►
From YouTube: Meshery Build & Release Meeting (Oct 14th, 2021)
Description
Meshery Build & Release Meeting - October 14th 2021
Join the community at https://layer5.io/community
Find Layer5 on:
GitHub: https://github.com/meshery
Twitter: https://twitter.com/mesheryio
LinkedIn: https://www.linkedin.com/showcase/mes...
Docker Hub: https://hub.docker.com/u/layer5/
B
A
D
B
B
Cool
and
so
while
we
prep
to
talk
about
the
v060
release,
it'll
be
nice
to
welcome
yasser.
B
You
today,
yeah
nice
day,
yeah
yeah
nice
to
nice
to
meet
you
good
to
yeah
you're.
I
recognize
the
profile
photo
from
yeah,
actually
yasser.
It's
a
well,
it's
a
community
tradition
to
force
newcomers
to
awkwardly
introduce
themselves
and
just
say
a
few
words
sort
of
you
know.
Where
are
you
dialing
in
from?
What's
your
current
focus?
What
brings
you
to
the
community?
You
know
it's
kind
of
an
icebreaker.
E
Well,
to
give
you
a
brief,
I'm
a
devop
architect,
I've
been
in
the
devops
arena
for
quite
a
while,
and
my
generally,
my
main
focus
now
is
go
language
and
I've
been
in
the
mesh
industry.
For
I
think,
quite
a
few
months
back,
I
have
used
the
istio
I
have
used
also
console.
I
have
used
traffic
and
several
other
mesh
solutions
and
I'm
very
pleased
to
be
within
this
community
tackling
the
most
important
issues
in
the
mesh
industry.
B
Yeah,
nice,
oh
that's,
perfect!
Yeah,
that's
great
yeah!
Gosh.
What
to
say
there
boy!
You
know
yasser
actually,
there's
a
lower
number
of
people
attending
this
call
than
usual.
I
mean
in
part,
because
we
just
are
this
particular
call
anyway,
because
we
just
got
done
switching
this
call
from
like
a
continuous
integration
focus
to
a
broader
sort
of
build
and
release
focus,
excellent.
So,
but
there's
a
lot
of
well
speaking
of
kind
of
the
speaking
of
those
important
initiatives
within
the
service
mesh
landscape.
B
I
was
just
chatting
with
a
couple
of
other
maintainers
on
service
mesh
interface
and
they
are
using
or
we
are
creating
an
well
an
smi
controller
sdk
so
that
it'll
make
it
a
bit
easier
for
different
service
meshes
that
are
implementing
smi
to
have
a
to
have
a
a
development
kit
to
create
smi,
compatible
controllers.
B
Part
of
that
kit.
There's
a
relationship
to
what
we're
talking
about
here,
like
part
of
that
kit,
will
include
a
github
action
for
smi
conformance.
B
Integration
is
still
definitely
in
scope
of
this
meeting,
but
we
also
recognize
that,
as
measuring
is
about
to
release
its
dot
six
release
we're
getting
closer
and
closer
to
a
1.0,
and
we
want
to
give
you
know,
you
know,
build
up
the
practices
around
test
strategies,
specific
test
plans
and
you
build
a
muscle
and
kind
of
routine
around
release.
You
know
hardened
releases,
so.
B
Yeah
yeah
nice
we've
been
doing
a
few,
others
have
joined,
that's
nice,
some
familiar
faces.
A
All
right
welcome
guys
to
the
machery,
build
and
release
meeting.
So
the
main
topic
for
today
is
definitely
the
v0.6
release.
A
I
think
we
are
ready
to
go
ahead
with
release,
but,
as
mentioned
in
the
midi
minutes,
there
are
a
couple
of
pending
issues
which
were
supposed
to
be
a
part
of
this
v0.6
release,
one
of
which
is
migrating
measuring
ctl
to
use
helm
charts
instead
of
the
kubernetes
manifest
file,
and
another
area
that
we
might
need
to
look
into
is
the
seed
content
for
measuring
cloud.
A
As
far
as
I
think,
these,
the
help
chart
issue
is
not
not
not
that
big
of
an
issue,
because
we
still
can
users
can
still
deploy
to
kubernetes
directly
with
the
current
implementation.
But
this
dc
content
for
measuring
cloud
might
be
an
issue
that
we
need
to
address
before
we
release
or
make
the
release
more
public
and
maybe
attract
more
users.
A
A
Thing
that
we
we
need
to
get
ready
is
the
release
blog
or
the
release
blocks.
So
the
this,
the
the
official
release
blog,
is
almost
ready,
like
from
my
part
like.
I
will
make
a
pr
like,
after
this
a
meeting
to
add
that
blog
and
some
of
the
other
one-off
blocks
are
still
pending,
but.
D
A
Is
that
can
happen
like
even
after
the
release
like
we
can,
we
can
keep
pushing
those
blocks
out
after
we
make
a
release
as
well
so
yeah.
With
that
context,
let's
discuss
discuss
about
these
two
items
in
particular.
First
then,
like
let's
see
if
we
can,
if
we,
we
can
in
fact
make
a
release
today,
itself.
A
So
yeah,
so
for
the
the
reason
I
mentioned,
seed
content
for
measuring
cloud
is
the
priorities
is
in
fact,
because
most
users
are
likely
to
choose
the
measuring
provider
when,
when
they
start
up
and
without
having
any
content
in
any
sample,
applications,
any
sample
filters
or
any
sample
service
mesh
patterns,
users
would
not
be
able
to
really
use
those
beta
features
so
yeah.
A
B
Yeah
or
we
didn't,
we
didn't
necessarily
have
to
go
out
of
order,
but,
but
I
mean
I,
I
would
propose
that
well,
one
that
that
this
week
is
a
good
week
for
announcements.
You
know
we
made
two
big
ones
already
and
that
I
think
we'll
I
think
in
general
collectively.
I
know
that
there's
a
bunch
of
people
signed
up
to
write
blog
posts
on
forthcoming
features,
but
I
think
we'll
probably
struggle
a
little
bit
to
get
all
of
that
written
up.
B
I'm
not
it's
not
about
the
writing
as
much
as
I
guess.
What
I
mean
to
say
is
we'll
struggle
a
little
bit
to
get
the
the
dot
six
release
out
and
adding
and
like
I
don't
like.
B
I
don't
know
that
she'll
hold
up
the
release
like
that
design
will
take
a
little
while
to
to
do
and
then
to
to
get
right,
we'll
just
we'll
end
up
pushing
the
dot
six
release
back
and
further
back
and
for
the
back.
These
are
beta
areas.
B
I
don't
know,
but
I'm
not
just
kind
of
thinking
aloud
like
one
thing
to
do
is
to
in
the
ui
itself
is
to
have
a
link
to
a
guide
that
says:
here's
how
you
here's
some
sample
content
go
go
find
it.
I
mean
that
way,
people
when
they
go
to
a
blank
page.
It's
like
oh
okay,
like
great.
B
Yeah,
I
don't
know,
but
the
thing
that
is
more
is
for
certain
kind
of
critical
and
we
know
affects
the
way
that
people
deploy
measuring
into
kubernetes
is
the
pervasive.
Is
the
pervasive
switch
to
using
helm,
charts
and
trying
to
make
it
there?
B
If
we
refocus
there
a
suit
like,
under
the
assumption
that
we
could
write
a
blog
post
or
a
a
doc
in
mesh
redox
that
then
links
people
to
how
to
load
up
sample
content.
Assuming
that
that
that
we
could
achieve
that.
B
If
we
talk
more
about
the
measure
operator
deployments
in
the
server
rudraksha
is
that
is
that
something
you're
focused
on,
or
is
that
something
that
darren
is
focused
on.
F
F
F
Mean
if
we
introduce
another
chart
for
that,
we'll
have
some
new
concerns
about
versioning
and
doing
stuff
related
to
that,
and
that
might
come
with
its
own
caveat.
So
I
guess
we
can
keep
it
the
way.
It
is
right
now.
So
it's
like,
we
don't
need
to
publish
a
new
chart
for
mystery
operator
and
it
can
stay
the
way
it
is
mystery.
Server
would
have
some
things
so
that
it
can.
It
will
only
install
the
operator
and
the
other
resources.
B
That's
fantastic!
That's
awesome,
that's
great
briefly,
just
as
we're
by
the
way,
as
it's
a
small
meeting
today
so
nice
to
warmly
embrace
yasser
yasser
quick
question
just
in
getting
in
gaining
getting
to
know
you
more
so
so
you
you,
you
have
some
familiarity
and
comfort
with
golang
and
then
and
then
I
guess
you
know
clearly
through
having
deployed
the
other
meshes
that
you
talked
about.
B
E
Well,
I
can
leave
with
you
for
quite
a
bit
of
time.
I
can
spare
some
hours
but
in
fact
I'm
just
getting
acquainted
with
the
repository,
I'm
just
being
familiar
with
the
code
with
the
repository
the
charts
are
variable
and
having
got
a
chance
to
build
a
quite
preferable
release
in
a
while.
E
So
I
think
I
I
need
some
some
links,
some
ready
links
from
where
to
start
from
what
to
start
with.
B
It
makes
to
makes
perfect
sense,
yeah,
totally
and
I'll
I'll.
For,
for
my
I'll,
send
you
at
least
maybe
just
one
I
was
gonna
say
a
couple,
but
maybe
just
one
just
because
so
that
you're
able
to
give
give
it
a
well
give
give
this
upcoming
measure
release
a
healthy
poke
yeah,
because
because
there's
some
rough
edges
in
there
still-
and
I
think,
we've
been
looking
at
the
we've
been
looking
at
those
edges
so
long.
E
The
building
strategy
is,
I
think,
you
have
written
some
kind
of
a
document
to
for
a
build
strategy
for
for
get
github
for
code
repositories.
Yeah.
I
think
I
I
guess
it's
in
the
interview
on
the
newcomer
document
that
you
are
showing
how
you
are
going
to
prs
how
to
initialize
these
pr's
and
from
where
yeah
it's
branching.
Also.
B
B
Yeah
yeah
there
is
yasser,
brings
up
a
good
point.
Actually
is
that
on
measuring
docks?
There
is
a
section
that
talks,
that's
just
dedicated
to
the
build
and
release
topic,
so
it's
like
the
yeah
and-
and
this
is
fairly
now
it's
fairly
accurate.
There's
one
thing
in
here,
though,
that's
changed
kind
of
recently
or
or
I
don't
know
that
this
talks
about
our
helm,
charts
and
so
rude
rocks.
B
It's
potentially
something
to
mention
to
darren
as
well:
it's
just
orto
asuko
asuko's.
He
just
scored
a
46
out
of
50
on
his
english
exam,
so
he's
so
actually
have
asking
him
to
write
a
little
bit
of
doc
on
helm.
B
Charts,
which
is
his
forte,
is
probably
a
good
exercise
for
him.
So
I'll
I'll
go.
Do
that
actually
I'll
include
you
guys,
but
I
think
it
would
be
good
for
him.
It
would
be
good
for
us
to
have
some
talks
about
how
the
project's
building
and
where
it's
publishing
the
helm,
charts
I
think,
on
the
venue.
If
you
go
to
the
very
top
there's
an
artifacts
table,
I
think
we
list
that
the
charts-
oh
no
okay,
yeah
they're,
oh
no,
okay,
boy,
yeah!
B
A
And
like
I
think,
we
have
actually
postponed
some
pr's
off
to
v
series,
seven,
like
just
so
that
this
looks
clean
like
we.
We
we
changed
priorities,
or
we
just
decided
that
some
of
these
did
not
necessarily
be
in
this
release.
So
that
is
one
thing
and
we
have
a
couple
of
open
issues
for
with
the
milestone,
some
of
which
are
documentation,
and
it
would
help
us
like
if
we
go
through
this
and
ensure
that,
like
we
don't
miss
anything.
A
As
far
as
I
know
like,
we
didn't
miss
anything
but
like
if
anything
like
went
through
the
cracks.
Please
let
me
know
as
well
and.
B
A
Yeah,
so
looking
through
this
appears,
most
of
this
stuff
could
go
into
subsequent
releases,
so
we'd
be
6.1.
It
shouldn't
be
much
of
an
issue
so
yeah.
A
Okay,
so
I
think
like
we
had,
we
discussed
most
of
this
stuff.
I
think,
like
push,
is
ready
to
finish
this
off,
so
we
have
the
proper
functionality
with
the
performance
configuration
file
as
well
as
we
have
some
user.
Our
user
experience
improvements
across
the
perf
commands
as
well,
so.
B
B
Yeah,
I
think,
yeah.
I
think
the
only
other
thing
it's
just
yep
the
sub
chart
for
mystery
operator,
or
you
know,
switching
out
that
mesh
reservoir
code
to
do
helm
not
manifest,
and
then
at
least
of
the
p.
I
don't
know
if
anyone
else
saw
any
other
prs
that
couldn't
wait.
4.7
it
looked
like
it
was
only
the
persistence.
We
have
some
documentation
ones
that
very
last
one
looks
a
little
old
and
I
guess
I
just
wonder
if
we're
even
like
what
we're
thinking
of
doing
with
this.
A
B
Oh
okay,
yeah!
If
given
given
the
severity
of
the
issue
like
it's
actually,
what
what's
interesting
is
given
the
severity
of
the
issue.
It
can
probably
wait
actually
aditya.
I
think
he
just
took
on
the
issue
a
little
bit
ago
and
then
I
think
from
and
then
he
hasn't
opened.
Does
he
open
a
pr
on
that?
I
think
there's
an
open
pr
right
now.
F
B
B
A
So
this
is
going
to
be
the
complete
block
and
we
will
have
other
individual
blocks
for
particular
set
of
features.
A
A
Regarding
the
the
individual
blocks,
people
have
signed
up
to
take
some
of
this,
but
I
don't
think
much
have
been
written
the
the
blogs
for
the
github
actions.
It's
an
open
pr,
it's
already
an
open
pr,
but
I'm
not
sure
about
the
other
ones.
Maybe
saying
like
is
there
any
blocks
here
that
have
been
written?
That
needs
reviewer
need
our
attention.
D
We
actually
need
our
attention.
Filter
person
who
has
been
assigned
has
not
been
responding
so.
D
D
Yeah,
I
was
just
wondering
if
I
can,
if
I
can
take
the
contents
of
filters,
and
can
any
of
you
give
me
an
opinion
like?
Could
I
be
able
to
do
that
given
to
something.
A
B
Are
we
blog
posts
about
specifically
about
the
filters
or
about
the
filter.
B
Yeah
well
so
yep
there's
a
little
bit
of
a
mixed,
so
mesherie
does
deploy.
Does
let
you
load
up
your
filters?
B
Does
let
you
configure
your
filters,
does
let
you
actually
deploy
one
filter
specifically
from
the
ui
and
that's
the
image
hub
filter,
because
that's
a
kind
of
a
baked
in
consideration
as
part
of
that
sample
app
and
you
can
separately,
you
can
deploy
the
sample
app
and
then
you
can
deploy
the
filter
and
so,
but
that's
just
for
one
as
we
genera
as
we
make
this
a
generic
capability
to
manage
any
webassembly
filter.
B
So
the
ability
for
measury
to
do
that
from
the
server
is
present
and
it's
in
the
new
service
mesh
patterns
method
and
under
that
methodology,
which
is
a
lot
more
flexible
and
that
that
raw
capability
is
present
in
measuring
server.
B
The
ui,
though,
for
up
deploying
them
and
applying
them.
B
B
It
assumes
that
you'd
use
your
current
context
and,
like
you
know
like
it's,
it's
under
form.
That's
why
it's
in
beta,
but
but
I
guess
the
question
here,
is
it
still
in
alpha
or
has
it
made
it
to
beta?
If
you
can,
if
it's
a
rude
rocks,
I
don't
know
if
you're
an
individual
that's
closer
to
this,
but
the
question
is
like:
if
you
select
a
filter
from
the
table,
can
you
deploy,
even
though
you
can't
necessarily
select
all
well?
I
guess
like
actually
through
the
pattern
you
can.
B
You
can
select
where
you'd
like
for
it
to
go,
what
what
workload
you'd
like
for
it
to
be
applied
to.
B
A
B
Yeah,
I
think
I
think
the
only
the
that
is
true
and
that's
basically
what
I
was
saying
about
measuring
server
being
able
to
do
it.
I
guess
the
question
just
more
or
less
becomes
it
doesn't
that
it
doesn't
have
all
the
flexibility.
Well,
it
does
like.
Actually,
if
you
give
it
the
right
yaml,
it
has
all
the
flexibility
that
you
want.
A
Yeah,
like
you,
can
technically
deploy
filters,
provide
that
they
are
stored
somewhere,
and
you
use
this
patent
file
as
a
configuration,
okay,
so
yeah.
So
basically,
there
is
a
filter
file
that
is
stored
somewhere
here,
awesome
binary.
So
basically
this
this
is
a
filter.
I
guess.
C
B
So,
okay,
that's
good!
We
should
you
know,
do
a
double
check
with
boot
crush
on
this,
but.
A
Yeah,
the
only
functionality
missing
here
is
measuring
actually
being
actually
taking
care
of
the
binary,
as
well
as
the
filter,
as
well
as
the
configuration
so
yeah
right
right.
B
Okay,
good
good,
although
if
you
go
back
to
the
example
that
you
were
just
on.
A
B
Okay,
so
the
user
experience
is
not
what
is
ultimately
like.
That's
it
right
there
right
the
right
there
yeah
the
value,
that's
the
configuration,
so
the
ux
is
not
quite
what
we're
hoping
for.
A
A
That
I'm
not
sure
like
we
used
to
discuss
it
on
the
wasam
filter,
school.
A
We
can
document
this
and
mention
that
this
is
a
beta
feature
and
maybe
like
we
have
a
couple
of
pattern
files
describing
awesome
filters.
So
we
can
also
maybe
talk
about
that
in
the
blog
post
as
well.
D
B
Oh
yeah,
that's
I
wrote
that,
like
that's
done,
like
that,
no,
no!
I'm
no
kidding
so
in
this
case
all
right.
The
thing
that
we're
looking
to
achieve
with
dynamic
management
uis
well,
we
kind
of
have-
I
haven't
written
this.
So
to
answer
your
question
up
front.
B
But
it's
not,
but
what
we
have
right
now,
it's
and
by
the
way
just
like
for
yasser.
In
particular,
I
know
I
keep
on
picking
on
yasser
because
because
I'm
excited
that
he's
here
but
like
this,
I'm
glad
that
we're
talking
about
docs
or
blogs,
because
well
it's
one
of
those
things
we
don't
have
enough
of,
and
so
just
as
you
familiarize
with
the
project
like,
don't
don't
hesitate
to
to
do
the
same
yep.
B
It
too
is
not,
ultimately
what
you
know
is
desired,
but
it's
closer
or
but
it's
a
new
yeah
it's
it
is
closer
to
what
we
would
have
in
terms
of
a
dynamic
management
ui
for
istio
and
specific.
B
The
fact
that
we
support
so
many
versions
makes
this
really
painful
to
look
at,
and
it's
also
not
obvious
like
this
is
this
is
supporting
all
of
the
steel
versions
of
the
various
auth
policies
shoot
this
particular
feature.
Like
I
don't
know
you
could
say
it
was
ready
for
beta.
B
Abi
kumar
he's
he's
working
conceptually
he's
working
on
collapsing,
these
into
versions
categorizing.
B
Okay,
good
and
then
once
once,
that's
there
that'll
be
a
lot
that
actually
might
be.
The
last
thing
to
be
done
is
that
that
that
just
needs
that
will
make
it
navigable,
the
the
other
part
of
the
challenge,
and
so
like
rodolfo.
This
is
a
good
thing
for
you
and
for
others
to
reflect
on
like,
for
my
part.
B
I
know
I've
been
staring
at
this
challenge
too
long
and
don't
have
I'm
not
bringing
good
suggestions,
but
let
me
describe
what
we're
trying
to
do
here
so
that
you
guys
can
reflect
on
what
you
think
would
be
a
good
user
experience.
So
that's
more
or
less
that
so
right
now.
The
interface
that
we're
in
is
that
someone's
taking
a
look
at
a
service
mesh
pattern
patterns
are
captured
in
yaml.
B
B
Kubernetes
manifests,
as
a
matter
of
fact,
we
just
gave
a
talk
on
this
at
service
meshcon
this
week.
I'm
trying
to
make
it
not
take
30
minutes
to
explain,
and
that's
just
to
say
that
patterns
are
on
templates
for
service
mesh
configurations
that
people
might
want
to
use,
there's
a
bunch
of
ways
to
configure
your
mesh,
and
so
how
do
you
know
if
you're
doing
it
right
or
if
others
are
doing
it
in
a
similar
fashion?
Well,
go
grab
a
pattern
like
how
many
retries
should
you
configure
in
your
environment?
B
B
Maybe
you
want
to
tweak
those,
maybe
best
practices
to
use
two
retries
per
service
just
sort
of
universally,
like
it's
the
magic
number,
maybe
I'm
making
it
up,
but
like
that
wouldn't
necessarily
be
far
off
okay,
so
that
pattern
can
be
expressed
in
yaml.
B
Someone
may
say
great:
I
see
the
the
best
practice
retries
pattern
in
the
list,
although
you
know
it's
not
going
to
do
it
for
me,
like
I
want
to
tweak
it
a
little
bit,
so
they
click
to
edit
it
and
so
right
now,
they're
in
edit
mode
and
the
ui
that
they're
seeing
well.
It's.
B
One
what's
happening
is
that
measuring
supports
all
of
the,
so
this
is
okay,
so
it
supports
all
the
construct,
in
this
case
we're
using
istio
as
the
example,
but
it
supports,
like
all
of
the
custom
resources
that
istio
has
so
authorization
policy,
a
destination
rule,
a
virtual
service.
B
B
So
the
ui
needs
to
be
enhanced,
like
enhanced
here
such
that
it's
not
showing
you
all
50
versions
of
the
author
of
that
custom,
resource
authorization
policy
and
then
all
50
of
the
destination
rule
and
all
50
of
the.
We
need
to
fix
that
user
experience.
B
B
If
you
do
me
a
favor,
if
you
scroll
down
a
bit
and
I'm
going
to
shut
up
here
in
a
minute
and
let
rudolfo
think
out
loud,
but
if
you
go
to
gateway
sure
just
like
this
there's
istio
gateway
is
an
ingress
gateway
and
egress
gateway
things
that
people
can
configure
gateways
help
ingress
gateway,
helps
you
redirect
traffic
once
once
it
reaches
your
service
mesh
and
directs
it
to
various
services
with
various
workloads
on
your
mesh.
B
So
you
know
you
need
to
configure
your
gateway
with
ports
and
protocols
and
redirect
you
know
different
rules
and
things,
and
actually
it
goes
on
for
a
long
time
like
there's
a
lot
that
you
could
potentially
configure,
and
this
user
interface
is
attempting
to
expose
all
that
configuration
so
having
described
the
problem,
I'm
going
to
step
back
and
and
kind
of
ask
you
know
if
you,
if
you
put
your
user
hat
on.
B
I'm
having
an
allergic
reaction
in
my
eye,
anybody
else
allergic
to
cats
and
have
a
wife
who
likes
to
have
cats
but
doesn't
care
if
you're
allergic
or
not,
because
oh
it's
painful,
okay,
yeah
to
to
rodolfo
like,
like
you
know,
gosh,
there's
just
there's
a
lot!
There's
it's
so
compact
right
or
it's
so
dense.
Rather,
if
you
so
to
rodolfo's
point
like
yeah,
it's
inevitable,
inevitable,
whatever
whatever
the
word
is
for,
like
I
can't
as
a
user,
I
can't
comprehend
what
the
heck
is
going
on.
B
I'm
not
confident
in
configuring
like
if
you
expand
one
of
these
accordions
again
I
get
lost
kind
of
quick
and
what's
going
on
and
and
that's
okay,
because
there
hasn't
been
a
lot
of
effort
given
to
really
re
refine
and
provide
the
right
information
architecture
in
this
accordion
to
rodolfo's.
Point
like
if
you
know
how
you're
ever
editing
a
word
doc
or
something
like
that.
Like
a
lot
of
you
know,
you'll
as
if
not
everyone
doesn't
do
that.
B
B
It
might
make
sense
that
what
you
want,
if
you're
going
to
focus
on
authorization
policy
and
if
it's
fairly
dense,
like
maybe
that
should
take
up
your
whole
screen,
like
you
know,
give
you
the
space
that
you
need
to
really
understand
what
all
these
inputs
these
fields
are
about,
and
then
let
people
configure
that
policy.
You
know
close
it
save
it.
B
G
B
Actually,
since
rodolfo,
since
you
brought
this
up
like
that,
that
actually
breaks
through
my
for
me
personally,
just
for
me
my
own
wall
around
how
to
around
what
abi
who's
just
joined.
His
current
challenge
in
terms
of
the
ux
here
has
been
to
group
all
of
these
components
by
service
mesh,
invert
well
by
service
mesh
and
then
by
service
mesh
version
and
actually
there's
a
toolbar
and
there's
kind
of
an
over
overarching
like
the
fact
that
you're
trying
to
apply.
B
B
You
know
like
a
retry
pattern:
that's
service
mesh
agnostic,
but
anyway,
for
the
time
being,
they
would
be
doing
this
configuration
in
context
of
a
service
mesh
type.
Secondarily,
in
context
of
a
service
mesh
version,
which
then
means
you
don't
have
to
show
them
all
the
components
for
all
the
versions.
B
Rather,
if
they
have
those
first
two
settings
which
kind
of
become
filters
if
they
have
those
configured,
if
they're
saying
well,
I'm
working
with
istio
and
I'm
working
with
this
version,
then
whenever
they
click
on
an
icon
on
that
toolbar
or
they
or
whatever,
that
ends
up
being
that
well,
it's
predetermined,
like
those
other
things
are
already
answered,
so
we
don't
have
to
show
all
the
versions
and
we
don't
have
to
show
all
the
service
messages
we
don't
have
to
show
all
the
versions
of
that
service
mesh.
It's
just.
C
Yeah,
yes,
yes,
I'm
just
thinking
like
what
else
we
can
do
like
the
collapsible
yaml
part
collapsing.
It
is
one
way
that
can
obviously
help
and
yeah
and
minimizing
what
is
the
last
thing
that
you
were
mentioning?
Was
it
minimizing
the
number
of
steel
versions
that
we'll
be
handling
like
from
50
or
so
to
15
or
25-ish?
B
Imagine
for
you
imagine
for
a
moment
if
you
will
that,
like
a
horizontal,
toolbar
kind
of
like
the
edit
pattern,
toolbar,
that's
there
right
now
and
that's
not
a
toolbar!
That's
a
cookie
crumb
that,
if
that
were
a
horizontal
toolbar,
that
the
first
two
selections
on
that
toolbar
would
be
service
mesh
type.
The
second
one
would
be
version
number
of
that
service
mesh.
C
B
And
that
the
the
very
leftmost
selection,
what
type
of
service
mesh
updates
the
values
that
are
available
for
the
version
field,
and
then
that
updates
the
value,
the
icons
that
are
available
on
the
toolbar
yeah.
Then,
when
someone
were
to
click
a
particular
icon,
which
is
a
custom
resource
like
authorization
policy,
what
they
would
get
is
well,
maybe
just
this
like.
It
wouldn't
necessarily
need
to
be
in
an
accordion
because
they
would
just
get
on
the
left-hand
side.
They
would
get
the
full
content
of
that
accordion.
C
Yes,
I
I
get.
I
get
that,
obviously
that
sounds
good,
actually
a
tool
bar
that
has
like
to
select
the
version
in
the
service
mesh
and
then,
according
to
those
configurations,
you'll
be
getting
all
the
things
that
you
can
edit.
So
that
sounds
good.
Actually,.
H
And,
and
and
and
for
that
reason
that
we
have
all
these
like
large
number
of
same
policies,
because
that
is
coming
from
different
persons
right.
B
H
No,
no,
no
I'm
saying
like
since
we
have
for
for
the
easter,
especially
we
have
different
versions
that
is
like
available
right.
So
that
is
so
that's
why
we
have
like
more
than
five
ten
fifteen
same
type
like
authorization
policy
dot
is
still
that
corresponds
to
just
is
it
correspond
to
like
the
certain
version
number
like
all
the
fields
here?
H
Okay,
okay,
okay,
sounds
good.
Okay!
I
will
be
doing
that
thing.
Okay,
I
was
just
a
bit
confused.
I
just
don't
get
full
time
too.
You
know
research
on
it.
So
probably
I
will
be
doing.
B
G
That
is
something
that
lee
explained:
that
if
you
select
one
service
message
type,
you
will
see
service
number
or
other
thing
that
is
related
to
the
service
mesh
type
selected.
And
then
you
will
see
some
versions
related
to
that
to
the
previous
options.
And
then,
after
selecting
all
these
three
filters
or
option,
you
will
see
in
a
whole
screen
the
general
edition
and
the
options
configuration.
A
All
right
so
yep,
thanks
rodolfo,
like
I
think
like
abby,
will
go
ahead
and
work
on
it,
since
we
are
almost
at
time.
Let
me
quickly
refresh
on
the
things
that
we
need
to
get
get
done
before.
We
make
a
release
so
related
to
seed
content
for
measuring
cloud,
since
it
might
take
some
design
effort
on
our
part
to
actually
implement
this
quickly.
What
we
can,
what
we
discussed
and
derived
it
was
create
sample
content.
A
So
even
if
a
user
uses
the
measuring
provider,
they
can
load
in
their
own
sample
applications
or
sample
filter
filters
from
the
url
we
provide,
and
another
missing
item
is
persisting
performance
test
results
from
measure
ct
like
it
is
being
actively
worked
on.
So
I
guess
we
can
wait
for
it
to
be
completed
and
then
make
a
release.
A
Is
there
anything
else?
I
miss
that
that
that
we
need
to
do
right
away.
F
Just
a
little
confusion
on
persisting
performance
systems,
there
are
two
issues
regarding
this:
one
was
around
the
smp
profile
and
results,
not
getting
persisted
due
to
that
and
the
other
one
was
when
you
create
a
performance
test
using
measuring
cpl.
It
gets
I
mean
once
you
create
to
run
a
test,
you
need
to
specify
flags
every
time
and
you
can't
just
reference
a
profile
that
you
created
earlier.
A
Okay,
like
could
you
share
the
issue
when
a
when
you
find
it
like
in
slacker.
F
A
A
All
right
anything
else,
we
missed
or
any
outstanding
pr
outstanding
tasks
that
we
need
to
complete
for
the
v06
release.
A
And
regarding
the
helm,
charts
what
we,
what
we
came
to
conclusion
was
we
can
probably
make
this
part
of
a
subsequent
release,
so
you,
the
users,
won't
feel
any
user
experience
changes
so
yeah.
That
is
that,
all
right,
everybody-
that's
we
are
out
of
time
today.
I
guess
we
made
some
progress
on
what
what
needs
to
be
done
for
the
next
release.
A
So
if
no
one
else
has
anything
to
add
right
now,
let's,
okay.
A
Oh
okay,
it
was
rushed
like
I
will
look
into
this
and
I
will
hit
up
views
as
well.
So
all
right!
Everyone
thanks
for
joining,
we'll,
see
you
probably
tomorrow
on
the
community
meeting
we'll
see
you
in
two
weeks,
bye.