and
if
you
wanted
to
come
and
get
some
help
interactive
help
getting
up
and
running
in
debugging.
What's
going
on,
we
have
the
kubernetes
selection,
and
that
is
when
we
have
the
right
slide
up.
So
people
can
we
have
the
kubernetes
octant
select
channel.
A
C
Let
me
take
this
one:
ok,
so
hello,
I'm,
Brian,
Lyles
I
am
I'm.
Actually
the
creator
of
octant
and
it's
just
nice
to
see
someone
actually
demoing
your
software.
It's
like
having
your
kid
grow
up,
but
when
it
comes
to
lens
and
infra,
and
what's
the
other
one
k-9s
really,
you
know
we're
just
a
group
of
people
who
are
trying
to
give
you
a
view
into
your
cluster
and
how
does
octant
differ
from
all
those?
Well,
no
one
has
a
resource
viewer,
and
that
was
like
be
one
resource
viewer.
We're
working
on
a
new
version.
C
Octants
goal
is
to
show
you
what's
going
on
in
your
cluster,
not
just
to
be
a
list
of
stuff.
Unfortunately,
we
have
to
have
a
list
of
stuff
to
make
it
useful,
but
really
like
Wayne
was
showing
we
know
like
when
your
ingress
service,
your
workload
is
wrong.
We
want
to
be
in
be
able
to
put
the
infrastructure
in
a
place
to
not
only
just
show
it
but
actually
say
your
thing
is
broken
because
of
this
reason,
and
a
lot
of
tools
are
just
meant
to
be
dashboards
or
show
you.
C
The
current
state
of
the
world
or
show
you
graphs
or
whatnot
auxins,
taking
a
different
approach.
It's
trying
to
make
you
a
better
user
of
kubernetes
and
what
we
do
some
of
the
other
stuff
yeah
yeah.
Definitely
but
that's
that's!
The
main
differences
are.
Our
actual
goal
is
to
make
you
mesh
with
kubernetes,
better.
A
B
This
doesn't
mean
that
it
won't
ever
be
a
cube,
CTL
plugin,
but
the
the
current
vision
for
octant
is
to
kind
of
make
it
something
that
you
can
run
and
easily
get
started
with
by.
You
know
pointing
it
to
configure
uploading
a
cube,
config
and
then
and
then
also
having
native,
or
you
know,
integrated
notifications,
so
that
you
can
see
when
things
that
have
are
going
wrong
or
when
you're
editing,
yeah,
moles
and
applying
them.
You
can
get
notified
of
status
errors
and
warnings,
and
things
like
that.
B
A
Cool,
so
next
one
by
Warren
is
with
all
the
manual
changes
to
the
yaw
Millinocket
and
the
original
source
of
the
cue
control
apply.
Apparently,
eventually
he
gets
out
of
date.
So
is
there
a
way
for
octant
to
tell
the
user?
You
need
to
update
your
original
file,
or
is
there
even
some
kind
of
get-ups
workflow
that
you
could
integrate
this
with
yeah.
B
B
But
that
is
something
that
I'm
very
interested
in
having
as
being
part
of
octant
is
to
kind
of
track,
some
of
the
edits
that
you
might
be
making
and
then
get
them
out
in
a
way
that
is
actually
useful
to
preserve
them
or
get
them
connected
directly
into
you.
Some
repo,
because,
as
you
saw,
there's
a
lot
of
like
metadata
information
that
could
inject
it
into
the
yamo
by
the
kubernetes
api
and
if
you
copy
and
paste
out
that
whole
thing
and
try
to
rerun
and
apply
it,
it
gets.
B
A
B
We're
definitely
targeting
and
people
who
are
developing
against
kubernetes
and
want
to
run
their
applications
there.
I,
don't
I,
wouldn't
say
that
it
is
not
useful
in
production
likely
if
you,
if
you
had
so
this
is
an
opinion.
This
is
not
anything
else.
If
you
have
edit
if
you're
cute,
if
you're
cute
config
has
the
ability
for
you
to
make
edits,
which
means
you
can
do
it
with
cheap
CTL
and
you're
running
octant,
then
there's
no
difference
between
having
that
edge
button
and
having
the
ability
to
apply
in
ooyama
with
cube
CTL.
B
Now,
if
you
do
not
want
somebody
to
be
able
to
do
that,
then
you
give
them
cute
config.
That
is
read-only,
so
they
can't
apply
the
edits
right.
So
so
no
I,
don't
think
it's
dangerous
I,
don't
think
it's
kind
of
productive
because
it's
it's!
It's
really
surfacing
what
you
can
already
do
anyway.
There's
no
and
you
don't
get
any
enhanced
permissions
or
anything
that
you
couldn't
normally
do
with
command-line
tools.
Just
because
you're
running
Acton.
C
A
C
We
can,
you
can
install
them
cluster,
we
don't
provide
the
directions
anymore
because
it
was
just
confusing
people,
but
there
is
a.
There
is
an
incantation
that
you
can
actually
run.
Octant
is
a
angular
app
and
a
go
app.
You
can
actually
run
that
inside
of
a
container
inside
of
a
class
there
and
then
use
of
service
and
an
ingress
to
expose
it.
But
that's
not
the
answer.
C
You
know:
60
70,
80,
90,
100
and
50
megabytes
memory,
depending
on
what
you
have
there's
no
way
that
anyone
can
run
this
in
cluster
with
a
lot
of
people
running
it
that
way.
So
the
idea
is
that
we're
going
to
have
to
do
some
some
magic
with
some
with
the
with
a
great
read-only
view,
and
then
some
trickery
around
the
readwrite
stuff
problem
is
that
kubernetes
api
does
not
make
this
easy.
It
was
made
for
different
types
of
like
one-off
tools
and
and
controllers,
but
not
for
real-time
interactive
tools.
C
C
C
No
one
has
no
one
actually
sat
down
and
figured
out.
How
do
you
actually
do
that
in
real
time
for
a
cluster?
It's
easy
to
make
a
one-off
configure
raishin.
But
what
do
you
do
whenever
you
have
deployments
that
generated
replicas?
The
replica
sets
that
generate
pods.
It
becomes
and
then
also
I
cheated
inside
of
octant
and
I
made
it
so
that
services
directly
connected
to
pods,
but
there's
actually
an
end
point
there.
C
We
I
would
love
that,
but
then
I
would
ask
people
well.
Why
do
you
want
that
and
actually
what
you're
probably
asking
for
is
a
better
tool
for
that
and
something
we're
thinking
about.
If
anyone
has
any
ideas,
all
open
source,
so
please
contribute
but
I
think
we
want
to
solve
that
problem
in
a
slightly
different
way
and
then
have
opposite
work
with
that
tool.
A
C
Supports
it
support
right
now.
I
would
wager
that
octant
supports
anything
any
kubernetes
greater
than
my
1.12.
We're
not
14
is
a
definite
I'm
running
it
on
with
1.18
I.
Have
it
I
have
it?
I
have
often
pointed
to
a
cluster
that
is
on
vSphere,
7
right
now,
so
yeah.
It
should
really
work
with
anything,
that's
kubernetes,
and
if
it
doesn't,
please
report
it
as
a
bug,
because
it
is
it's.
A
C
Well,
actually,
there's
an
easy
answer
for
this.
You
don't
so
here's
here's
a
very
important
thing.
Vmware
also
has
another
product
called
TMC
tongzi
mission
control,
which
is
actually
in
that
realm.
We
are.
We
are
very
targeted
in
our
open
source
efforts
to
ensure
that
it
is
one
cluster
one
namespace
at
a
time.
A
Good,
if
you
have
any
other
questions,
feel
free
to
put
them
in,
because
if
we
are
with
the
last
one,
octant
has
great
discovery
capabilities.
Do
we
have
to
subscribe
to
all
of
octant?
So
you
say
the
full
platform.
Can
we
leverage
parts
that
make
sense
kinda
and
that
octants
just
inside
some
other
application,
but
still
retain
some
of
the
extensibility
of
the
plugins.
C
Often
is
apache
v2
as
long
as
you
would
hear
by
the
Apache
v2
license.
We
do
whatever
you
want
now
we're
not
going
to
support
it,
of
course
not,
but
but
any
reasonable
request,
so
anything
that
we
have,
and
that
is
not
internal
but
anything.
It's
in
github,
comm,
/,
VMware,
Hines
news,
/,
octant,
/
package,
oh
I,
will
say
we
will
support
that,
but
anything
else.
No,
we
don't
have
a
stable
API
for
that,
but
we'll
I'll
definitely
talk
to
you
about
it.
If
you
were
taking
me
somewhere.
A
Cruel
there
are
lots
of
people
saying
thank
you
for
the
efforts,
especially
for
the
newbies
in
kubernetes
world,
so
I
think
that's
not
a
question,
but
that's
definitely
to
mention.
Last
but
not
least,
from
Connor
today.
Are
there
any
plans
for
better
integration
with
git
or
more
specifically,
github
within
octant,
so.
C
I'm,
assuming
this
means
around
like
a
get-ups
type
thing,
I,
don't
know,
wait.
B
D
D
A
Thanks
for
the
demo
is
that
a
question
I
will
send
that
onwards
as
well.
Thank
you
so
much
coming
so
also
Thank
You
Wayne
for
the
great
presentation
and
that's
all
we
have
for
today.
Thank
you
so
much
for
joining
and
have
a
great
day
everyone,
the
webinar,
recording
and
the
slides,
will
be
online
later
today,
and
we
are
looking
forward
to
see
you
at
the
next
webinar
by
CN.
Cf
take
care.