►
Description
Defend engineers working with PM to breakdown upcoming issues into components, clarify requirements, and identify work boundaries.
A
A
We
can
find
a
way
to
avoid
having
a
synchronous
meeting,
but
for
right
now
considering
some
of
the
conversations
going
on
around
trying
to
be
more
accurate
and
our
deliverables
and
what
we
can
get
done
in
a
certain
iteration,
as
well
as
making
sure
that
we're
working
with
our
PMS
to
try
and
break
down
big
pieces
of
work
into
smaller
deliverable
slices
of
technology.
We
thought
that
for
right
now
having
a
synchronous
conversation
that
would
be
helpful
going
forward,
maybe
we
can
find
a
way
to
do
this
asynchronous
way
through
issues
or
other.
A
You
know
slack
other
partners,
but
for
right
now.
This
is
where
we're
at
so
this
meeting
is
focused
on
container
and
app
security.
I
know:
there's
two
groups
right
now:
vacant
systems
there's
a
application,
infrastructure,
security
and
runtime
application
and
security,
but
given
the
direction
that
we're
going
to
try
and
combine
these
into
one
group,
I
figured,
it
makes
sense
to
have
one
meeting
for
them.
A
Sam
is
the
new
PM
for
that
area,
so
he
is
gonna,
be
crucial
to
this
conversation
and
I
would
hope
that
Sam,
you
can
help
help
drive
as
much
as
possible.
The
discussion
around
the
issues
that
we're
gonna
be
looking
at
today
go
away
MFR.
So,
for
the
sake
of
time,
I
did
send
out
three
links
to
look
at
today
and
Sam
I
hope
if
we
can
make
it
through
those
three.
A
If
that's
gonna
be
its
progress,
it
may
not
be
all
of
the
issues
that
you'd
like
to
have
discussed
for
the
upcoming
release,
but
I
think
that's
pretty
much.
All
we're
gonna
get
time
for
and
a
half-hour
chunk
based
on
feedback
on
how
long
this
took
for
other
groups.
I
know
secure,
is
playing
around
with
this
as
well
I'm
trying
to
have
these
types
of
discussions
earlier
on.
So
any
questions
before
you
get
started
and
jump
right
into
looking
at
the
first
issue
of
the
list.
A
So
since
this
is
the
first
time
we've
done
this,
how
I
would
like
to
run
this
in
this
is
obviously
flexible,
so
that
I,
like
I,
like
Sam,
you
can
give
us
sort
of
a
high-level
description
of
the
issue,
so
we
don't
have
to
read
through
everything
and
then
we
can
jump
into
to
questions
and
suggestions,
ways
that
we
can
make
this
smaller
because
already
feels
like
a
big
story.
There
is
you.
B
Sure
so
you
know
really.
The
problem
that
we're
trying
to
solve
here
is
to
make
it
easier
for
users
to
turn
laughs
off
and
to
change
it
between
logging
and
blocking
mode
as
needed,
especially
in
light
of
the
recent
performance
bug
that
we
ran
into,
and
you
know
just
trying
to
help
ensure
that
they're
able
to
control
when
the
RAF
is
being
used
and
when
it's
not
you
know
for
any
product,
that's
sitting
in
line
in
a
customer
environment
that
is,
you
know,
sitting
in
front
of
a
production
grade
application.
B
If
we
ever
introduced
a
bug
or
have
a
problem
there,
the
customers
got
to
have
the
flexibility
to
shut
the
thing
off
in
a
very
easy
way.
So
you
know
recognizing
that
there
are
ways
that
they're
able
to
do
that
today.
None
of
them
are
entirely
optimal.
We're
trying
to
make
this
a
lot
easier
for
users
to
just
to
toggle
the
thing
off
if
it
ever
goes
goes
wrong.
B
The
proposal
was
to
put
it
on
the:
let's
see
the
operations
kubernetes
page,
one
of
the
big
drivers,
for
that
is
because
we're
shipping
modsecurity
out
to
all
users
all
the
way
down
to
get
LabCorp
and
I
feel
strongly
that
the
ability
to
have
this
kind
of
toggle
also
needs
to
go
all
the
way
down
to
core.
You
know
that's
something
that
anybody
who's
using
mod
security
needs
to
have.
B
Was
anticipating
this
to
be
a
real
time,
toggle
dear--my?
Otherwise,
if
that
would
cause
problems,
but
you
know
again
I'm
2
weeks
on
the
job
here,
but
I
would
expect
this
would
be
a
real
time
toggle.
You
know,
I
view
the
use
case.
As
you
know,
we
discover
a
latency
bug
that
just
showed
up
seven
days
after
we
turned
on
laughs
and
it's
impacting
our
production
environment.
I
can't
crash
my
whole
production
environment
I
just
need
to
shut
off
the
laughs.
Let
me
go
do
that
right
now.
C
C
C
B
C
I
think
the
other
concern
I
had
was
around.
We
have
like
a
product
category
of
infrastructure
as
code,
and
the
kind
of
the
direction
with
give
I
have
managed.
Apps
is
more
towards
using
a
like
a
cluster
management
project,
so
you
essentially
store
your
help.
Configuration
in
everybody
bow
thereby
make
it
easy
to
like
track
configuration
changes
and
so
I
think
if
we
wanted
to
stay
in
line
with
that,
this
would
have
to.
C
B
Mean
so
well
infrastructure,
as
code
has
a
lot
of
advantages.
It's
not
without
its
disadvantages
and
ease
of
use
and
speed.
You
know
I,
just
kind
of
see
being
able
to
turn.
This
thing
off
is
a
pretty
basic
task
that,
as
an
end-user,
you
know
needing
to
go
implement
that
in
code,
when
I'm,
just
trying
to
quickly
shut
it
off,
is
a
little
bit
cumbersome
for
accomplishing
that
task.
D
May
be
less
about
perhaps
less
about
how
the
it
gets
turned
off
then
the
time
it
takes.
So
if
it's,
you
click
the
button
to
turn
it
off
or
to
turn
it
on
and
push
and
it
kicks
off
a
deploy.
The
deploy
takes,
what
the
deploy
takes.
You
know
it's
starting.
It's
minutes
that
may
not
be
as
fast
as
some
users
may
like,
but
since
we're
infrastructure
is
code,
that's
reasonable
in
an
infrastructure
as
code
world,
perhaps
yeah.
B
I
think
the
timeframe
of
minutes
is
okay,
I'm
more
concerned
about
you
know
somebody
goes
in
there.
You
know
because
we've
defaulted
last
on
and
you
know
so
they
did.
It
may
not
have
even
realized
that
last
just
got
turned
on
for
their
environment
and
now,
although
all
of
a
sudden,
that's
causing
a
problem,
you
know
what's
the
learning
curve
and
ease
of
use
for
them
to
intuitively
figure
out
on
their
own.
B
How
to
turn
this
thing
off
I
mean
if
you
gave
somebody
kind
of
a
blind
test
of
this
thing
and
said
you
know
here,
go
figure
out
how
to
turn
flap
off
and
it's
the
first
time
they've
ever
interacted
with
laughs.
You
know
if
they
have
to
go.
You
know
whether
it's,
whether
it's
an
environmental
variable
or
whether
it's
logging
in
and
issuing
a
cute
cuddle
command
or
yeah
I
mean
that's
like
they
have
to
realize
that
it's
there
they
have
to
correctively,
go
look
up
our
documentation
and
figure
out
how
to
do
it.
B
A
A
E
B
I
would
rather
keep
mine
focused
on
the
end
customer
experience,
which
would
involve
both
the
back
end
and
the
front
end,
because
if
you
just
do
one
of
the
pieces
that
customers
not
going
to
get
the
value
and
then
kind
of
what
you
guys,
let
engineering
break
that
down.
I
know
we
don't
have
subtasks
on
issues
yet
and
there's
been
talk
about
how
to
work
around
that
with
get
lab.
But
you
know.
F
C
Trying
to
think
about
what
that
would
look
like
and
what
that
essentially
would
look
like
is
you
could
trigger
a
pipeline
with
maybe
an
environment
variable
or
an
option
to
trigger
this
that
would
get
processed
by
the
back
end
or
or
via
API,
that
pretty
much
exists
now
using
some
conglomeration
of
configurations,
options
into
a
pipeline,
so
I
would
almost
say
that
we
have
the
backend
portion
and
that's
why
it
gets
really
tricky
on
how
we
break
this
down
more
I.
Guess.
A
C
Well,
I
get
I
just
one
thing
that
just
came
to
mind,
though,
when
you're
saying
that
was,
we
need
some
kind
of
stateful
tracking
of
this.
Currently
we
do
not
track
like
the
current
state
of
the
active
application,
and
so
if
we
have
a
toggle
here
too,
that
that's
going
to
actually
say
blocking
note
is
on.
C
We
need
to
know
where
we're
getting
that
information,
and
is
that
a
live
query
to
the
cluster
with
that
we're
just
that
we're
caching,
somewhere
or
as
I
mean
that
were
actually
being
that
we're
actually
storing
a
staple
copy
of
the
state
of
the
cluster
with
the
negate
live
application
itself,
and
that
would
definitely
require
more
complex
back
in
word.
So.
C
D
C
That's
a
good
question,
so
it's
going
back
to
that
one.
The
infrastructures
code
managed
issue,
I,
guess
in
an
ideal
world
the
way
that
we
would
imagined
that
ending
up
is
through
this
UI.
You
set
block
that
commits
it
internally
to
a
repo,
so
it
keeps
the
repo
up
to
date
with
the
UI
through
the
sync,
which
is
exactly
how
our
wiki
works.
By
the
way
our
wiki
product
is
backed
by
a
git
repo
and
then
on
that
commit
it
triggers
a
deploy.
C
The
pipeline
pushes
it
out
everything
works,
so
that
would
be
like
the
complete
full
cycle
without
working,
both
tracking
a
three
infrastructure
as
code
way
providing
a
UI
and
then
allowing
it
to
be
Saints.
I
just
don't
know,
I
think
that
that's
often
the
stick
in
terms
of
everything
we
need
to
get
in
place
for
that
to
occur,
so
I'm
trying
to
figure
out
what
the
best
way
is
between
here
and
there
to
get
there,
because,
maybe
internally,
it's
all
bash
scripts,
doing
something
but
I.
E
H
The
reason
I
asked
was:
if
what
is
there
today
is
it's
possible,
but
not
extremely
user-friendly,
we're
not
the
way
we
would
like
to
do
it
as
the
infrastructure
is
code
under
configuration
perspective,
I
guess,
Sam
I
would
see
that
as
potentially
a
place
where
you
could
cut
that
off
a
sort
of
MDC
one
is,
let's
redo
the
non
UI
driven
portion
of
it.
Make
it
easier,
the
more
you
know,
maybe
clear,
through
documentation
how
to
do
this.
Setting
the
groundwork
for
the
wiring
it
up
for
a
UI
is
a
separate
one.
H
Is
it
I
think
Lindsay
the
way
that
I've
seen
it
done
here
and
like
sort
of
general
opinion
on
this,
if
something
cannot
be
broken
down
into
a
smaller
unit
of
customer
value?
If
there's
a
split
between
front
end
and
back
end
engineering,
work
task
wise,
that's
for
engineering
to
manage
and
we're
looking
for
cut
lines
for
product
management,
so
that
I
would
just
kind
of
thrown
it
out
there
that
maybe
that's
one.
G
A
I
think
the
challenge
might
be
in
defining
the
back
end
pieces.
That
could
happen
ahead
of
time
earlier
all
right.
The
front
end
has
been
fairly
well
fleshed
out
with
designs
and
available
in
this
ticket.
So
I
think
it
would
make
the
most
sense
then,
to
create
a
ticket
that
represents
the
details
of
the
backend
work
to
make
that
happen.
So.
A
Again,
we're
treading
the
new
territories
here,
so
I
don't
want
to
between
this
and
grooming
and
I
want
to
make
sure
that
we're
being
respectful
of
everyone's
time,
because
I
know
we're
still
executing
on
the
the
current
release.
But
I'm
kind
of
looking
at
Lucas
is
the
person
who
just
gave
this
great
explanation
of
what
can
happen
in
advance
and
the
back
end.
Could
you
maybe
start
an
issue
on
that
and
we
can
sure
and
take
that
occurring
for
tonight?
Yeah.
A
Everyone
to
think
times
a
mistake
that
we're
I
don't
think
we're
expecting
to
have
everything
completely
grooms
coming
out
of
this
meeting.
It's
really
just
that
conversation
of
are
all
the
big
questions
answered.
Have
we
broken
these
down
into
small
pieces
as
we
can
and
we
understand
the
boundaries
of
working
or
we
have
to
be
working
in?
You
know
totally
different
areas
of
the
the
product
that
we're
used
to
or
new
technologies
that
we're
not
very
familiar
with
so
I
think
we've
achieved
at
least
the
breakdown
version
of
this
question.
A
A
So
Mir
this
one's
already
been
assigned
to
you
somehow
I'm,
not
sure,
and
it's
already
been
back
and
weighted,
so
this
is
I'm
guessing
one.
It's
been
around
for
a
year,
so
I'm
assuming
we've
talked
this
one
into
the
ground.
At
this
point
anyone
wanna
so,
let's
start
say,
would
you
like
to
summarize
to
us
the
intent
of
this
issue?
A
B
A
A
C
A
B
B
B
And
I
need
to
think
of
with
Becca
I,
don't
know
if
this
is
a
hundred
percent
done
or
only
ninety
percent
done
again.
This
wasn't
originally
my
issue,
but
if
you
guys
have
questions,
let's
get
those
apps.
Now,
if
it's
not
ready
to
change
status,
that's
okay!
I
can
take
any
action
items
and
incorporate
them
back
into
the
issue.
G
A
I
Sexual
about
discovered
Lucas
did
a
lot
of
on
the
back
end
of
the
research
process
initially,
so
centralizer
is
a
way
to
inject
a
certain
annotation
into
the
engrossed
definition
in
Cuban,
Edison,
Phoenix
and
grace.
By
doing
that,
we
can
define
a
custom
what
security
rules,
and
they
already
partially
using
this
notation
in
our
water
deployment
process-
is
the
way
how
they
enable
mod
security
in
the
first
place
for
the
naughty
boy
helm
charts.
I
So
this
work
actually
was
reduced
to
just
changing
helm,
charts
to
add
ability
to
inject
custom
mode
security
rules
for
the
notation.
So
it's
several
related
tasks,
the
several
related
tasks
to
that
specifically
cross
validation
and
tracking
of
the
statistics.
But
this
was
extracting
two
separate
issues
because
on
the
both
ends
we
have
some
complications.
So
this
particular
issue
was
just
about
annotation
and
ability
to
add
custom
rules,
no
UI
involved
as
far
as
as
far
as
I
understand.
A
I
No,
this
is
something
that
way
it
touched
while
discuss
the
police
issue
because
of
the
change
of
the
direction
and
restoration
team.
Certain
things
there
are
two
different
ways
we
can
deploy
certain
things
like
in
graceful,
for
example,
and
the
new
ways.
The
way
that
demonstration
team
is
pushing
as
proper
direction,
which
is
a
GUID
based
deployment,
fall
called
cost
application.
It
has
some
limitations
and
specifically
state
control
of
the
applications,
and
this
blocks
heavily
any
kind
of
UI
right
now.
I
A
A
I
They
were
created
two
issues.
Front-End
issues
not
but
I,
had
an
assumption
that
your
ex
work
was
blocked
by
that
and
it
stopped.
But
I
just
checked
the
issue.
Apparently
your
ex
discovery
issues
that
we
just
share.
It
was
successfully
completed
so
but
I
don't
know
what,
because,
as
I
understand,
you
I
work
supposed
to
help
enough
to
successfully
expand
mocks
already,
but
on
that
issue
there
is
commend
that
they
will
be
tested.
First,
white
Marx's
himself
will
be
tested.
A
Okay,
this
is
an
interesting
situation
we
can
I
can
make
if
you
work
with
Sam
to
get
a
UI
issue
created
once
I
have
read
through
intense
amount
of
discussion,
that's
going
on
there
to
make
sure
that
we
don't
drop
that
work
and
that
we
can
keep
some
track
on
the
blocking
is.
Is
this
the
issue?
That's
blocking
it
now.
What
is
the
issue
exactly
or
is
there
an
issue
that
is
linked
here?
That
is
what
we'd
be
looking
for
to
know.
If
we're
not
aware
of
honor
block
on
the
UI
I
need.
A
C
A
So
the
backend
work
that
Arthur
is
describing
has
already
been
broken
off
and
we
have
smaller
issues
reflecting
the
backend
work,
that's
been
completed
and,
and
so
needs
to
be
done
and
you're
saying
put
those
under
this
epic
or
is
all
of
that
work?
That's
already
happened
from
the
back
end
represented
here
in
this
issue.
I
do.
C
A
E
E
You
see
the
proposed
under
proposal
that
is
like
mini
moon
and
it's
kind
of
the
back
end
part,
and
then
you
have
like
next
I,
don't
know
if
next
is
kind
of
next
issue
and
then
that
would
include
they
use
the
wizard
with
the
UI
properly.
That's
what
I
got
from
this
okay,
so
the
back
end
part
de
Mars,
there's
two
pending
in
Mars,
one,
the
commutation
and
one
for
one
of
them
modules
that
we
handle
and
then
the
UI
part
would
be
the
only
part
left,
but
it
goes
against.
E
I
I
I
A
We
will
look
for
the
exist,
the
front,
end
tickets
and
if
it
doesn't
exist,
we'll
create
it,
and
we
can
talk
about
in
future
funding
breakdown
meetings.
I
guess
my
big
question
is:
do
we
need
an
additional
issue
created
for
any
remaining
work?
That
needs
to
be
done
off
of
this
going
into
the
future?
A
I
A
Okay,
so
we
didn't
get
to
the
third
issue
on
our
list
today.
You
know
I
hope
that
over
time
as
we
can
start
to
create
these
issues
smaller
to
begin
with,
these
conversations
will
go
faster.
I'll,
stick
up
with
with
Sam
about
how
we
want
to
handle
the
the
rest
of
the
issues
that
at
least
the
other
one
that
was
on
this
list
and
any
other
ones
that
we
didn't
include
to
talk
about
for
12:9,
and
we
will
reschedule-
or
we
can
talk
about
this
next
week.
A
I
think
there's
some
outstanding
questions
on
how
often
we
should
have
this
meeting.
You
know
the
last
thing
I
want
to
do
is
in
Dane
everyone,
a
too
many
synchronous
meetings,
so
done
is
still
to
be
determined
and
we
will
get
everyone's
feedback
and
just
discuss
it
further,
but
not
on
this
call
right
now.