►
From YouTube: Meshery Development Meeting (Dec 16th, 2020)
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
A
A
A
A
A
A
No
okay:
let's
send
a
link
to
the
meeting
minutes
in
there.
There's
a
couple
of
us
that
have
just
joined
and-
and
let's
start
real
briefly
is-
is
anyone
on
the
call
for
the
first
time
or
anyone
that
hasn't
had
a
chance
to
introduce
themselves
on
this
call
before.
A
Some
of
you
might
have
introduced
on
other
calls,
but
it's
good
to
do
it
on
the
different
calls,
because
not
every
different
people
come
to
different
ones.
So
so
I
guess
yeah,
I
think
everybody's
been
for
the
most
part
introduced
and
those
that
might
not
have
been
a
few
of
you
are
going
to
talk
about
your
work
today.
So
so
you
will
be
all
right.
A
I
think
we've
got
a
decent
agenda
lined
up
if
you,
if
anyone
has
any
topics,
please
put
them
in
actually
husseina
is
here,
which
means
we
need
to
talk
about
eks
a
little
bit.
A
And
I'm
hoping
that
markin
will
join
as
well.
A
Hussaina
would
would
you
mind
messaging
markin
in
slack
just
to
ask
him
if
he
might
be
able
to
join.
He
had
had
some
great
suggestions
and,
for
my
part,
I
ran
out
of
time
to
follow
up
on
them.
A
Okay,
thanks
by
the
way,
just
hussein
as
you
go
to
do
that
this
is
a
good
note
for
for
everyone.
I
don't
know
if
everyone's
aware,
but
you
there's
some
convenience
links
to
each
of
the
meetings
like
to
each
of
the
zooms
and
they're
under
like
meet
dot
layer.
Five
that
I
o.
So
in
this
case
it's
meet
dot
layer,
five
dot,
io,
slash
meshery!
A
A
So
just
fyi.
A
All
right,
first
topic
up
actually
maxine
hi,
hey
we've,
so
abhishek
has
just
joined
and
I
think
we
can.
You
know
let's
talk
about.
Let's
talk
about
provisioning,
istio.
D
A
Yeah,
let's
talk
about
your
your
item.
You
want
to
take
us
through
the
issue
that
you
recently
bumped
into.
D
A
D
Okay
yeah,
so
I'm
right
here
in
the
in
the
dashboard,
and
so
everything
is
okay,
the
mini
cube
is,
I
can
ping
everything
and
I'm
actually
following
the
workshop
to
to
set
up
his
tio
and,
for
example,
first
of
all,
I
don't
even
have
the
same
menu
as
shown
into
the
into
the
the
tutorial
like
when
I
click
on
the
plus
I
don't
have.
I
don't
have
the
same
options,
so
I
can
still
ping
these
to
adapters.
D
A
And
if
you
click
on
it,
that's
a
good
human,
readable
message,
that's
good,
and
then
we
get
some
additional
details.
Unfortunately,
we
don't
get
an
error
code,
which
would
be
really
nice
to
look
up
like
specifically
a
measuring
specific
error
code.
That
then
tells
you
what
you
should
probably
do
to
fix
it.
A
D
But
here's
where
you
can
ping,
because
I
can
I
can't
ping
everything,
but
so
I'm
supposed
to
to
connect
to.
A
Are
you
running
mini
cube,
yeah?
Oh
yeah,
yeah.
Sorry,
it
says
it
right
in
front
of
my
face.
I
wonder
if
it
isn't
the
measuring
adapter
that's
having
a
challenge
speaking
to
mini
q,
as
opposed
like
that
specific.
D
Doesn't
seem
to
be
just
histo,
because
I
think
if
I
try
to
run
any
action
from
any
adapter,
I
think
it
will
have
exactly
the
same.
Yeah.
A
Which
I
actually
oh
good
mark
martin
is
on
which
so
I'm
excited
about
that,
because
mark-
and
I
don't
mean
this-
is
very
relevant
to
your
discussion,
maxime
so
mark
I
wanted
to
apologize.
Actually,
we
had
a
conversation
like
really
early
my
morning
and
I
never
got
back
to
finish
our
conversation,
so
so
I
just
wanted
to
apologize,
but
glad
that
marquin
is
here,
because
I
think
he's
about
to
answer
your
question
or
try
to
point
out.
A
D
D
A
D
And
where
shall
I
put
this
in
your
terminal.
A
In
your
power
in
your
not
powershell,
I
guess
but
your
well,
no
yeah
powershell.
D
A
A
I'm
sorry
in
the
measuring
ui.
D
D
A
A
D
A
This
is
that
marking
was
the
mvp
of
our
last
mastery
dev
call,
and
we
can't
we
can't
have
him
being
the
mvp.
Two
two
calls
in
a
row
two
weeks
in
a
row.
So
that's
we
need
to
move
on.
D
D
A
Great
and
then
now
I
could
one
last
thing
that
markin
was
asking
to
potentially
do
just.
I
think
for
additional
info
is
like
when
you're
in-
and
you
might
have
done
this
before,
but
when
you're
in
your
terminal,
your
powershell
terminal,
if
you
run,
I
think
you
did,
but
if
you,
if
you
do
it
now
that
you're
successful,
if
you
run
mescheri
ctl
system
logs.
A
Things
should
look
a
this
is
really
like.
Some
of
this
stuff
looks
like
issues
and
because,
and
some
of
us
just
debug.
A
A
I
don't
know
anyway,
you're
you're
you're
good
to
continue
with
the
release
to
your
work
yeah
and
should
be
now
with
all
of
them.
So
good.
There
is
a
follow-up
to
this
topic.
This
mini
cube
topic
so
but
maxime
thanks
for
letting
us
walk
through
this.
This
is.
E
A
E
Heidi,
but
I'm
just
wondering
if
we
shouldn't
change
just
anything
during
startup
of
mercury
or
something
else
just
to
do
this,
connecting
in
the
background
under
the
hood
just
because
so
many
so
many
users
are
just
trying
to
overcome
rest-
and
this
is
overwhelming
then
so
so
the
one
of
the
idea
is
just
to
add
it
to
the
measuring
ctl
system
start
for
example.
Yes,
I
don't
know
if
this
is
the
the
best,
but
this
is
one
of
the
possibility,
because
this
connection
is
needed
every
time
after
every
measuring
startup.
E
Yes,
if
you
stop
them
as
restart
it
again,
the
connection
to
the
mini
cube
network
bridge
will
be
required
once
again,
because
the
the
the
measuring
container
container
will
will
lose
this
connection.
Yes,
so
this
this
is
needed
after
each
start,
so
I
think,
connecting
to
the
mercuricity
of
system
start
just
adding
this
one
command
to
this,
and
the
same
issue
as
in
with
mini
cube,
is
with
kind.
Yes,
if
you
install
the
kind
the
and
do
not
install
the
measure
of
your
home,
but
the
docker,
you
will
just
get
exactly
the
same
issue.
E
A
Yeah
yeah,
I
think
that
yeah
it's,
which
is
why,
both
in
the
on
principle
of
what
you're
saying
is
hey
like
this
is
this
is
this
is
a
fairly
significant
hindrance
to
you
know
in
a
number
of
use
cases,
and
it
like
very
clearly
needs
to
be
addressed.
It's
not
only
needs
to
be
addressed
just
from
a
like
for
any
project,
but
particularly
for
this
project.
We
want
to
hopefully
as
a
design
principle
and
something
of
a
point
of
pride
for
the
project
is
to
make
sure
that
it's
brings
a
great
user
experience.
A
Just
you
know
makes
things
it
has
the
nice
default
assumptions
and
helps
part
of
the
reason
that
people
are
going
to
use.
It
is
because
networking
is
confusing
and
hard,
and
if
we're
not
helping
overcome
that,
then
then
we're
not
help
really
helping
solve
the
problem
in
general
that
people
are
having
so
the
I
don't
off
the
top
of
my
head.
A
I
think
my
so
my
from
my
perspective.
It's
yes
and
then
the
the
particulars,
and
it's
yes
like
it's,
not
a
one-time
or
the
the
particular
approach
that
we
might
take,
is
a
recurrent
like
it's
a
recurring
fix
that
kind
of
needs
to
happen
like
you
were
saying,
sort
of
on
start
at
which,
like
it
it
bugs
me
it.
A
Because
it
actually
to
your
point,
you'd
said:
like
hey:
we've
got
a
mini
cube
issue.
We've
got
a
kind
issue,
there's
probably
a
couple
of
other
environments.
We
have
issues,
there's
there's
another
one
where
when
people
want
to
run
a
performance
test
and
they
want
to
give
measuri
the
end
point
like
the
url
to
one
of
their
services.
A
Well,
sometimes
those
those
services
are,
people
want
to
give
them
that
service,
by
name
or
by
ip
address
like
kubernetes
ip
address
and
meschery.
Since
it's
in
its
own
private
ip
address
range
network.
It
doesn't
you
know,
that's
another
example
of
where
it
doesn't
know
how
to
talk
or
how
to
reach
this
other
network.
A
Even
if
it's
not
meaning,
cube
or
kind
or,
and
so
it'd
be
nice.
If
yeah
to
to
step
away
and
come
back
with
well
and-
and
I
don't
I
don't
have-
I
don't-
have
a
good
suggestion-
I
might
toss
out
some
and
and
then
they're
not
going
to
be
great
and
it's
going
to
be
something
like
well.
A
A
A
Ctl
only
allows
you
to
deploy
to
docker
today
tomorrow
it
should
be
supporting
kubernetes
like
we
should
be
doing
that
fairly
quickly,
but
the
the
problem
will
persist,
and
so
one
of
the
things
I
think
that
would
even
if
the
only
approach
is
look,
that's
the
only
thing
that
we
can
figure
out
is
like
it's
best
for
meshri
as
an
app
to
sit
in
a
bridge
network,
and
so
it's
bet,
and
the
only
thing
to
do
is
like
individually
connect.
These
containers
and
you've
got
to
do
it
every
single
time.
A
Either,
to
unintelligently
like
in
a
greedy
way,
connect
mescheri's
bridge
network
to
other
networks
that
look
interesting
ones
that
are
called
mini,
cube
ones
that
are
called
kind
or
or
is
to
make
mastery
ctl
a
little
more
intelligent
and
have
it
understand
more
about
the
environment
like
what
type
of
mini
cube?
Is
there
what
the.
E
E
Okay,
so
I
think
the
quite
quite
easy
idea
is
to
connect
a
measuring
containers
to
each
bridge
network
on
the
hosts,
so
we
can
filter
via
driver
type
if
it's
breach
connect
to
all
name
of
this
bridge,
and
it
will.
This
is
the
grivy
approaches
the
degree.
So
this
is
very
simple
idea.
Yes,
and
it
will
work
such
approach,
which
will
work
with
kind
with
mini
cube
with
some
because
mini
it's
also
the
network
name.
This
is
the
default
name
on
the
of
the
on
the
cluster.
E
The
kind
will
be
default
on
the
kind
cluster
if
anyone
will
name
the
cluster
other.
The
network
name
can
be
also
changed.
So
if
we
filter
the
other
driver
type,
we
are
breach
and
connect
to
the
name
corresponding
to
this
breach.
It
will
always
work.
This
is
not
the
best
solution
for,
but
but
it's
easy,
it's
easy,
it's
it's.
D
A
A
E
E
A
So
I
don't
have
a
better
suggest
yeah.
It
would
be
nice
to
brainstorm
a
little
bit,
there's
a
there's,
a
document
that
is
a
perfect
place
to
to
sort
of
outline
these
thoughts.
A
Even
if
it's
the
we
just
arrive
at
the
exact
same
solution
that
you're
just
suggesting,
but
it's
in
part
just
to
say,
here's
why
these
other
things
won't
work
and
and
to
kind
of
have
that
on
on
paper,
I'll
I'll,
put
the
link
in
the
meeting
minutes
and
chat
and
and
I'll
say
well
maxine.
Thank
you
for
thank
you
for
this.
I
think
we're
ready
to
martin
martin.
If
we're.
A
A
So
so
this
is
the
other
dock.
By
the
way,
this
one's,
probably
one
of
our
earlier
docs
it,
I
think,
even
in
the
the
design
principles,
it
should
probably
talk
about
yeah,
facilitating
getting
mastery
running
faster,
making
sure
that
our
user
experiences
are
cons.
We
have
consistent
that
they're
familiar
so
so
we
should,
you
know,
seek
to
be
inspired
by
other
popular
clis
in
the
space
cube,
ctl
and
docker,
and
but
but
this
talks
about
why
we
use
docker,
compose
and
other
things.
A
Next
up,
sudan.
F
F
To
say
what
are
the
features
being
released,
the
bug
fixes
and
the
maintenance,
particularly
so
I
have
been
working
to
create
for
each
release.
I
I
have
been
working
to
create
a
change
log
that
will
create
a
file
that
in
under
misery
docs,
so
it
will
be
a
automatically
updated
currently
in
progress.
A
Nice
and
very
good
I
figured
we
could
talk
about
the
approach
just
a
little,
because
I
think
there's
there's
two
that
you
can
basically
choose
from.
I
think,
whichever
I
think
you'll
arrive
at
the
same
place
and
whichever
one
is
maybe
easier
to
repeat
the
the
problem
statement
for
everyone
here,
that's
to
say
that
cush
trivedi
had
has
been
working
on
continuous
integration
for
the
projects
for
quite
some
long
time
and
part
of
the
each
of
the
40-something
repositories
that
are
here.
A
They
for
the
most
part
follow
a
fairly
similar
set
of
content.
You
know
github
actions
of
of
workflows.
Anyone
that
wants
to
come
to
understand
those.
It's
just
in
any
of
the
repos
you'll
go
to
dot,
github
and
you'll
see
a
couple
of
workflows
either
both
under
this
folder
and
basically
everything
under
here
can,
you
know,
is
of
interest.
A
One
of
those
things
that's
ubiquitous
for
all
of
our
repositories
is
the
use
of
a
bot
called
release
drafter
and
it
every
time
that
a
pr
emerges.
This
release,
drafter
bot
kicks
up
and
it
identifies
it
looks
at
the
type
of
issue
that
was
just
resolved.
The
type
of
merge
that
was
just
done,
it
labels
it
and
it
drafts
out
what
would
ultimately
be
release
notes.
A
So
that's
what's
called
release
drafter.
So
if
I
were
to
cruise
over
to
the
releases
for
measuring
there's
the
current
release,
that's
public
is
04.27,
but
there's
been
prs
that
have
merged
since
in
the
last
six
days,
so
these
prs
have
merged
and
that
release
drafter
bot
is
just
ongoing.
A
Adding
you
know,
building
up
the
release,
notes
the
change
log
and
then
one
of
our
a
step.
That's
perpetually
plagued
us
is
the
lack
of
automation
between
these
change
notes
and
what's
publicly
published
in
this
jekyll
nashery
doc
site
we're
perpetually
behind
on
publishing
the
notes,
because
it's
just
one
of
those
steps
that
gets
forgotten
about
so
so
what's
being
worked
on
here.
A
Is
the
automation
of
that
and
kind
of
two
ways
that
you
might
consider
to
chew
this
off
on
sedan,
which
is
which
is
is
use
of
either
a
jekyll
data
file
or
a
jekyll
collection,
and
you
might
be
familiar
with
these
concepts
already,
but
I'll
touch
on
them.
Briefly,
in
case
you
aren't
and
for
the
benefit
of
others,
are
you
familiar
with
both
of
those
concepts.
A
F
Okay
yeah,
so
it
yeah
it
currently
using
the
release
draft.
We
there
is
no
option
to
create
a
file,
so
green
automatically
has
something
called
yeah.
If
you
can
see,
we
can
generate
the
change
clock
as
we
want.
We
can
point
out
to
a
template
as
released.
G
A
Let
me
sorry
interrupt.
Let
me
change
the.
Let
me
change
the
ask
on
you
if
I
can,
which
is
to
say
the
release
drafter
that
we're
using
now.
Ideally,
we
want
to
try
to
continue
to
use
it,
use
it
to
generate
the
notes
and,
and
is
that
your
approach
are
you?
Are
you
switching
that
out
with
something
else.
F
No,
I'm
not
gonna,
I'm
not
going
to
switch
that.
I'm
like
this
is
for
only
for
getting
the
change
log
to
get
the
pull
request,
like
whatever
the
release
drafter
like
records,
the
pull
request
that's
been
merged
will
be
like
to
get
the
doc
yeah
like
create
amd
file.
I'm
not
I'm
currently
not
changing
anything,
and
I
have
added
few
things
like.
I
have
not
changed
any
functions
of
the
release
structure.
A
Okay,
yeah,
it's
confusing
to
me
the
reason
I
think
that
is
because
is
because
it
looks
like
we're
republishing
or
re-identify.
I
guess
the
th.
My
thought
was
that,
and
it
doesn't
have
to
be
this
way.
A
It's
just
this
I'll
explain
why
I'm
confused,
which
is
if
release
drafter,
is
identifying
the
different
types
of
pr's,
and
you
know,
formatting
the
release,
notes
that,
upon
release
like
when
we
make
a
tagged
release
that
it
would
be
at
that
point,
the
some
automation
some
we
would
need
to
just
you
know,
capture
those
release,
notes
and
save
them
persist
them
to
a
markdown
file,
and
so,
but
of
what
I
see
here
it
looked
like
we,
you
were
kind
of
redoing
what
release
drafter
does
and
so,
but
that's
probably
me
just
being
confused.
F
Yeah,
like
I
wanted
to,
I
won't.
I
didn't
want
to
repeat
the
same
structure
here,
but
I
I
was
trying
to
find
out
a
way
that
we
can
capture
the
what's
been
drafted
in
a
release
draft,
so
the
same
approach
you
were
saying
like
you
were
meaning
to,
and
I
couldn't
find
a
way
that
to
capture
it
and
pass
it
to
a
file
like
create
a
md
file.
F
A
Also,
okay,
so
in
addition,
so
we
would
be
generating
two
change.
Logs
two
release
notes:
two
sets
of
different
release:
notes
right
using
two
different
tools.
I
can't
I
can't
imagine
that
there
isn't
a
utility
that
can
that
you
know,
while
we're
doing
a
release
and
release
drafters,
generating
those
notes
that
it
can't,
in
that
same
workflow,
that
we
can't
pass
along
those
release,
notes
to
a
different
job.
You
know
to
a
different
step
and
have
it
have
something
else
other
than
the
release
drafter
bought
persist
that
to
a
file
seems
weird.
A
F
Like
I
think
you
got
it
me
wrong,
like
we
are
not
generating
two
release
notes
like
currently,
it
generates
only
the
change
lock
here.
Okay,
but
it
has
the
functionality
to
generate
a
release,
dropped.
Also,
release
note
also
the
gen
user
generates
or
creates
a
change
log
file.
That's
like
what
we
have
in
the
current
release.
Notes
like.
If
you
have
a
current
release,
note
it
gets
it
and
it
creates
a
change
log
like
if
I
can
share
the.
A
A
Using
but
okay,
maybe.
A
Maybe
we'll
come
come
back.
I
think
I
have
a
better
understanding
as
to
to
what
you're
saying
if
we
can
avoid
generating
the
release.
No,
what
what
release
drafter
creates
that
would
it
would
just
be
ideal
if
we
can
take
that
and
persist
it
to
file,
and
it
sounds
like
that's
what
you're
you're
more
or
less
trying
to
do,
and
so
yeah,
okay,
nice.
F
A
Like
to
today,
this
doc,
this
page
is
just
one
long
page,
but
that
tomorrow,
you'd
probably
you'd
create
this
such
that
there's.
Each
individual
release
is
its
own
markdown
file
and
you'd
have
so
you
basically
have
a
collection
you
sort
of
iterate
through
and
the
net
result
is
probably
that
there's
still
just
one
really
long
page,
but
but
it's
actually
behind
the
scenes
broken
into
a
bunch
of
individual
markdown
files.
A
Is
that
right,
yeah,
nice?
Okay,
anybody
have
any
comments
on
that
or
feedback
for.
Anyone
worked
with
this
with
gren
before.
A
F
A
Nice
good
well
not
to
not
to
ease
off
all
the
pressure,
but
the
issue
that
and
by
the
way
you
might
want
to
list
the
issue
that
you're
fixing
the
issue
number
here.
But
when
you
do,
when
you
look
at
it,
if
that
issue
was
created
a
year
14
months
ago
or
something
so
we've
been
waiting
for
for
14
months,
we
can
wait
a
few
more
weeks,
cool
cool
thanks
for
thanks
for
doing
that.
That's
great.
A
All
right,
another
quick
topic
kush
had
landed
a
measuring
ctl
feature
called
channel.
Has
anybody
tried
it?
Anybody
tried
this
command.
A
And
another
question:
can
anyone
tell
me
what
this
command
should
do?
What
you
would
expect
that
it
does.
A
A
So,
let's,
let's
see
if
we
do.
A
Does
okay
all
right?
Well,
so
I
pressed
enter
please
subscribe
to
either
the
stable
or
edge
release
channel,
okay,
yeah,
I'm
cool
yeah
switches
between
edge
and
latest,
very
good
beneath
very
good,
okay
good.
So
at
least
a
few
of
you
are
interpreting
the
command
in
the
right
way
like
we're
not
far
off
like
it's.
A
You
know
in
terms
of
its
naming,
just
it's
not
far
off
from
what
it
is
supposed
to
do
so
yep,
that's
right!
It's
supposed
to
help
people
switch
between
the
two
different
release
channels
that
we
have
okay.
So
if
you
just
do
that,
it
it
says.
Well,
hey
look.
You've
got
to
choose.
One
is
what
the
response
is.
Okay,
so
in
my
mind
and
by
the
way,
anirudh
or
or
anyone.
A
A
Yes,
okay,
no!
Okay!
It's
one!
One
item
for
me
just
like,
and
I
I
think
it's
important
for
us
to
walk
through
this,
because
it's
it
helps
us
as
a
community
grow,
because
so
that
others
can
bring
forth
a
consistent
experience.
What
we
would
do
is
this
is
a
good
message.
We
should
probably
also
give
them
the
literal
usage
like
usage
and
give
them
an
example
of
what
they
would
need
to
put
since
we
don't
I'm
going
to
type
dash
h
for
help
and.
A
One
thing
that
we're
doing
we're
just
as
a
side
note
and
is
that
we
have
this
listed
twice
where
the
mastery
ctl
gamma
config,
is
so
side
note
from
the
system
channel.
Is
anyone
willing
to
take
this
on
to
go?
Remove
this
duplicative
listing.
A
Anybody
anybody
anybody
is
anybody.
If
someone
doesn't
sign
up,
I'm
gonna
sign
someone
up.
A
I
wonder:
would
you
mind?
Oh,
very
good?
Okay,
you
were
volunteering
before
I
could.
Thank
you,
hussain
yeah.
We
did
we're
okay,
all
right.
So
then,
back
to
the
user.
Experience
here
about
system
channel
help.
Okay,
so
it
says
all
right,
so
you've
got
two
commands
two
flags
that
you
need
to
use
here:
you're,
either
gonna
use
dash,
dash,
dash,
set
or
dash
dash
switch.
A
That's
weird
like
this
s
kind
of
matches
the
set,
but
this
c
doesn't
match
the
switch
okay.
It's
not
the
most
intuitive
release
channel
to
be
set
for
mesri
and
its
adapters.
A
Okay,
I
don't
know-
maybe
we
would
rather
just
say,
set
the
release
channel
for
messy
and
it's
adapters,
but
anyway
a
little
bit
of
wording,
but
anyway,
okay,
so
I'm
a
user.
I'm
going
through
this
release
channel
to
be
switch
release
channel
to
be
switch
for
my
stream
to
sit
down.
Okay,
that
english
is
a
little
off,
but
this
might
we
might
just
start
with
the
verb
like
set
release
channel
for
you
know
or
and
maybe
instead
of
set-
maybe
we
would
say
subscribe
because
we're
kind
of
subscribing
to
anyway
all
right.
A
So
I
will
switch
with
a
c
to
stable,
successfully
setted.
A
Successfully
setted
channel
okay
we're
going
to
fix
this
this
verb,
but
and
maybe
rather
than
said
it
we
would
say
subscribe.
It
looks
like
as
we're
doing
that
we're
invoking
and
we're
updating
measuring
as
well,
which
makes
sense
if
you're
gonna,
so
by
the
way.
So
so
I'm
kind
of
walking
through
this
for
everybody
to
one
there's
a
couple
of
issues
here.
A
Is
that
we're,
if
you,
if,
if
you
go
to
perform
an
update
of
measuring,
if
you
do
a
mastery
ctl
system
update
in
my
environment,
meshre
server
is
not
is
not
running,
which
I
assume
is
what's
happening
here,
like
the
connections
refused,
and
so
I
couldn't
you
know.
So
it's
not,
or
rather
I'm
sorry.
Docker
is
not
running.
That's
what's
not
running
well,
both
meshes
server
and
docker
are
both
not
running,
and
so
we're
not
doing
a
good
job
of
reusing
existing
feature
functions.
A
So
husseina
had
built
in
a
function
into
measuring
system
measuring
ctl
systems
start
that
would
detect
whether
or
not
docker
was
running,
and
if
it
isn't
it
would.
You
know
capture
that
error
and
it
would
actually
ask
you
if
you
would
like
to
start
docker,
and
so
we
should
be
reusing
as
much
as
possible
as
we're
going
through
this
code
that
we're
using
the
same
functions.
So
we
don't
bump
into
errors
like
this.
A
A
At
some
point,
you're
going
to
start
with
measuring,
get
comfortable
you're
going
to
use
the
stable
release,
channel
and
you're
going
to
be
using
things
and
then
soon
you're
going
to
be
wanting
to
get
to
things
new
features
that
have
yet
to
go
into
the
stable
channel,
so
you're
probably
going
to
want
to
switch
to
the
edge
release,
and
so
this
command
will
be
helpful
to
you
and
the
user.
Experience
might
be
something
for
you
to
comment
on
so.
A
It's
a
it's
an
excellent
command.
It
probably
needs
a
little
fine
tuning,
but
really
helpful
to
quickly
switch
back
and
forth
if
you're
not
aware
every
time
that
a
pull
request
is
merged
into
the
master
branch
in
the
mesherie
repo
and
actually
the
same
is
true
for
every
measure:
adapter
as
well,
all
of
the
adapter
repos.
A
Every
time
a
pr
is
merged,
an
edge
really
the
edge
release
is
updated.
An
edge
release
is
made
available
to
the
world.
Like
you
know,
a
few
minutes
later
after
the
workflow
is
done,
the
stable
release
or
a
new
stable
release
is
made
as
and
when
a
human
goes
over
and
gives
it
a
release
number
and
you
know,
pushes
out
my
release
so,
okay,
so
I
think
there's
a
couple
of
issues
to
be
raised
here,
so
husseina
you've
got
one
on
the
and
the
other
and
hussein.
A
Actually
you
know
considering
that
if
you
don't
mind,
there's
a
second
one
in
here
that
you
might
consider
and
that's
to
ensure
that
we're
invoking
the
same
thing.
Docker
detection.
C
Yeah
the
pre-requisite
function.
A
Right,
okay,
cool
all
right
and
then
next
so
on
a
route.
I
know
that
we're
not
in
a
position
to
to
demo
context,
but
is
this
one
that
you
want
to
talk
to
people
about
about
your
you
know
your
design
and
what
you're
going
to
deliver.
B
I
don't
have
either
now
other
than
the
dog
which
I
created,
so
he
was
speaking
about
it.
People
would
not
have
an
idea,
so
the
that
the
idea
on
or
the
main
focus
here
is
to
have
a
switchable
context
as
how
you
have
in
cubed
ctl.
For,
given
that
this
we
would
have
context
in
terms
of
measuring,
so
that
would
be
switching
between
different
measuring
contexts,
so
the
messy
server
could
be
in
any
place.
B
A
Nice,
so
the
the
link
to
to
to
this,
so
this
document
is
measuring,
cli
commands
and
documentation.
This
is
the
the
meschery
ctl
dock.
If
you
will
it's
the
design
spec
from
s3ctl
and
there's
a
section
at
the
very
bottom
that
anirud
wrote
up
and
it's
it's
been
reviewed,
he's
he's
off
executing
stuff.
If
folks
have
feedback
for
him,
I
think
then
he's
open
to
it.
There
is,
you
know,
it'll,
be
a
very
simple
command
to
start
and
yeah,
and
so
we
should
dina
has
a
great
question.
A
So
we're
going
to
spend
the
rest
of
our
time
on
this
to
talk
about
this,
which
is
which
I
think
is
really
good.
It's
something
that
we
like
a
month
ago.
I
don't
know
how
many
I've
talked
about
this.
I
don't
know
how
many
times-
and
it's
been
my
fault-
that
we've
never
written
it
down
and
onion
root.
Finally,
has
I
think
one
thing
that
would,
I
think,
collectively
we're
trying
to
define
here
is:
are
two
things
two
terms.
A
So
if
you,
if
you've,
been
a
messerie
user,
you've
and
you've
gone
to
the
the
gear
icon
in
the
upper
right
hand
corner
in
the
ui,
you
go
to
the
settings
page
on
the
settings
page,
there's
an
environment,
tab.
The
environment
tab
is
where
you
load
up
your
cube
config.
If
you
have
to
manually
load
your
cube,
config,
that's
what
you're
doing
so
any
route
through
this
work
is
trying
to
help
sort
of
con
construct.
A
couple
of
terms
a
couple
of
constructs
in
the
measuring
project.
A
One
of
those
is
is
called
context
and
this
would
be
a
measuring
ctl
construct.
A
So
this
command
is
working
on
its
mastery
ctl
system
context,
and
it
will
basically
perform
the
action
of
tracking
the
location
of
mesherie
server.
A
It
will
let
you
give
a
name
to
if
you
have
multiple
mesh
reserver
deployments
it'll,
let
you
give
a
name
to
those
deployments
and
switch
between
them.
So,
just
like
you
might
do
cube
ctl
like
just
like
any
route
said
just
your
cube.
Ctl
config
set
content.
It's
not
going
to
be
the
same
syntax,
but
just
the
same
concept.
A
The
reason
for
this.
It's
not
because
there's
people
running
five
different
measuring
servers
and
they
need
to
switch
between,
like
that's,
not
a
common
use
case
right
now.
But
the
reason
for
this
is
because,
if
you
look
through
messy
ctl's
code,
there's
some
hard
coding
in
there
to
this
location,
local
host,
hard
coding,
assuming
that
you've
got
you're
using
measuring
ctl
in
the
command
line
and
you've
got
mesh
reservoir
deployed
locally.
A
A
This
will
allow
us
to
have
a
measuring
s,
a
meshery
server
construct
that
we
have
today
it's
a
very
weak
construct.
It's
called
environment,
and
this
will
be
focused
around
your
kubernetes
cluster,
and
it
will
eventually
then
allow
you
to
switch
between
which
kubernetes
cluster
you're
trying
to
administer,
and
that's
actually
a
very
common
use
case
that
people
have
so
so
adina
had
sort
of
asked
the
question
I
think
the
first
of
which
hopefully
we
answered,
but
the
second
one
is
yeah.
Okay,
right
adina
is
that
does
that
help?
Is
that
and
clarify.
C
So
after
getting
the
context,
are
we
switching
the
context
using
any
other
cli
or
how.
A
B
Simply
create
new
context,
so
I'm
currently,
I
have
three
particular
commands
in
mind
inside
context:
that's
create
delete
and
edit,
so
you
can
create
delete
or
edit
any
of
the
contacts
that
point
to
separate
server
endpoints
and
that
would
actually
free
us
from
localhost
where
actually
messages
everything
hard
coded.
B
Oh
yeah,
that's
that's
the
entire
part
here.
So
you
switching
between
mexico
context
for
now,
like
the
initial
part,
is
simply
switching
between
urls,
where
our
backend
has
been
hosted.
A
Underwood,
would
you
document
this
a
little
bit
further
about
how
it
where
it
is
that
someone
would
give
I'm,
assuming
that?
What
it
would
be
here
is
like
the
name
of
that
context,
and
then.
B
B
Use
this
dock
yeah,
I
just
use
its
introduction
to
context
right
yeah,
so
this
is
actually
a
bit
different
here.
So
I'll
just
update
the
image.
Okay.
A
Cool
okay,
so
maybe
you
win
you'll
you'll
message
in
the
slack
channel,
probably
when
there's
some
of
these
questions.
A
B
Okay,
so
you
have
context,
and
so
this
is
a
yaml
file.
You
have
context,
then
inside
context,
you
have
the
different
context
that
you
would
add
or
delete
right.
So
currently
I
have
local
and
gk
and
yeah
this.
These
are
just
template
urls
for
now
and
then
so,
I'm
currently
in
working
on
it
and
once
it's
done,
maybe
we
can
work
on
taking
this
further
to
actually
get
it
integrated
with
system,
config,
mini
cube
gear
and
all
the
different
commands
that
work
with
different
messages
right.
B
So
this
is
the
initial
way
on
how
context
is
you
know,
starting
to
get
implemented,
I'm
halfway
through
so
next?
I
guess
next
meeting
yes,
yeah.
A
Cool
nice
thanks
for
walking
us
through
this
when
when
this
is
a
little
when
you
when
this
is
a
little
more
written
out
yeah,
if
you
message
in
the
mastery
ctl
channel
folks
can
come
give
comment,
but
it's
looking.
A
It's
looking
good,
it's
nice
and
simple
to
start
with
even
simple
things
like
just
tracking
the
location
of
mesh
reservoir
comes
with
all
of
the
burden
of
like
doing
crud
on
that
you
know
creating
the
entry
deleting
it
updating
the
entry,
so
yeah
is
the
briefly
is
the
edit
that
you're
thinking
of
here
this
would
more
or
less
open
up
would
do
basically
vi
on
the
config
file.
B
A
B
You
could
like
you,
could
change
the
bits
measure
url
url,
but
yeah.
This
is
the
syntax.
B
And
oh
yeah,
so
the
idea
here
is
also
to
pass
in
tokens.
So,
as
you
said,
we
need
auth.
We
need
to
have
authentication
simplified
right,
so
you
just
need
to
pass
in
token
path
here
and
that
would
automatically
save
it
with
the
context
so
the
next
time
you
run
the
command.
It
would
pick
the
token
from
the
context
location.
A
D
A
Well,
hopefully,
this
will
do
not
only
break
us
free
from
some
hard
coding,
but
also
maybe
break
us
free
of
what
one
of
the
things
that
we
were
recently
talking
about.
It
was
the
subject
of
the
of
conversation
was
like
the
horse
before
the
cart
issue,
with
having
to
go
out
and
retrieve
a
token
and
pulling
it
back,
and
I
think
I
think
the
conversation
that
we
were
all
having
was
sort
of
headed
toward
questioning
the
convenience
like
questioning
whether
or
not
that
approach
that
prior
approach
should
continue
forth.
C
That
yeah,
so
this
create
context,
would
require
auth
dot
json
as
a
prerequisite,
and
it
would
be
stored
in
the
context
and
the
context
would
be
reused
again
right.
A
Oh
I'll
answer
real
quick,
yeah
yeah,
that's
right,
it
doesn't
yep,
that's
right.
It
doesn't
th.
This
first
step
doesn't
circumvent
some
of
the
awkwardness
of
having
to
go
over
download
the
token
save
it
somewhere.
There's
so
there's
some
there's
further
steps
on
that.
Ultimately,
hussein.
Actually,
I
think
the
step
that
we
ultimately
want
to
get
to
with
is
to
probably
get
away
from
tokens
and
get
into
well.
I
guess
this
would
still.
A
It
would
still
be
a
token,
but
we
might
use
oauth
as
given
that
we,
you
know
it's
a
long
conversation
because
there's
there's
the
local
provider
there's
remote
providers.
A
A
Very
good
wow
we're
five
over
thanks
everybody
for
for
demoing
and
sharing.
Today
it
was
very
nice.
I'm
good
to
see
you
all.
Martin,
I
don't
know
I
I
desperately
tried
to
to
not
have
you
received
the
mvp
award
again
this
time,
because
I
just.
A
Anyway,
thanks
for
letting
me
harass
all
of
you
guys
it's
fun
so
next,
so
in
a
week
from
now,
what's
16
plus
seven
we're
close
to
christmas
at
that
point,
we'll
still
meet
I'll
still
be
here.
If
you
know
anybody
else
shows
up
we'll
still
be
meeting
next
week
on
wednesday,
but
on
friday,
that's
christmas,
and
so
we've
gone
ahead
and
cancelled
the
community
meeting
next
friday,
ten
days
from
now
so
other
than
that,
we're
we're
on
so.
A
Okay,
yeah
see
everybody
in
slack,
we'll
see
you
on
friday.