►
From YouTube: Think Big Release Management #6
Description
Welcome to the 6th Release Management Think Big
A
All
right
so
we're
here
for
our
sixth
think
big
for
release,
management
and
Hana.
You
added
a
couple
of
really
good
items
for
our
agenda.
We'll
spend
this
session
talking
about
deploy
freezes
and
if
we
have
some
time
we'll
go
ahead
and
dive
into
some
of
the
feedback
that
we've
received
on
our
prototypes.
So
I'll,
let
you
do
the
voiceover
and
it
started
awesome.
B
I
just
wanted
to
share
some
of
the
the
findings
and
discussion
points
that
the
Shah
and
I
went
through
last
Friday,
so
I
had
to
hear
the
notes
and
also
the
recording
and
I
pretty
much
were
disgusting.
You,
the
whole
deploy
phrases,
teams
and
VCS
is,
and
it
was
awesome
because
Sean
just
gave
me
a
lot
of
insight
from
a
back-end
point
of
view.
That
will
definitely
impact
the
user
experience
and
the
in
the
front-end.
So
Meghan
is
not
here
yet
okay,
but
I'm
already
discussing
some
of
those
things
with
him.
B
A
sink
and
I
gotta
hear
some
of
the
learnings.
The
first
one
is
that
yeah,
we
can
definitely
add
a
time
zone
to
the
mo
file
and
also
represent
this
to
the
user.
You.
Why?
Because
I
had
a
question
related
to
that.
So
if
you
set
a
deploy
freeze
in
a
period
well,
how
can
we
communicate
this
to
users?
They
are
other
side
of
the
of
the
world
and
you
want
to
hear
Jackie
doing
a
voice.
This
about
the
embassy
yeah.
B
C
So
sir
I
guess
with
I
mean
Clark
Ron's
really
designed,
for
you
know
server
management
from
the
perspective
of
a
server,
and
so
so
you
can
put
it
so
I
guess.
Some
of
the
questions
we
had
were
around
you
know
blocking
out
holidays,
for
example,
am
I
getting
too
ahead,
but
because
you
can
do
that
in
a
con,
but
it's
you
know
you
have
to
go
and
kind
of
to
be
tricky
to
put
it
in
there,
but
as
an
MVC
we
can
certainly
support
it.
C
I
guess
they
I
guess
you
know
NB
C.
We
can
also
support
time
zones.
No
I,
don't
see
any
problem
with
her
in
the
mo
I
guess.
The
issue
would
be
proposed
in
BC
is,
like
so
say,
you've
got.
You
know,
an
organization
has
got
you
know
10
or
50
or
100.
You
know
projects
and,
and
they
want
to
go
and
put
in
a
holiday
into
every
project.
A
C
A
Allowing
that
to
be
like
the
entry
point
for
then
somebody
to
we
could
say
as
an
NBC
that
this
is
at
least
a
linking
options.
So
you
can
then
link
to
your
projects,
repo
and
change
this
yellow
file,
rather
than
like
uploading,
a
yellow
file
at
the
group
and
having
that
trickle
down,
because
I
feel
like
that
option.
Like
instrumenting
global
blackout
periods
at
the
group
level,
would
be
across
a
lot
of
stages
and
I.
C
Well,
in
fact,
I
wasn't
that's
a
good
point
when
you
know
I
wasn't
thinking
about
having
again
father
got
pushed
down,
I
was
thinking
about
kind
of
having
a
combination,
so
the
mo
files
inside
the
project
and
then
somewhere
else,
the
pipeline
checks
am
I
allowed
to
run.
You
know
some
type
of
master
calendar
and
that's
all
that
that
came
out
of
the
feedback
from
the
infrastructure.
People
were
saying
they
have
this
need.
You
know
they.
C
A
C
C
Well,
we'll
say
a
date:
they
would
be
a
good
starting
point,
because
so
we
have
a
high
volume
application,
but
we
don't
have
a
lot
of
applications.
So
that's
actually
you
know
we're
just
being
used
and
you
know
in
everything
that's
great,
but
we
don't
have
the
problem
of
trying
to
implement
14
or
August
kind
of
10,
but
we
don't
have
100
projects,
yeah
and
so
yeah.
I
think
that
and
I
seem
to
be
pretty
interested
in
this
feature
and
in
fact,
they've
kind
of
written
some
stuff
themselves.
C
A
Hi
Nathan
how's
it
going
where
just
to
recap
to
bring
you
up
to
where
we're
talking
about
we're
kind
of
honing
in
homing
in
on
what
is
the
first
MVC
going
to
include
in
the
amel
file
and
making
sure
that
the
issue
description
reflects
like
the
smallest
change
that
will
make
so.
The
first
kind
of
topic
that
were
iterating
on
right
now
together
is
including
in
the
ML
file
time
zones,
so
that
if
someone
is
in
a
global
team,
they
see
the
blackout
period
in
their
time
zone
rather
than
in.
A
D
C
They're
also,
we
high
in
a
we
were
talking
also
about
how
like
what
actually
are.
We
are,
we
blacking
out
exactly
and
we're
blacking
out
a
release
only
to
production,
and
then
how
do
we
define
production?
Can
we,
if
there's
a
problem
with
production,
do
we
have
some
way
of
overriding
it
does?
How
does
it
affect
the
existing
deployments.
A
A
Environment
so
it'll
be
all
environments,
but
the
very
exciting
and
interesting
part
is:
what
do
we
do
if
there's
a
production
failure
and
it's
a
deploy
freeze,
so
we
would
need
to
like
how
do
you
auto
do
a
rollback
or
how
do
you?
How
can
we
support
automatic
role,
blacks,
regardless
of
deployed
freezes?
I
think
is
probably
the
answer.
So
it's
like
investing
further
in
rollbacks
so
that
the
owner
or
administrator
can
override
the
deploy.
Freeze
and
really,
you
know,
redeploy
the
last
successful
release
or
the
last
successful
pipeline
yeah.
A
D
One
thought
I'd
said:
while
you
were
talking
you're
talking
about
how
it's
gonna
be
for
all
environments
and
I
guess
this
goes
back
to
my
question
about
how
this
would
look
at
the
animal,
because
I'm
kind
of
envisioning
that
each
environment
specification
would
have
a
sub.
He
called
blackout
dates,
and
so
you
could
kind
of
configure
it
on
a
per
environment
basis.
But
maybe
that's
were
you
thinking
more
of
a
global
blackout
that
would
automatically
apply
to
all
environments.
C
Yeah,
that's
a
good
point:
I
didn't
I
hadn't
thought
about
about
the
at
the
environment
level
like
that
that
it's
more
and
it'd
be
more
intrigued
to
do,
but
well,
we
can
kind
of
have
it
at
both
levels
like
have
it.
You
know
for
the
yam
or
fall
as
a
whole
and
then
potentially
have
you
know
entries
at
the
at
the
at
the
lower
level
that
over
either
higher
level
I
don't
know
yeah.
D
And
I
was
wondering
if
it
even
needs
to
be
at
the
job
level,
because
right
now
we
might
call
a
job
deploy
right,
which
will
you
know?
Association
is
a
server
and
and
push
them
files,
but
get
land
itself
doesn't
really
know
that
that's
actually
a
deployment.
It
doesn't
really
distinguish
Mena
testing
job
and
a
deployment
job,
so
I
think
it
might
have
to
be
at
the
job
level,
where
you
mark
certain
jobs
to
say
this
job
can't
execute
during
the
deploy
freeze,
mm-hmm.
D
C
E
A
Like
the
documentation
should
be,
if
you
set
it
to
ploy,
freeze
you'll
need
to
go
back
in
there
at
the
end
of
your
period
and
retry
you're
missed
I'm,
not
sure.
If
that's
what,
if
we
should
have
like
notifications,
that
say,
hey
your
this
has
been.
You
know
this
was
skipped
like
it's
a
to.
Do
that
then
create
that's,
then
created
you
might
want
to
consider.
How
that
how
that
looks,
hi.
C
B
Exactly
so,
I
looked
into
the
documentation
to
see
where
deployments
are
displayed
to
users,
and
we
have
this
in
the
merge
request.
The
deployments
view
the
deployments
detailed
view.
This
did
not
so
my
suggestion.
The
embassy
is
that
whatever
we
can
identify
that
a
job
was
skipped
due
to
freeze
the
update
information
instead
of
just
saying,
skip
to
skip
because
of
deployment
fees,
but
also
the
merge
request.
B
We
add
the
information
about
when
the
deployment
freeze
kind
of
like
when
it's
a
yeah
when
the
phrase
is
a
currently
active
and
then
afterwards
change
the
message
so
that
the
user
you
know
knows
that
they
have
to
manually,
go
there
and
retry
that
one
step,
because
right
now
yeah
we
can
definitely
create
a
to-do,
but
with
that,
for
example,
any
military
users
would
generate
a
task
to
do.
You
know
if
that
the
the
system
admin
is
that
whoever
is
assigned
to
a
specific
merge
request,
but
they
get
in
because.
A
I,
don't
think
it'll
be
worth
doing
it
to
do
for
this,
because
I'm
pretty
sure
we'll
fast
follow
with
a
retry
automatically
feature
we're
like
once
it
ends
we'll
run
the
pipeline
again
or
I.
Guess
the
job
again
right,
so
is
that
what
we're
gonna
call
this
now
that
it's
a
blackout
period
or
a
deploy,
freeze,
/
job
like
how
do
we
want
to
or
is
it
you.
A
C
Like
hair
periods,
more
general
right,
it's
just
like
we're
just
not
going
to
basically
execute
what's
inside
this
job,
whatever
that
might
be
because
he
again
Hitler
doesn't
really
know
whether
it's
doing
a
deploy
or
not
specifically,
it
could
be
anything
could
be
just
some
cute
control
scripts
or
something
in
any
stage.
So.
C
A
C
C
E
A
C
A
C
A
We
interviewed
a
customer
about
this.
They
gave
me
two
examples
of
deploy
phrases
that
they
use
the
first
one
and
they
call
them
moratoriums.
So
this
first
customer
said
when
I
am
redoing
my
QA
environment,
because
apparently
they
have
to
pull
a
copy
down
from
production,
anonymize
the
data,
and
they
don't
want
people
to
playing
that
staging
environment.
A
So
they
put
a
moratorium
on
that
and
they
let
people
know
that
they
have
to
spin
up
their
own
feature
branch
in
order
to
test
their
code
and
then
they
can
deploy
to
the
test
environment
once
it's
available.
The
second
use
case
is
once
they
have
like
a
company
event
which
they
would
have
weekly
company
events.
They
employed
a
moratorium
on
Friday
to
Monday
what
people
would
be
deploying
all
through
staging,
which
would
be.
C
A
C
A
D
Yeah
we
might
want
to
I
just
thinking
through
all
the
different
ways
and
that
we
could
do
that
in
one
implication
of
doing
it.
That
way,
which
I
think
makes
most
sense
to
me.
But
now
we
could
theoretically
have
a
pipeline
that
has
multiple
different
deploy.
Freezes
like
one
job,
has
a
certain
deploy
freeze.
Another
one
has
a
different
one
like
it's
a
edge
case,
but
it
would
make
it
hard
to
show
in
the
UI
like
determine
what
is
this
pipelines?
D
D
I'd
have
to
look
into
that,
so
that
was
another
idea.
Another
a
third
idea
I
had
to
is
that
we
could
maybe
tie
into
only
except
those
are
like
conditions
that
a
job
is
run
and
maybe
right
now
we
already
have
some
special
values
like
merge
requests.
You
can
say
only
run
this
job
for
merge
requests,
so
we
could
say
accept,
run
this
job
except
if
there's
a
current
deploy,
freeze
and
then
that
deploy
freeze
could
be
defined,
maybe
globally
or
even
outside
the
amyl.
C
E
Think
that
could
be
super
powerful
if,
because
I
shot,
I
think
you
were
matching
this,
but
an
example
where
there
may
be
like
an
emergency
where
you
need
to
like,
let's
blackout
right
now,
folks
right
away
if
they
had
to
go
edit.
This
yeah
well
file
to
do
that.
That
would
be
kind
of
a
bummer
but
I
think
the
dashboard
you're
talking
about
plus
and
maybe
Nathan's
idea
of
just
adding,
except
if
and
then
it
can
go
check
this
kind
of
settings
area
right.
D
A
So
I
wasn't
following
everything
in
that,
because
my
animal
knowledge
is
not
like
super
clean,
so
I
might
need
help
with
updating
the
actual
ish
description
if
the
requirements
are
different
than
what
we're
saying,
because
reconciling
the
issue
right
now,
it
sounds
like
we
should
make
the
update,
for.
We
want
to
be
able
to
specify
time
zones
but
I
think,
like
that's
just
more
like
technical
proposal,
details
that
we
should
just
add,
but
it's
out
of
like
the
acceptance
criteria,
we
have
today
kind
of
Maps
exactly
what
you
guys
are
saying.
C
A
Okay,
with
that
yeah
I
think
as
long
as
they
ask
is
we
want
people
to
be
to
set
a
deployment
freeze
for
any
given
environment.
So
that's
what
that's
where
our
customers
have
asked
for,
but
we've
just
discussed
that
environment.
It's
kind
of
a
tricky
word,
because
the
environment
doesn't
necessarily
have
the
rules
or
the
logic
that
would
prevent
a
deployment.
Yeah.
C
A
C
And
so
in
so
from
the
UI
perspective,
Nathan
I
guess
what
we
will
need
to
do
is
I
guess
we
need
to
make
sure
all
this
is
exposed
to
the
API
and
then,
whenever
the
page
is,
is
loaded,
they'll
have
to
be
a
call
to
check
the
the
amyl
enough.
That
will
be
expensive,
that
there
would
be
the
simplest
way
and
then
put
this
banner
or
whatever
it's
kind
of
very
saying.
There's
a
blackout
period.
D
Yeah,
that's
the
one
difficulty
about
storing
the
blackened
information
in
the
animal
yeah
is
that
we
then
have
to
reference
it
just
to
display
in
the
UI,
which
is
a
little
odd
and
possibly
more
the
more
I
think
about
it.
I
kind
of
liked
it
yeah
yeah,
the
more
I
think
about
the
I
kind
of
like
the
idea
of
storing
the
blackout,
dates
outside
and
then
just
referencing
them
from
within
yeah.
D
A
I
think
this
is
where,
like
originally
Nathan,
we
only
have
this
conversation
with
the
UX
experience,
which
is
like
hey.
People
are
used
to
specifying
this
in
the
mo
fire.
We'll
see.
Icd
is
all
in
the
ammo
file,
but
now
we're
learning
that
well
from
a
back-end
perspective,
it's
going
to
be
perfectly
be
as
performant
as
it
could
be.
If
we
store
this
outside
of
the
and
will
file
so
I,
don't
I
think
it's
okay!
Well,.
C
We
do
have
some
precedents
as
well
because
like,
for
example,
if
this
simple
I'm
just
thinking,
what
would
the
simplest
version
of
this
be
and
would
be,
maybe
add
another
one
of
those
accordions
in
the
configuration
they
have.
You
know
like
a
period
or
something-
and
you
know,
and
just
have
a
simple.
You
know
relatively
simple
form
in
there
and
because
if
you
look,
for
example,
a
just
like
merge
strains,
for
example,
you
know
that
behavior
is
a
lot
of
the
a
lot
of
the
behavior
of
the
CI
is
not
actually
only
in
the
amal.
C
C
A
A
C
Because
because
there'd
cuz,
the
forum
would
because
the
forum
has
to
have
the
form
could
be,
you
know
relatively
complicated
right,
because
you
know
you
gotta
have
repeating
dates
and
then
times
within
those
dates.
So
well,
you
might
just
have
okay,
you
might
just
have
times
because
you
might
have
a
blackout
period.
You
know
every
day
from
7:00
p.m.
to
8:00
in
the
morning
or
something.
But
then
so
you
know
regular
chrome
stuff,
but
then
what
we
could
just
use
the
cron
form
every
go
field
in
this
book.
C
You
know
just
put
cron
format
in
there
and
then
we
also
need
to
incorporate
dates
and,
and
so
and
then
you
might
have
repeating
groups
right
and
in
all
that
would
roll
up
to
one
setting
one
name
to
setting
holiday
calendar
or
something.
So
so
relatively
you
know
you
do
it
just
as
a
rails
forum,
but
it'd
be
relatively
complicated
form.
You
know
two
or
three
levels.
D
Too,
we
also
could,
if
we
were
interested
in
simplifying
it,
might
make
sense
to
not
do
the
cron
part
as
part
one.
We
could
even
just
have
like
a
more
of
a
calendar
view
or,
like
date,
pickers
oh
yeah
yeah.
That
would
be
less
useful
because
you
have
to
go
in
you'd
have
to
like
update
a
year's
worth
of
blackout,
dates
and
then
come
back
and
keep
doing
it,
but
that
might
be
a
short-term
solution.
You
know
that.
B
A
When
I
spoke
to
users,
this
is
a
feature.
That's
pretty
like
60%
are
the
technical
team.
The
build
managers
that
are
implementing,
deploy,
freezes
and
40%
have
to
interact
the
deploy
freezes.
So
this
is
one
of
our
first
feature
sets
where
we
start
noticing
the
bifurcation
of
get
labs
users
being
front-end
dominant
versus
wanting
to
use
everything
in
the
amel
file.
But
since
the
split
is
still
fairly,
technical,
I
am
almost
leaning
toward.
We
should
do
the
investment
in
the
animal
file
first,
but
I'm
open
to
to
your
guys's
thoughts.
A
C
C
Nbc
or
you
know,
spec
or
anything,
but
oh
I
mean
we
could
even
do
a
wireframe,
maybe
because
yeah
when
we
start
to
think
about
dates
and
times
it
can
get
complicated
right
when
it
comes
to
forms
and
yeah
just
to
make
sure
we're
not
missing
something
because
I
think.
If
we,
if
we
go
one
direction,
it
will
probably
end
up
saying
on
that
direction.
A
A
Agreed
that
that
would
be
a
really
great
option,
but
we
didn't
have
like
we
didn't
build
out
mocks
and
we
didn't
put
mocks
in
front
of
them.
This
is
more
like
hey,
conceptually
would
having
a
yellow
file
configuration
satisfy
your
need
for
being
able
to
set
deployment
freeze,
but
when
we
look
at
the
second
and
third
iterations
I
thought
that
we'd
be
expanding
more
into
the
UI
anyways,
so
so.
C
I
read
somewhere
and
I,
it
was
kind
of
when
I
was
onboarding
until
I
can't
quite
remember
where
it
was
there.
At
one
stage
there
was
a
the
gamma
file
was
a
UI
or
there
was
that
you
know
it
was,
and
so
so
I
guess
with.
Maybe
we
should
I
should
try
and
find
that
it,
because
that
way,
you
know
why
did
we
make
the
decision
to
go
all
the
yeah
more
you
know
and
and
away
from
from
pages
you
know,
there's
probably
flexibility
but
yeah
I'd.
D
I
think,
regardless
it's
probably
worth
writing
down
all
these
proposals
like
how,
because
I
think
there's
two
distinct
way.
It's
one
is
kind
of
what
we
can
talk
about
where
you
configure
the
dates
outside
and
then
reference
it
in
the
animal
or
we
do
everything
in
the
amyl.
We
include
the
dates
in
the
animal
and
then
the
question
would
be
like:
where
did
we
put
that
it
probably
is
worth
digging
a
little
further
into
this
issue
and
actually
putting
all
those
proposals
down
just
so,
we
can
think
through
them
and
yeah.
C
B
C
A
Because
I
feel,
like
our
current
proposed
a
little
high
on
I,
like
what
you
have
built
out
in
our
prototypes,
satisfies
what
our
validation
confirmed
like
as
the
needs
and
the
problem
that
we
were
solving.
But
I
do
think
that
we
were
looking
at
just
a
sliver
of
what
customers
are
using
things
like
zvo
labs
to
do
with
deploy
freezes
so
like
the
biggest
use
case
for
moratoriums
or
deploy
freezes,
were
reoccurring
times
or
holidays.
A
C
A
You
just
talked
yourself
out
of
any
voice
over
that
someone
had
to
do.
You're
like
that
that
we're
just
gonna
talk
about
it.
So
when
I
look
at
like
the
things
that
we
currently
have
today,
hi
Anna,
it
looks
like
it's
just
the
date:
picker
right
from
an
MVC
side,
right,
okay
in
the
front
end,
and
that
date,
picker
was
going
to
reuse,
a
component
that
we
already
have
available
to
get
lab
library
right
and
then
it
was
going
to
be
fed
from
the
amal
file
like
the
date
specified.
Is
that
correct?
D
C
D
D
B
A
Yeah,
the
actual
MVC
issue
didn't
have
a
date
picker
in
it.
Okay,
we're
at
time
but
I
think
if
we
can
iterate
on
the
MVC
issue.
I
would
like
to
get
this
buttoned
up
before
we
kick
off
12.10.
Obviously,
because
we're
supposed
to
be
shipping
it
then
so,
let's,
let's
hammer
down
on
what
makes
sense
and
I'm
inclined.
Do
you
do
what's
the
lowest
hanging
that
enables
our
users
to
have
deploy
freezes
and
get
lab
so
whatever?
A
Whatever
you
guys
think,
is
the
best
technical
implementation
for
that
I'm
happy
to
make
sure
the
requirements
meet
that
need.
So
just
let
me
know
what
I
can
do
to
help
make
that
clear
inside
the
description,
but
I
really
look
forward
to
hearing
more
about
your
proposals
and
different
options,
and
let
me
know
how
high
on
and
I
can
help
okay.