►
From YouTube: DASH Behavioral Model WG Sep 1 2022
Description
PR207 Marian will approve right now
PR201 includes more than just renaming APIs to be compatible w/SONiC software; Fixes make file, Autogeneration of utils, etc…
Chris to play w/autogen today
PR202 Marian to try this today now that he has his environment back
Chris Sommers to continue working with ACR and Runners for images
A
Our
first
behavioral
model
meeting
I'm
displaying
my
screen
here
and
if
I
remember
correctly
last
time,
we
were
mostly
all
waiting
on
193,
which
seems
to
have
been
approved,
merged,
etc.
And
so
I
was
wondering
if
everyone
has
become
able
to
move
forward
now
with
their
items
that
were
waiting
on
193.
B
A
Oh,
it
is
okay,
okay,
and
is
there
anyone
on
the
call
who
needs
an
explanation
of
what
193
was
or
anything
or
are
we
all
on
the
same
page.
A
We're
good
okay!
So
so
this
one
here
is
first
and
then
this
one
and
marian
thanks
for
doing
that.
A
For
193.,
so
did
anyone
want
to
pick
up
with
their
pr
for
the
week.
C
I
had
207,
this
is
the
case.
A
C
Previously,
these
tables
only
had
two
actions,
no
parameters,
so
we
were
not
able
to
add
the
default
default
only
tag
to
the
deny
action,
because
that
would
that
would
not
generate
any
any
parameters
at
all
in
the
slide.
There
would
be
no
psi
attributes
generated
so
change
the
psi,
auto
generation
tool
to
kind
of
just
have
the
permit
action
alone
and
remove
the
deny
action
if
the
default
drop
default,
only
tag
is
set.
B
Action
attribute
it's
just
that
the
only
choice
left
is
permit.
C
B
D
Mario
and
zurdy
interaction
with
pr201,
which
is
passing
now,
and
what
do
we
want
to
do
about
that?
One.
B
B
Yeah,
so
I
don't
think
any
anything
will
anything
from
the
207
will
affect
this
one,
but
probably
I'll
just
rerun
the
pipeline
anyway.
D
I'll
play
with
that
today,
I
want
to
check
out
the
auto
generation
and
you
know
see
how
you
replace
the
hard
code
so
I'll
play
with
that
later.
I'm
glad
you
got
to
that.
D
Probably
this
afternoon,
that's
good,
that's
a
good!
That's
a
real
good
addition.
D
Yep,
will
there
be
any
impact
or,
let's
say
any
collision
with
what
mukesh
just
did
or
are
they
kind
of
in
orthogonal.
B
As
I
said,
99,
no
because
mukesh's
change
is
related
to
enum,
which
is
the
value
for
attribute.
My
changes
are
on
in
the
scope
of
api
structure
name,
so
they
shouldn't
intersect.
A
Okay,
cool
yeah,
we
just
we
were
waiting
on
193,
which
was
a
outbound
pipeline
packet
test.
We
merged
it
and
we
merged
187.
It
was
default,
drop
support
for
p4
table,
miss
blah
blah
blah.
So
just
those
couple
of
things
and
then
marion
did
a
lot
of
work
on
201.
So
that's
what
we
went
through.
Then
we
have
behavioral
model
project
here,
which
is,
we
will
do
a
view
by
status
here,
and
so
we
did
add
a
few
things
to
the
backlog
here.
A
A
So
mukesh
was
147.
Was
that
done
now.
C
Yeah
this
this
last
year
was
the
last
piece
once
that
is
merged.
We
can
close
this.
B
A
Okay,
cool
awesome
and
then
that'll
move
into
the
done
category
and
then
there's
a
lot
of
in
progress
here
for
the
are
we
still
waiting
on
a
group
or
resource
to
come
back
for
for
connection
tracking
or
simulator
development?
That
kind
of
thing.
B
A
B
Integrate
the
code
into
into
our
forks,
so
it
will
need
changes
to
the
docker
files,
basically
to
the
dockers
themselves,
because
we
are
using
public
names
for
the
compiler.
B
So
right
now
the
the
work
is
ongoing.
A
We'll
check
back
on
this
then
check
back
next
week,
maybe
or
a
week.
B
A
D
B
A
So
these
these
are
the
ones
at
the
top
that
we're
just
kind
of
waiting
on
assignment.
If
there's
you
know,
I
mean
I
mean
some
of
these
have
to
wait
like
writing
apis
for
state
machine
testing.
Obviously
we
have
to
wait
on
that
one
and
counters.
We
need
to
figure
out
what
we're
counting
before
we
start.
You
know.
Writing
counters.
That
kind
of
thing,
mario,
mario
from
pinsondo,
amd
added
these
two
here,
which
will
be
good
ones
to
take
on
so
yeah.
That's
all
I
have
for
this
week.
D
Okay,
so
just
just
a
few
things
I'll
actually
take
this
screen
over.
So
thank
you
see.
D
So
I
mentioned
yesterday
a
couple
of
pr's
which
I'd
like
to
move
forward
soon.
This
is
just
documentation,
mainly
this
one.
It's
now
going
to
be
blocking
me
from
subsequent,
let's
say,
docker,
work
and
also
mara,
and
the
work
you're
doing
with
bmb2
eventually
will
have
to
coordinate
all
these
docker
files
and
images
and
sort
things
out.
D
Great
right,
just
a
few
victims
or
guinea
pigs
to
try
it
out
to
make
sure
I
didn't
miss
something
critical,
because
the
whole
point
of
this
is
to
finally
fix
this
technical
debt,
and
then
I'm
working
on
I
mentioned
yesterday
briefly
about
using
the
acr
repository
for
our
images.
Instead
of
my
docker
hub
and
I've
actually
got
a
branch
and
this
branch
is
under
azure
dash
and
I'll.
Explain
why
it's
called
publish
docker,
acr,
publish
dockers
to
acr.
D
This
was
created
for
us
by
microsoft,
I.t
people
christina's
help
and
I'm
now
pushing
all
these
images
to
sonic-acr,
and
I
can
use
them
to
build
from
so
I'm
going
to
do
some
more
refinement
of
this,
but
once
we
do
this,
then
we'll
have
to
do
off
to
do
a
big
merge
so
that
all
the
new
images
are
pulled
from
acr
and
the
result
will
be
the
next
time.
People
use
all
that
merge
code.
They'll
have
to
do
a
one-time
pull
of
all
these
new
images
because
they
have
a
new
repo
name.
D
Even
if
they're
the
same
image,
they'll
have
a
new
repo
name
in
their
path,
so
you'll
be
pulling
those
to
your
environment,
but
so
far
it's
gone
surprisingly
easy.
All
I
had
to
do
was
find
the
right
piece
of
code
to
steal
from
another
place
and
you
know
put
it
in
the
right
place
and
it
logs
me
into
acr.
D
Otherwise,
it's
a
standard
github
action
file,
so
this
is
very
encouraging
for
me,
and-
and
so
you
know
with
this
going
on
the
docker
permissions
once
that's
merged
off-
to
use
those
images
instead
of
these
older
ones,
and
I
have
to
come
up
with
a
workflow
to
do
this.
One
thing
I
I
said
a
moment
ago
was:
I'm
doing
this
in
a
branch
off
of
azure,
I'm
not
doing
it
in
a
fork.
D
D
All
that
means
is
it's
kind
of
it's
kind
of
a
special
once
in
a
while
thing
to
do
is
to
create
a
branch
and
push
push
new
images
to
acr,
but
that's
the
workflow.
You
know
you
can't
just
do
a
fork
and
push
things
to
acr
unless
you
have
account
access,
so
I
need
to
really
figure
out
the
workflow
here.
D
Go
through
it
a
few
times,
maybe
get
someone
to
try
also
who
acts
in
the
in
let's
say
as
a
maintainer,
so
this
will
be
kind
of
a
maintainer
activity
and
it
would
be
good
to
have
a
few
people
who
act
as
maintainers
to
go
through
this
with
me
in
these
trial
phases,
and
we
work
out
the
kinks,
I
shouldn't
be
the
only
person
who
knows
how
to
do
this.
We
don't
want
single
points
of
failure
right.
We
want
multiple
people
who
can
handle
this
so
I'll.
D
Try
to
work
it
out
I'll,
try
to
document
it
we'll
go
through.
Some
practice
runs
together
and
then
you
know
start
using
it.
Does
that
sound
like
a
plan
right,
there's,
probably
a
lot
there,
but
once
I
document
it,
I
think
it'll
be
clear.
You
just
can't
push.
You
know
images
willy-nilly
to
a
docker
register.
You
have
to
have
account
access
and
that's
why
I've
been
doing
it
so
far
to
my
own
docker
hub
and
it's
been
occasional,
but
we
need
to
make
this.
You
know
just
a
joint
kind
of
activity
now.
C
D
Great
I
had
I
had
you
in
mind
because
you've
spoken
up
about
these
kind
of
things
in
the
past,
when
you
found
room
for
improvement
like
the
doc
to
make
file
perms,
and
all
that
that
was
you
kind
of
pushed
me
over
finally,
getting
rid
of
that
technical
debt,
and
I
thank
you
for
that.
This
is
fun
anyway.
We
get
to
learn
some
new
things
along
the
way,
I'm
not
a
I'm,
not
a
ci
wizard.
I'm
learning
this
as
I
go
so
anyway,.
C
D
D
A
Fun,
I've
been
watching
p4
youtube
videos
from
andy
fingerhead's
teammates,
trying
to
learn
new
things
there.
So
yeah,
it's
been
fun.
D
Cool
and
hannah
to
your
point
on
this,
you
posted
this
make
run
all
test.
Fails
this
issue
205
and
I
spent
some
time
this
morning
getting
to
the
bottom
of
it,
and
I
think
I
know
what's
happening
and
I
have
a
a
quick
work
around
which
I
just
put
this
up.
D
D
I
don't
know
why.
I
need
to
work
with
our
teams
to
figure
this
out,
but
I've
got
a
way
to
reproduce
it,
and
you
know
you
can
take
a
look
at
this
mukesh
and
hanaf,
and
I
have
been
having
a
dialogue
on
this,
and
so
I
did
some
investigating.
You
could
take
a
look
at
that
and
if
you
just
do
a
little
experiment
and
and
confirm
that
the
quick
work
around,
which
is
just
deleting
all
you
have
to
do,
is
delete
this
from
one
make
target.
C
D
Oh
you're
saying
running
this
re-enables
things.
Yes,
that's
interesting,
you
know
it
might
actually
be
a
like
a
side
effect
of
docker
compose.
I
I
don't
know
what's
causing
it.
So
that's
why
I
have
to
go
to
dig
down
to
the
next
level
and
probably
get
some
people
more
experienced
than
this,
but
I,
finally
they
have
a
way
to
reproduce.
That's
very
simply,
I
think,
but
that's
that's
good
information
too.
D
So
anyway,
we're
working
on
this.
It's
it
seems
like
it's
a.
It
only
happens
once
so.
If
you
re-ran
the
test
hannah,
it
probably
would
pass.
C
Okay,
yeah,
let
me
let
me
try
that
you
know
definitely
I'll
give
it
a
shot
and
then
thank
you
so
much
for
helping
to
root
cause.
This
issue,
and
both
you
and
mukesh
have
been
pretty
helpful
and
then
I
you
know
I'll
definitely
give
it
a
shot
for
whatever
the
suggestions
you
have
awesome
we'll
go
over
the
hump.
Definitely.
D
E
Right,
chris,
you
may
be
aware
of
that.
There's
some
sys
control
linux
commands
when
you
create
virtual
ethernets
that
by
default,
linux
enables
certain
kind
of
ip6
control
packets
to
be
generated
across
them.
I
put
in
shadow
some
commands
that
can
disable
those
I
don't,
but
that
might
not
be
the
p6
packets
you're,
seeing
so.
D
E
Finding
those
are
the
commands
that
I
was
I
was
linking
you
to
so,
if
you're
already
using
those,
then
I
don't
know
what
what
you're
seeing.
D
Yeah,
it's
it's
been
an
ongoing
saga,
so
we're
getting
closer.
D
E
Yeah,
just
like
that
glad
you
already
know
about
that,
so
I
might
my
knowledge
is
spent
there.
D
A
So
I
guess
the
last
thing
would
be
then
maybe
an
intro
it
looks
like
eddie
rowan
has
joined
an
eddie.
I
don't
know
if
we've
met
before.
Maybe
you
could
do
a
quick
intro.
C
A
Okay:
okay,
we
have
that,
I'm
sorry,
your
name
looked
familiar
and
but
I
couldn't
pull
it
off
the
top
of
my
head.
So
great,
thank
you
and
thanks
for
coming
to
the
meeting.
A
So
yeah
I
appreciate
it
so
everybody,
that's
all
I
have
unless
you
have
anything
else
and
feel
free
to
reach
I'll,
stop
the
recording
feel
free
to
reach
out
chris
me
at
microsoft.com.
If
you
need
to
email
me
I'll,
stop
the
recording
and
post
it
like
normal,
and
I
hope
you
have
a
good
day.