►
From YouTube: 2021-07-20 CNCF TAG Observability 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
B
I
guess
matt
I
saw
that
putting
stuff
on
the
agenda,
so
I'm
assuming
matt
will
join.
A
Great
by
the
way
I
had
today
a
recording
of
an
episode
of
open
observability
talks.
I
hosted
richie
I'll
compliment
him
in
his
absence,
so
he
can't
object
and
he
did
a
fine
job
representing
tag
observability
and
its
role
and
its
part
in
the
due
diligence
of
open,
metrics
and
open
telemetry
and
stuff.
So
if
you're
interested
in
listening,
I
highly
recommend
good
pr
for
us.
B
B
B
Hey
there's
richard.
C
Here
we
seem
to
have
a
problem
with
our
invite.
I
couldn't
join
the
call
just
now
without
logging
in
I
I
poked
amy.
You
can
see
them.
B
Maybe
that's
that's
you
richie.
Everyone
else
managed
to
literally
help
people.
D
Okay,
we
can
take
a
note
to
do
that.
Apologies
for
being
in
late
richard
did
you
already
sort
of
open
proceedings?
I
think
the
recording
started
as
a.
D
So
so,
just
a
point
of
order.
This
is
a
cncf
sponsored
meeting.
The
cncf
code
of
conduct
applies,
as
usual,
we're
kind
to
our
colleagues,
and
we
abide
by
that.
Please
don't
put
anything
in
the
comments
or
do
anything
that
is
in
violation
of
conduct,
so
we've
got
some
guests
today.
Richard
do
you
want
to?
C
D
Well,
with
that
welcome
to
folks
that
are
joining
us
from
the
octane
team,
I
will
let
you
introduce
yourselves,
because
you'll
do
a
better
job
of
that
than
I
take
it
away.
Wayne.
E
Hi
yeah
wayne
wentzel
I
work
on
octant
and
yeah
I'll,
be
showing
it
off
today,
just
kind
of
walking
through
it
for
folks
who
haven't
seen
it
before
and
answering
any
questions
people
might
have
about
it.
We
have
some
other
members
of
the
team
here
in
attendance
hanging
out
so
I'll.
Let
them
introduce
themselves
I'll
hand
it
off
to
david.
F
Hey
everyone
good
morning,
david
espejo,
here
open
source
community
manager
for
often
and
thank
you
so
much
for
your
time
and
for
giving
us
this
space
to
showcase
the
project.
Thank
you
and
I'll
hand
it
off
to
liam.
B
Hey
guys,
I'm
liam,
I
am
an
intern
with
the
octa
team
and
I'm
currently
working
on
plugins
and
I
will
pass
it
off
to
cara
hi
there.
My
name
is
carrie
yemenis,
I'm
the
engineering
manager
for
the
octant
team-
and
I
think
milan
is
the
last
person
here
we
have
from
the
team,
hello,
everybody,
milan
kleinscheck
software
engineer,
octane
team,
glad
to
be
here.
F
B
B
Hello,
I
am
luis
super
developer
for
the
arkham
team
yeah.
I
think
wayne
is
missing
or
I
don't.
B
E
E
So
yeah,
so
thank
you
yeah,
it's
great
to
be
here.
Is
there
anything
before
I
get
started
with
the
presentation
for
the
meeting?
Okay
awesome!
Well,
let
me
share
my
screen
and
we'll
get
going
a
quick
poll
in
the
in
the
chat
for
folks
who
who
has
seen
octant
before
or
is
familiar
with
it
at
all.
D
I
guess
I
could
start
and
say
in
my
day,
job
we've
been
using
it
almost
since
its
inception
as
a
way
to
debug
workloads
and
and
just
type
less.
E
Excellent
all
right
and
is
there
anyone
who'd,
be
willing
to
just
keep
an
eye
on
the
chat
for
me
as
I
go
through
things
and
call
out
any
questions
that
might
come
up
david.
Do
you
mind
doing
that
yeah
sure
excellent?
Thank
you
yeah.
So
is
this
large
enough
font
size
good
enough?
Does
it
need
to
be
embiggened
or
is
everyone
yeah.
C
E
C
C
E
Yes,
thank
you
all
right,
so
this
is
octant.
Let
me
just
give
a
brief
overview
of
of
kind
of
what
our
mission
and
goal
is
with
octan,
so
we
like
to
think
of
octant
as
a
platform
for
kubernetes
developers
to
enable
them
to
troubleshoot
troubleshoot
debug
and
view
their
workloads,
an
analogy
that
I
think
long
term
we're
trying
to
shoot
for
is
kind
of
being
like
the
the
the
integrated
development
environment
for
for
kubernetes.
So
so
you
know
with
our
plugin
system.
E
E
If
you
will
the
there
they're
they're,
not
plugins,
because
they're
they're
baked
into
the
core,
but
they
leverage
all
the
same
api
systems
that
a
plug-in
author
would
use
and
the,
and
so
what
I'm
going
to
show
today
is
kind
of
walking
through
those
defaults
that
we
ship
and
I'll
I'll
also
talk
about
some
of
the
the
plugins
that
are
out
there
in
the
ecosystem
that
I
have
installed
here.
So
with
that
I'll
just
kind
of
get
into
it
and
folks
can
can
stop
me
and
ask
questions
as
I
go.
E
So
this
is
what
you
will
see
when
you
first
start
octant.
This
is
the
what
we
call
the
namespace
overview
octant
will
you
can
provided
a
namespace
at
startup
or
it
will
just
kind
of
pick
one
by
default.
It'll
pick
default
so
I'll
switch
to
that
namespace.
Just
so,
you
can
see
that
there
is
a
default
namespace
right
now.
What
this
is
doing
is
is
going
and
pulling
down
all
of
the
resources
that
are
in
that
namespace.
E
This
namespace
is
not
very
interesting,
which
is
why
I
was
using
the
the
kubeflow
one
for
this
demo
namespace
overview
is,
has
an
opinion
on
the
things
you
want
to
see
for
overview,
so
we
we
show
all
of
the
the
core
kubernetes
resource
types.
Then
we
also
list
your
crds
that
you
have
or
your
custom
resources
necessities
your
custom
resources
that
you
have
installed
in
this
namespace
as
well,
so
and
then.
Finally,
we
list
the
events
that
have
happened
within
that
namespace.
E
So
going
back
to
that
cubeflow
namespace.
We
were
in
before
since
it's
a
little
more
interesting
I'll
start
to
walk
through
a
set
of
resources,
specifically
like
deployments
and
we'll
just
I'll
just
kind
of
walk
through
a
flow
of
looking
at
a
deployment
and
watching
it
watching
it
move
through
the
system,
and
then
we
can
also
look
at
some
some
debugging.
E
E
We've
got
our
card
sizing
kind
of
auto
sizes.
So
you'll
see
this.
This
nice
large
edit
button
here,
but
we
have
the
status
as
as
it's
reported
from
the
api,
any
pods
that
are
associated
with
the
deployment
conditions.
E
This
is
a
report
here.
This
is
actually
coming
from
the
aqua
security
plug-in
that
I
have
installed
so
so
this
is
this
config
audit
reports
section
that
you're
seeing
here
is
not
part
of
of
core
octane
you,
you
would
go
run
this
command
to
generate
a
scan
from
starboard
and
then
that
audit
report
would
show
up
here.
We
also
have
the
the
pod
template
and
any
events
associated
with
this.
Now
we
make
best
effort
to
kind
of
display
this
information
in
real
time.
E
So,
if
I
come
in
here-
and
I
just
say
up
up
the
replicas
of
this
deployment-
you'll
see
that
we
get
a
alert
that
I've
done
so
and
then
at
some
point
the
ui
will
refresh
and
we'll
see
that
these
pods
are
going
into
pending
one
of
the
ways
you
can
also
observe.
This
is
through
what
we
call
the
resource
viewer,
which
you
saw
as
it
just
flipped
to
green.
To.
E
Let
me
know
that
this
deployment
was
okay
and
you
can
also
instantly
kind
of
see
your
last
event,
which
was
that
we
scaled
it
up
to
three
and
you
get
to
kind
of
walk
through
I'll
zoom
out.
So
you
can
see
it
all,
but
you
get
to
walk
through
how
your
resources
are
all
interconnected
with
pods
it's
a
little
different,
just
because
they're
a
grouping,
so
you
can
see
here
are
the
three
individual
pods
that
are
here
and
then
you
also
get
your
your
last
event.
E
It
doesn't
explicitly
call
that
out,
but
that's
where
that
that
comes
from
one
of
the
things
that
is
nice
about
the
resource
viewer
is
when
things
aren't
functioning
correctly.
You
kind
of
get
insight
into
that.
So
if
I
scale
this
down
and
I
go
into
the
resource
view-
you
can
instantly
see
that
my
this
pod
group
right
it
has.
E
It
has
now
shown
these
two
pods
as
being
kind
of
in
a
warning
state,
because
they're
they're
going
away,
and-
and
we
see
that
here-
that
they
are
being
deleted
and
we'll
also
see
the
event
here-
that
there's
that
we've
scaled
down,
we
can
go
into,
for
example
like
the
replica
set
and
once
we
load
the
replica
set,
we'll
also
see
that
same
set
of
pods
that
it's
controlling
and
we'll
see.
E
Here's
the
the
events
specific
to
that
replica
set
the
resource
viewer
for
that
works
in
the
same
way,
except
that
the
replica
set
will
be
the
root
resource,
whereas
before
we
were
looking
at
it
right,
the
pop
the
deployment
was,
was
the
route
and
now
everything's
back
to
being
healthy.
E
We
still
manage
to
show
that
there
are
things
happening,
because
this
is
still
associated
with
it,
but
once
that,
once
things
are
fully
removed
here,
like
this
pod
here
as
it's
being
deleted,
once
that's
gone,
when
we
go
look
at
the
resource
viewer
for
deployment,
we'll
just
see
nothing
there,
but
we'll
be
able
to
get
to
the
event
history
and
and
kind
of
have
an
idea
of
what's
happening.
E
One
of
the
nice
things
about
octet
as
well
is
when
you're
viewing
a
resource
as
it's
being
deleted.
You
get
to
kind
of
see
the
events
stream
through
and
then,
as
it
goes
away,
it
just
kind
of
tells
you
hey
that
thing's
that
thing's
gone.
So
it's
it's
a
nice
way
to
kind
of
watch
things
happen
and
then
now
you
can
see
no
associated
resources
with
this
thing.
E
It
doesn't
know
that
like
because
there's
nothing
here
so
I'll
scale
that
back
up
just
because
I
don't
know
if
someone's
actually
using
it,
and
so
so
all
of
our
resources.
They
have
a
similar
there's,
no
crown
jobs
here
right
now.
If
there
was,
you
would
see
them
and
then
there's
actually
a
you.
Can
pause
them
resume
them
and
and
and
trigger
them
from
the
ui,
but
the
the
overview
page
here
will
actually
list
all
of
these
resources
that
you're
seeing
in
the
left-hand
menu
into
a
single
page.
E
You
can
filter
these
lists.
So
if
you
say
you
only
care
about
failed
pods
right
now,
you
can
go
in
and
select
failed,
or
should
you
only
care
about
stuff,
that's
running,
you
can
don't
need
you
running
et
cetera.
E
So
here
we
have
one,
that's
an
error.
I
actually
have
no
idea
why
it's
an
error,
oh
because
no
replicas
exist,
so
we're
gonna.
I
haven't
looked
at
this
yet
so,
but
I'm
gonna
go
see
if
we
can
use
octan
to
kind
of
help
us
figure
out
why
no
replicas
exist
for
this
thing
and
well
that's
progressive
time
down
yep
and.
E
Interesting,
so
let's
go
drill
down
what's
happening
here,
so
the
service
has
no
endpoint,
that's
different
than
that
error.
Here
we
go.
So
if
we
look
at
our
pod
last
event,
we
can
see
that
it's
attempting
to
set
up
a
volume
for
web
cert
object
and
it
is
not
there
so
so
the
spec
for
for
these
pods,
the
template
for
these
pods
is,
is
incorrectly
configured
or
the
resource
that
it's
expecting
to
be.
E
There
is
not
there,
which
is
this
volume
mount
here,
so
that
that
is
that's
an
example
of
using
octan
to
kind
of
quickly
identify
hey.
What's
the
problem?
Oh
and
you
can
actually
go
into
this
event
and
and
see
the
full
event
here
and
and
go
from
there,
so
so
that
that's
kind
of
the
that's
kind
of
the
quick
overview
of
this
namespace
overview
section.
You
know
each
each
each
resource
is
is
kind
of
called
out
and,
and
you
can
go
through
and
and
our
back
is.
E
We
have
some
things
in
the
backlog
to
improve
it,
but
for
now
you
can.
At
least
you
know,
you
can
see
your
roles,
you
can
go
into
your
roles
and
you
you
can
see
kind
of
what
the
rules
are
for
them.
There's
a
long-term
goal
of
being
able
to
actually
get
like
the
the
the
all
of
the
relate
because
like
if
I
go
here,
I
you
don't
actually
see
what's
using
it
right.
E
So
long-term
goal
is
to
be
able
to
actually
go
and
say,
like
hey,
show
me
everything,
that's
using
this
role
and
all
of
the
permissions
that
are
inclusive
with
it.
So
so,
with
with
that
the
namespace
overview.
Any
questions
before
I
move
on.
I'm
gonna
go
through
these
kind
of
quickly
because
we
don't
have
a
ton
of
time.
So
yeah
I'll
go
through
each
one
and
pause
for
questions,
and
then
I
would
like.
D
To
make
sure
that
you
know
if
we
do
kind
of
run
short
on
on
time,
and
and
we
have
some
some
other
things
at
the
back
end
of
the
hour,
I
do
want
to
make
sure
that
the
plug-ins
and
and
just
the
the
overall
strategy,
what
they
are,
why
cncf
projects
in
particular
might
care.
That's
probably
at
least
from
my
perspective,
one
of
the
more
interesting
things
to
talk
about
for
cncf
projects
that
that
might
want
to
leverage
this
open
source
tool.
E
Yeah,
absolutely
in
fact
I
can
I'll
go
into
that
I'll
cover
cluster
overview
very
quickly
and
then
I'll
go
into
plugins.
So
cluster
overview
is
very
similar
to
namespace
overview,
except
it
is
focused
on
cluster-scoped
resources
so
like,
if
you're
familiar
with
running,
like
cube,
ctl,
api
resources
command
and
and
there's
that
list
of
resources
that
come
back
as
as
cluster
scoped.
E
That's
what
we
try
to
focus
on
here,
as
well
as
some
information
around
nodes
and
and
details
that
are
specific
to
the
to
the
cluster,
and
so
we
do
best
effort
to
try
and
go
and
list
all
of
the
known
custom
resource
definitions
that
you
have
and
and
just
and
show
them
here.
This.
This
list
is
when
you
click
into
the
custom
resource.
E
You
can
also
see
the
versions
we
don't
have
a
diff
between
versions
at
the
moment,
but
that
is
something
that
has
been
requested
since
here
I
have
152
I'll.
Just
extend
this.
To
make
it
a
little
easier
to
see,
you
can
see
that
one's
in
process
of
being
reconciled,
but
yeah
as
we
go
into
these
the
detail.
Views
for
custom
resources
is
not
super
useful.
Yet,
but
it
has
enough
there
to
to
to
kind
of
at
least
give
you
an
idea
of
you
know
the
group
kind.
E
You
know
all
the
names
that
you
would
expect
to
have
on
your
api
resources,
your
api
version
output
and
yeah,
but
I
think
that
they're
again
similar
to
other
things
within
auction,
the
resource
viewer
for
customer
resource
definitions.
E
It's
not
very
useful
right
now,
there's
probably
ways
to
make
it
more
useful,
but
we
just
haven't
got
there
yet
nodes
is
a
is
another
one
that
that
folks
use
and
and
have
requested
to
have
enhancements
for,
but
it
is,
you
know
your
list
of
nodes,
the
information
about
them,
the
set
of
conditions
for
the
various
types
and
are
not
types
but
the
the
status
of
the
various
conditions
and
the
resources.
You
know
your
information
around
cpu
memory,
etc.
So
this
one's
fairly
straightforward.
E
I
think
that's
it
for
notes.
We
do
have
one
thing
I
wanted
to
call
out
because
I
use
it
all.
The
time
actually
is.
If
you
do,
control
enter
in
octant.
You
get
this.
You
get
this
like
quick
switcher,
which
will
let
you
kind
of
fuzzy
switch
between
areas
of
octane,
and
it's
really
really
nice,
because
it
also
does,
for
example,
get
the
plug-in
menu,
which
is
a
good
segue
into
plug-ins.
E
So
everything
that
you
see
in
the
left-hand
menu
is
what
we
would
refer
to
in
octane
as
a
module
and
everything
below
this
cluster
overview
is
a
plug-in.
These
are
either
written
in,
go
or
typescript
and
they
connect
into
octant
through
our
plugin
api
and
what
octant
provides
them
access
to
is
kind
of
the
current
state
of
the
client.
E
So
what
namespace
you're
on
what
filters
you
might
have
selected,
and
also
potentially,
what
resource
you're
you're
looking
at
if
you're
looking
at
a
specific
resource
octant,
will
provide
that
information
to
the
plugin
so
that
it
can
use
that
to
make
cluster
calls
through
our
plugin
api,
to
get
more
information
about
resource
list
resources
and
and
generally
just
like
communicate
using
the
kubernetes
resource
types.
Now,
plugins
can
act
as
modules
as
well.
These
module
plugins,
which
are
all
of
the
ones
you
see
here
on
the
left
hand,
side
are,
they
are
controlling.
E
They
get
the
name
space
as
well
as
the
client
but
they're
controlling
their
their
calls.
So
they're
saying
hey
give
me
a
list
of
this.
For,
for
my
crd,
I
want
to
list
all
of
these.
These
resources,
all
these
objects
out
of
it
and
I'm
going
to
display
them
as
tables
or
you
know,
as
cards,
and
I
think
they're
they're
like
this
is
the
so
this
plug-in
here
is
the
starboard
plug-in
it's
a
security
scanning
tool.
Aqua
security
has
made
this
plug-in
and
it
integrates
with.
E
E
That
is,
there's
there's
a
couple
different
ways
that
plugins
can
send
information
into
octane.
One
of
them
is
freeform
through
their
own
page
and
the
other
one
is
through
what
we
call
the
printer,
the
status
printer
summary
printer
and
what
those
are
is
essentially
the
area
within
octant
that
you
want
to
render
your
component
that
is
determined
by
the
type
of
printer
that
you
use
with
your
plugin.
E
The
plugins
themselves
can
define
we,
we
have
a
set
of
pre,
predetermined
components
that
are
based
off
of
clarity,
ui
and
so
plugins
plugin
authors
can
use
those
components
and
populate
them
with
data
to
render
render
these
displays
briefly.
Folks.
E
D
E
Yeah
clarity,
ui,
is
a
is
a
css
framework
that
also
has
an
angular,
so
octant
on
the
back
end
is
is
built
with
go
as
as
well
as
angular
on
the
front
end,
there's
a
go
runtime
serving
up
json
data
and
that's
rendered
through
an
angular
front-end
piece.
Our
angular
front-end
piece
uses
clarity,
ui,
which
is
a
css
and
layout
library,
that's
open
source
and
our
octane
there's
a
there's,
a
website
which
is
reference.octant.dev.
E
E
Each
of
these
is,
is
a
is
a
plugin
that
was
created
by
a
group
like
so
the
jenkins
x,
plugin
was
was
created
by
the
folks
at
jenkins
x.
There's
a
k-native
plug-in.
I
don't
have
any
k-native
services
installed
right
now,
so
it's
not
particularly
useful,
but
if,
if
I
did
have
them
installed,
you
would
see
your
services
configurations
and
routes
here.
I
actually,
I
don't
think
I
have
any
pipelines
set
up
in
jenkins
x,
right
now
either.
E
So,
but
this
is,
you
know
this
there's
a
demo
of
the
jenkins
plug-in
running,
which
you
can
find
online
I'll,
get
a
link
for
it
and
link
it
in
the
chat
that
shows
kind
of
how
this
operates
and-
and
you
can
follow
your
you-
can
follow
your
deployment
pipelines
as
they
run
through
your
cluster
and
it's
it's
it's
pretty
useful,
and
then
this
is
more
of
a
an
ops
page.
E
That
just
shows
you
all
the
various
stats,
as
your
jobs
are
running
and
then
plugins
here
is,
is
kind
of
the
list
of
of
plugins
that
are
installed
and
then
the
capabilities
that
they
have
registered
with
octane.
This
is
still
very
much
a
dev
view
of
plugins.
It's
it's
very
much
an
early
view
of
like
here's.
E
E
So
and
it's
using
the
tab
printer
or
the
tab
config
for
those
as
well.
So
that's
why
you
see
that's,
why
you'll
see
it
in
the
summary
view
and
the
printer,
as
well
as
a
unique
tab
for
for
that,
and
so
just
to
call
those
out
in
the
namespace
overview,
because
I
think
it's
helpful.
E
E
I'm
running
an
old
version
of
the
plugin
with
a
new
version
of
octant,
so
that
is
my
fault.
It's
not
the
starboard
folks
fault,
so
that
that
plug-in
tab
is
from
starboard
and
would
show
more
data
and,
as
I
mentioned
before,
this
config
audits
right.
So
this
is
this
is
the
printer?
That's
that's,
printing
to
the
summary
page,
and
this
is
the
tab.
That's
registering
a
new
tab
for
the
plugin
and
then
this
is
the
module
that
the
plugin
has
said.
I
want
to
create.
E
D
Cool
I'm
going
to
leave
the
witnesses
in
the
interest
of
time
might
also
want
to
just
show
people
briefly
like
the
logs
and
sort
of
debugging
workflow,
but
I
do
want
to
make
sure
we
leave
some
time
for
q.
A
if
folks
have
questions
just
jump
in
don't
be,
don't
be
shy.
I
spent
a
little
time
with
wynn
a
week
or
two
ago,
and
I
can
attest
that
he's
not
fragile
and
and
yeah.
E
Yeah
so
quickly,
we'll
show
so
we
have
logs,
which
you
can
you
can
filter
down.
So
you
say
you
know
we
defaulted
to
five
minutes
because
as
a
debug
tool,
we
kind
of
we.
We
feel
that
like
and
I
for
me
personally
and
other
folks
that
we've
talked
to
like
the
likelihood
that
that,
especially
with
your
like
outer
loop,
local
development,
you
want
to
see
the
last
five
minutes.
E
E
You
know
configurable,
but
it
does
default
to
that
five
minute
window
and
then
we
also
defaulted
to
just
grouping
all
the
container
logs
together
this
we
found
just
more
useful
because
we
do
provide
filtering
and
we
do
put
the
container
as
a
prefix
to
all
the
log
entries.
So
this
way
you
don't
have
to
go
individually
spelunking
through
all
of
the
containers
to
get
that
aggregated
view,
which
is,
is
pretty
helpful.
E
We
also
provide
a
terminal
which,
if,
if
there
is
a
shell
provided
it
it
will
or
if
the
what
I'm
saying
is
if
the
image
has
some
form
of
shell
installed,
so
sh
bash
for
windows,
containers
command
or
powershell,
whatever
I
don't
know
the
exact
ones
for
windows
containers,
but
it's
there.
But
this
is
the
interactive
terminal
that
you
can
run.
I
don't
know
if
there's
anything
you
yeah,
so
the
the
it
it's
it's
full
compliant.
So
it'll,
it'll
it'll
run
things
like
vim.
E
E
Sam
fu,
one
of
the
members
of
the
team,
has
been
working
on
a
a
kind
of
a
neat
debugging
sidecar
terminal
attachment
thing
that
we're
hoping
to
have
here
in
the
future,
where,
in
the
case
where
your
container
itself
doesn't
support,
direct
like
interactive
shell,
because
it
doesn't
have
a
shell
installed,
you
can
attach
a
side
car
to
it
with
a
shell
and
and
drop
into
a
debugging
mode
on
that
on
that
container,
for
the
for
the
containers
that
that
support
it.
E
So
it's
it's
pretty
cool
we're
hoping
to
to
have
that
pretty
soon
or
sometime
soon,
not
this
next
release,
but
sometime
and
that's
yeah
that
I
think
that
covers
the
oh
there's
also
one
other
thing:
that's
pretty
neat,
which
is
right
from
within.
Let
me
see
if
I
find
something
that
you
know
it'll
be
easier
just
to
go
to
services,
see
if
there's
one
in
here.
E
So
there
is
a
port
forward
button
which,
when
you,
you
can
start
a
port
forward-
and
you
will
get
this-
I
you
can't
see
it.
It
opens
in
a
browser
but
yeah,
and
it's
just
showing
me
an
error
page,
because
this
end
point
is
just
a
cache
endpoint
that
doesn't
really
serve
anything
useful,
but
that
is
a
local
host
port
forward.
That
is,
is
opening
up.
You
know
59
901
locally
to
that
443
endpoint
in
my
cluster.
E
It's
a
pretty
pretty
useful.
It
works
for
services.
It
works
for
anything
that
has
a
a
set
of
discoverable
endpoints
on
it.
You
can
you
can
port
forward
to
so
you'll
you'll,
see
it
from
like
within
pods
you'll,
see
it
on
certain
deployments.
You'll
see
it
within
services
that
that
you
can.
You
can
activate
a
port
forward
so,
and
that
is
a
very
useful
thing
for
quick,
debugging
and
testing.
E
Where
you,
you
know,
you
imagine
you're
you're
in
your
you're
in
your
container
that
you're
trying
to
debug
it's
not
it's
running,
but
it's
not
working
quite
right.
You're,
looking
at
logs
you're
in
a
terminal
in
the
terminal,
you're
messing
around
with
with
relaunching
your
your
server
right,
whatever
g
unicorn
say,
and
then
you
have
your
port
forward,
set
up
right
and
then
all
of
a
sudden,
you're
refreshing
that
page
and
it
starts
to
work
you're
like
oh.
E
E
Using
and
one
other
thing
we
do
we're
we're
starting
to
work
on
on
this.
We
we've
had
this
kind
of
in
flight
for
a
while.
It's
very
you,
know,
poc
proof
of
concept,
but
this
is
kind
of
we're
trying
to
create
this
idea
of
like
a
workloads
or
an
applications
and
a
quick
view
dashboard.
E
So
you
come
into
your
namespace
and
what
you
see
is
is
all
of
your
workloads
and
if
there's
metrics
or
stats
for
them,
you
you
kind
of
get
a
quick
view
of
them
and
within
these
cards,
ideally
we
would.
We
would
have
some
form
of
like
what
we
would
call
like
an
object
status
or
a
score
or
tally
of
like
hey,
what's
kind
of
a
running
health
of
this
thing
over
the
last
you
know
small
window
of
time.
E
You
know
five
minutes,
ten
minutes
something
something
reasonable
for
an
in-memory
time
series
for
octan
to
actually
pay
attention
to
and
care
about,
and
then,
when
you
click
into
these,
it's
kind
of
you
know
that
hybrid
view
of
hey
that
application
card,
as
well
as
the
resource
viewer
that
we
explored
previously,
where
I
can
now
see
hey.
E
Here's
my
it's
my
last
set
of
events
and
and
here's
here's
all
of
the
things
that
are
connected
to
this,
and
this
is
very
much
a
work
in
progress
and
we
would
love
any
feedback
or
thoughts,
suggestions
about
what
might
be
useful
to
to
have
in
an
applications
type
view
like
that
and
for
rendering
applications.
We're
working
with
the
kubernetes
apps.kubernetes.io
set
of
recommended
application.
Labels
is
is
where
we
start,
and
then
we
also
do
like
owner
and
parent
ref
as
a
way
to
to
further
connect
things
that
might
be.
E
Related
pr
is
absolutely
welcome,
so
the
question
from
matt
young
matt
young
was
how
can
folks
engage
with
the
project?
Pr's
welcome
yes,
prs
issues.
We
have
a
github
discussions
as
well
open
to
feedback
prs
any
in
any
form
plugins.
If
you're
interested
in
crafting
or
creating
plugins,
you
can
come
into
the
octane
channel
on
the
kubernetes
slack
and
and
ask
about
you,
know,
plug-in
authoring
and
committing
to
upstream.
All
of
those
things
yes
would
love
to
have
folks
get
involved.
D
Cool
yeah
questions.
G
Hi,
it's
olga,
I
I
just
joined
and
I
was
wondering
how
does
the
octant
compare
to,
for
example,
canines,
or
maybe
there
are
other
solutions
like
that?
What's
what's
the
benefits,
is
it
like
different
kind
of
gui
or
is
it
or
plugins
else.
E
E
You
know
kubernetes
view
anyway,
yeah
it's
it's
mostly
our
plug-in
model
that
that
makes
us
different
and
also
we're
really
trying
to
focus
on
that
developer.
Target
of
of
like
early
like
early
outer
loop
early
inner
loop
like
real,
like
less
because
we
we
have
a
lot
of
actions
in
here
like
you
can
delete
things
you
can
edit
the
yaml
right
in
place.
E
You
can
do
a
lot
of
things
that
you
don't
want
to
do
in
a
production
cluster
like
when
you're
deployed
to
production
with
octant
or
when
you're
deployed
to
production.
There's
a
lot
of
activities
that
you
can
do
with
octane
that
you
might
not
want
to
do
which
can
be
solved
by
just
scoping
down
the
config
but
yeah
our
our.
E
E
So
so
we
tend
to
stray
away
from
some
of
the
more
op
c
type
things
and
and
ops
views
that
that
a
lot
of
other,
like
kubernetes
viewers,
have
built
in
just
because
they
kind
of
distract
the
they're
they're
distracting
a
small
team
from
the
focus
of
making
this
thing
as
good
as
possible
for
developers.
F
Yeah,
thank
you
olga
for
the
question.
I
will
add
that
also
you
know,
being
myself
a
kubernetes
administrator
some
time
ago
often
has
something
different,
because
it's
aiming
to
to
approach
the
troubleshooting
experience.
For
example,
debugging
experience
you
know
showing
more
than
relationships
between
objects
in
a
kubernetes
environment.
You
know
they're
in
in
coordinates
cluster.
There
are
a
lot
of
objects,
resources,
controllers,
etc.
F
So
the
the
goal
of
often
is
to
to
accelerate
developer
experiences
like,
for
example,
troubleshooting
by
making
out
a
view.
You
know,
look
graphical
sense
on
the
relationships
and
everything
inside
coordinates.
D
So
I
got
a
question
privately
in
slack,
you
know,
is
it
a
local
desktop
app
or
you
know,
I
think
when
a
lot
of
people
saw
this
last,
it
was
probably
something
running
on
localhost
that
you
run
as
a
local
angular
thing.
Could
you
could
you
briefly
just
give
a
description
of
like
you
know
what
is
octan?
E
Yeah,
so
octan
is
a
application
there.
There
is
a
web
server
component
to
it
that
is
written
in
go,
but
that
is
just
there
to
be
consumed
by
an
electron
front
end
it.
It
is
an
application
with
a
long-term
goal
of
being
being
able
to
leverage
some
of
that
tight,
desktop
integration
through
allowing
you
to
set
up.
E
If
you
want
to
be
alerted-
or
you
know,
imagine
you're
you're
doing
some
things
within
your
cluster
through,
say
the
aws
console
or
something
like
that,
and
you
have
octane
running
minimized.
If
I
minimize
this,
it
goes
down
into
the
tray,
so
you
octan's
running
down
in
the
tray
and
then
something
happens,
something
you
know
you
apply
some
new
yaml
and
it
breaks
or
a
pr
emerges
and
and
and
your
your
dev
cluster
has
been
updated
and
now
a
deployment
fails,
and
you
see
the
little.
E
You
know
toast
pop
up,
saying:
hey
this
and
you
click
on
it
and
now
you're
you're
staring
right
at
your
deployment
that
had
just
failed
due
to
the
pr
that
you
had
submitted
again.
That's
very
much
that
that
local
development
developer
centric
kind
of
model
that
we're
that
we're
pushing
for,
and
I
think
that's
why.
The
analogy
to
you
know
kind
of
being
the
the
vs
code
for
for
kubernetes
local
is,
is
or
kubernetes
developers
running
locally
is
what
is
what
we
are
hoping
to
achieve.
G
One
more
question:
so
how
would
you
kind
of
estimate
or
evaluate
how
long
would
it
take
for
a
new
person,
a
new
developer,
who
never
worked
with
octant?
How
long
will
it
take
to
get
up
to
speed?
Is
it
just
like
a
matter
of
installing
it
and
then
going
into
the
cluster
or
to
really
use
it?
You
need
to
figure
out
which
plugins
to
use
etc,
like
configure
dashboards.
E
Yeah,
that's
a
great
question,
so
the
everything
that
I've
shown
ex
minus
the
stuff-
that's
in
starboard
and
jenkins,
xk
native
on
on
the
plugin
menus
on
the
left
ships
with
octet
by
default.
When
you
install
octant-
and
you
run-
you
run
it-
it
will
do
best
effort
to
find
your
cube
config.
E
It
looks
it
actually
when
you
start
it
up
it,
it
tells
you
hey.
This
is
where
I
found
your
your
current
path.
Now,
if
it
can't
find
a
valid
cube,
config
file
it'll,
actually
prompt
you
to
paste
one
in
we're
we're
working.
E
I
I
don't
recall
if
we
I
should
know
this,
but
I
think
it
will
prompt
you
to
paste
one
in
or
allow
you
to
do.
The
your
local
file
picker
another
benefit
of
us
being
an
application
local
file
picker
to
go
to
the
cube
config
that
you
want
to
use
and
then,
once
you
provide
it
with
a
cube
config,
it
will
attempt
to
connect
to
that
context.
E
To
that
cluster,
and-
and
you
know,
if
you
have
the
permissions,
it
will
auto
auto-
discover
your
namespaces
and
start
to
start
up
your
informers
so
that
it
can
populate
caches
to
show
you
your
resources
and
and
provide
you
updates.
All
of
that
just
kind
of
works.
You
don't
have
to
configure
or
set
up
anything
as
long
as
you
have
the
kind
of
the
the
the
way
we
say
is
if,
if
cube
ctl
works,
octan
should.
F
F
You
know
time
it
takes
for
a
new
developer
to
to
get
up
to
speed
so
yeah
compared
probably
with
other
options.
It's
it's
quite
fast
and
yeah.
You
have
a
an
option
there.
So
I'll
inform
you.
Where
is
located
the
keep
config
file
and
you
just
need
to
feed
that
provide
the
keep
config
file.
That's
it,
but
but
it's
still
a
work
in
progress
to
improve
that
even.
F
D
Great
as
we're
running
on
time,
are
there
any
other
questions
or
is
there
anything
you
want
to
highlight
wayne
or
other
members
of
the
auction
team
about
sort
of
the
the
project's
positioning
in
the
open
source
landscape?
Or
you
know?
What
might
you
say
to
you
know
the
hundreds
of
projects
in
the
cncf
that
that
may
or
may
not
be
interested
in
in
plug-ins
as
a.
E
Yeah,
absolutely
so
I
would
say:
well
we're
you
know,
apache
2
license,
so
that
makes
it
easy.
E
We
also
are,
if
you
are
interested
in,
if
you
are
a
project
that
has
some
crds
that
go
along
with
your
project
and
you
want
to
have
a
ui
for
those
objects
that
ultimately
get
created
from
those
crds
and
you
don't
want
to
have
to
have
it-
have
a
ui
server
deployed
and
write
a
ui
and
write,
typescript
or
javascript,
and
and
and
do
all
of
that-
and
you
just
want
to
make
it
convenient
for
people
to
get
information
about
your
objects,
interact
with
them
and
and
and
surface
useful
information
about
it.
E
E
We
we
kind
of
we
put
you
in
a
box
essentially
and
say,
like
here's,
here's
what
you
can
do
and
it's
and
you
can
pick
the
typescript
avenue
or
you
can
pick
the
go
avenue
and
and
and
here
are
your
components
and
and
here's
your
resources
from
the
server
right
like
it's,
a
very
nice
experience
for
people
who
just
want
to
write
go.
E
D
Awesome,
thank
you
so
much
for
for
joining
us
and
doing
a
show-and-tell.
This
will
be
on
youtube
later
in
our
in
our
youtube
channel.
I
noticed
we.
I
think
we
have
some
new
faces
here,
that
we've
not
seen
before.
A
Quick
one
just
to
make
in
case
I
missed
it
behind
this
project.
It's
just.
Is
it
just
vmware
tanzo?
Is
there
any
other
involved
parties
behind
supporting
this
project.
E
So
right
now,
primarily
upstream's
all
vmware
employees.
There
are
external
contributors
who
have
contributed
in
the
past
and
we
would
love
to
get
more
external
folks
contributing
to
the
project
and
make
it
not
siloed
by
vmware.
That
would
be
really
great.
E
A
That'd
be
important
to
encourage
also
the
collaboration,
I
guess,
with
the
cncf
ecosystem,
to
make
sure
that
it's
open
for
for
contributions
and
for
additional
maintainers.
Thank.
E
You
yeah,
absolutely,
we
have
it's
documented,
both
our
process
to
becoming
a
contributor
like
becoming
an
octane
owner
and
being
part
of
like
essentially
getting
the
commitment
is
all
documented,
and
it
follows
some
of
the
guidance
that's
provided
just
as
in
kubernetes
upstream.
D
No,
no,
I
again,
I'm
happy
if
folks
just
want
to
discuss,
but
I
do
want
to
just
provide
a
brief
opportunity
for
folks
that
have
not
been
to
one
of
our
meetings
before
to
introduce
yourself,
optionally
and
and
also
in
the
document.
The
google
doc
feel
free
to
put
your
contact
info
there
and
sign
in
as
an
attendee.
G
Yeah,
I
don't
know
I
mentioned
myself,
I'm
all
the
couple
I'm
working
for
adobe
is
focused
on
observability,
so
I'm
trying
to
look
into
everything
related
to
observability.
G
My
big
focus
is
to
look
into
external
developer
observability,
which
is,
I
think,
a
little
bit
more
rare
than
just
observability,
which
is
usually
focused
on
the
internal
teams
who
observe
their
applications.
I'm
trying
to
understand
how
we
can
provide
observability
capabilities
for
external
developers
who
build
products
based
on
adobe.
B
Solutions
and
hi,
my
name
is
avi
press,
I'm
the
founder
and
ceo
of
a
startup
called
scarf,
and
we
make
tools
to
help
open
source.
Maintainers
have
observability
about
their
containers
and
how
they,
you
know,
distribute
them
via
different
container
registries,
and
they
can
actually
understand
like
how
those
containers
are
getting
pulled
down,
and
you
know
like
what
cloud
environments
and
companies
are
using
them.
So
it's
a
little
bit
of
a.
G
G
D
Welcome
thanks
for
joining
us
richie.
Is
there
anything
else
you
want
to
cover
today.
C
No,
I
mean
there's
if
dalton
and
I
did
did
a
podcast
thing
today-
that's
linked
in
the
in
meeting
docs.
I
think
the
question
will
just
move
to
next
time.
This
will.
This
would
be
the
second
now
the
third
time
I
think
in
our
history
that
we
actually
stop
on
on
the
10
before
we
can
also
run
over,
but
I
think
we
should
actually
try
and
not
to
actually
give
people
their
toilet
breaks.
B
F
It's
what
benefits
could
prometheus
operator
get
from
donating
the
project.
D
Yeah,
do
you
want
to
give
a
brief
overview
for
those
that
might
not
already
be
in
our
slack
or
might
not
know
I
I
know
what
you're
talking
about,
but
do
you
want
to
just
describe
sort
of
the
the
set
set
the
table
for
discussion
and
slack?
If
you
will
arthur.
F
We
are
creative
creator
team
is
discussing
if
we
want
to
donate
the
project
or
not,
and
we
have
a
lot
of
doubts.
A
lot.
C
So
as
to
specific
benefits,
once
you
are
in
incubation
state
stage-
and
that
is
obviously
quite
some
time
after
the
sandbox
stage,
you
get
pr
support
by
cncf,
which
means
I
think,
only
for
graduated.
You
can
actually
have
like
like
a
webinar
every
year
or
some
such
you
get
on
the
maintenance
track
of
kubecon.
You
get
one
certain
slot
for
incubated
and
I
think
two
with
101
and
deep
dive
for
all
graduated
projects,
and
you
get
you
get
process
support
like.
C
If
you
have
your
call,
you
can
put
this
on
the
cncf
calendar
you
can.
You
can
get
something
on
youtube
under
the
cncf
umbrella.
C
Those
are
the
most
direct
ones
like
there's,
obviously
fame
and
fortune.
It
might
make
sense
to
now
I'm
switching
heads.
You
could
also
consider
putting
this
under
promises
or
promises
community,
but
if
not
no
worries
back
to
the
other
head.
Those
are
the
main.
C
D
I'll
follow
up
in
slack
as
well.
You
might
want
to
reach
out
to
cheryl
who's
vp
of
ecosystem
at
the
cncf.
There
are
some
decks
that
I
could,
I
think,
there's
probably
more
updated
versions
from
our
mind.
The
the
the
link
I
have
is
about
three
quarters
of
a
year
old,
but
there
is
some
materials
and
collateral
that
that
goes
into
in
in-depth
what
it
means
to
join,
why
you
would
want
to
join
and
there's
a
whole
arm
of
the
cncf.
D
That's
aimed
squarely
at
answering
that
question
and
that
you
can
engage
with
so
I'll
follow
up
in
slack
with
some
links.
If
you
like.
D
D
All
right,
well
yeah,
so
thanks
again
everybody
we
have
all
manner
of
business.
We
could
go
through
in
github,
but
I
think
that'll
be
for
the
next
meeting.
There's!
No,
I
I
think
we're
plus
four
minutes.
So
sorry,
richie
we're
not
gonna
end
right
at
50.
yeah.
I
mean,
if
there's
nothing
else.
Thank
you.
Everyone.
D
And
and
thank
you
to
the
octane
team
for
joining
us
and
giving
a
demo
we'll
post
this
to
our
youtube
channel
later
on
today,
I
will
get
into
doing
that.
I've
not
always
been
super
timely,
but
we'll
we'll
make
sure
this
is
posted
and
thank
you
again.