►
From YouTube: Release:Release Management Think Big #12
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
So
welcome
to
our
is
it
13th
or
12th
12th.
Think
big.
There
you
go.
Hannah
has
a
numbering
right
from
our
last
meeting,
just
as
a
recap,
we
dove
into
a
couple
of
outstanding
items
around
the
release
pages
use
case
and
how
we're
trying
to
solve
this
idea
of
distributing
binaries
with
the
releases
page.
This
topic
we're
going
to
uncover
a
little
bit
more
information
around
the
user
story
of
managing
deployments
at
scale
across
multiple
projects.
So
a
couple
of
issues
that
came
up
during
our
analysts
review
of
continuous
delivery
and
release
automation.
A
It's
this
idea
of
managing
deployments
a
scale
so
with
thousands
of
environments
over
hundreds
of
different
target
production
environments
and
today
get
lab
if
you're,
not
using
a
deploy
board
with
multiple
projects
can
get
a
little
unwieldy.
So
for
people
who
are
using
traditional
or
legacy
style
deployments,
release
management
isn't
particularly
friendly
to
check
the
health
of
that.
It's
not
particularly
helpful
if
people
are
looking
to
approve
a
global
release.
A
Everything
is
very
much
about
per
merge
request
per
pipeline
per
environment,
so
giving
our
personas
like
release,
managers
and
development
team
leads
that
are
creating
builds
and
creating
production
releases.
The
tools
to
manage
a
scale
is
really
important,
so
some
of
the
things
that
hi,
Anna
and
I
are
working
on
invalidation
cycles
are
like
door
for
metrics,
which
are
the
four
metrics
that
are
industry
standard
for
deployment
tracking,
but
also
the
CI
CD
director
dashboard,
which
is
about
confronting
zbo
labs
as
value
stream
metrics.
A
But
there's
other
kinds
of
details
that
we
could
irritably
consider
adding
to
get
lab
to
make
this
journey
a
little
bit
more
accessible
for
our
people,
so
just
wanted
to
kind
of
set
the
context
of
what
we're
trying
to
accomplish,
which
is
tracking
deployments
at
scale
and
get
lab,
and
let
you
all
sort
of
talk
through
what
you
think
would
be
a
good
way
to
solve
that
problem.
We're
thinking
big,
so
No
Limits
here.
Let's
just
just
talk
about
it,.
B
A
Most
likely
target
are
people
who
are
trying
to
manage
hundreds
of
production
environments.
So
if
we
think
about
the
largest
companies
that
have
a
production,
microservice
environment,
that
then
feeds
into
a
customer
facing
experience
in
a
different
kind
of
like
trajectory,
they
may
have
hundreds
of
production,
protected
global
environments
and.
B
A
Filtering
so
I
would
say
as
an
MVC,
we
may
only
display
production
environments
because
a
release
manager
is
going
to
potentially
be
focus
just
on
the
end
state
and
looking
at
the
health
of
my
current.
But
then
the
future
iterations
could
be
being
able
to
include
other
lower
level
environments.
Okay,.
B
Yeah
yeah
I
mean
I
think
this
is
such
an
awesome
picture.
I
think
it's
really
gonna
shoot
us
out
of
the
ballpark,
and
so
the
other
thing
I
guess
I
was
thinking
about.
Is
you
know
we
have
all
this
instrumentation
and
logging,
and
you
know
Prometheus
and
all
this
stuff,
and
maybe
we
not
after
the
initial
MVC
you
might
consider
you
know
how
can
we
link
from
that
environment?
You
know
on
the
dashboard
to
to
get
all
this
other
instrumentation.
They
were
already
doing
easily,
which
of
course
takes
us
outside
of
release
management.
A
That
you
brought
that
up
there's
this
feature
that
Ori
is
working
on
right
now,
calling
post-deployment
monitoring.
So
let
me
I
just
copy
pasted
from
a
weird
weird
link.
Sorry,
so
this
feature
that
she's
working
on
is
about
attaching
alerts
from
monitoring
inside
of
the
environments
page,
so
she's
kind
of
designed
this
independently
of
what
we're
doing
so
it'll
give
us
an
opportunity
to
kind
of
look
back
and
think
about.
How
do
we
extend
this?
A
It's
almost
like
a
myopic
focus
on
deployed
words,
alerting
and
apply
it
more
toward
the
new
environment
dashboard,
but
you're
absolutely
right.
There
is
an
opportunity
to
create
a
beautiful
loop
between
release
and
monitor
in
a
more
thoughtful
way
and
I.
Don't
know
if
you
all
have
noticed,
but
Sid
created
this
video
about
the
at
lassi,
an
integration
with
Ops
Genie.
There
might
be
an
opportunity
for
us
to
create,
like
a
web
hook
on
deployments
in
to
ops
Jeannie
and
have
that
be
an
integration
path
of
release
to
Atlassian.
B
B
You
know,
we've
now
got
elasticsearch
as
a
managed
component,
and
you
can
see
down
the
track
that
you
know
these
the
we're
now,
almost
in
a
space
where
we
compete
with
like
data
dog
or
you
know,
we
must
like
be
able
to
compete
with
something
like
that
as
part
of
the
full
stack,
which
would
be
I
mean
that's
a
lot
further
down
the
track.
Obviously,
but
we
we
do
have
the
pieces
already
in
place.
We
just
kind
of
don't
have
a
holistic
view,
and
maybe
this
dashboard
is
a
place
for
that
you're.
A
Reading
my
mind:
I
love
whenever,
whenever
I
hear
these
affirming
words
of
like
the
things
I'm
thinking
and
whenever
you
say
it,
because
it
just
shows
that
like
yes,
this
is
a
need
that
we
should
solve,
because
you're
totally
right,
like
being
able
to
you
almost
do
like
a
data
dog
light,
that's
cheap,
yeah,
it's
what
the
market
means,
because
everyone
is
like.
Okay,
the
dog
is
so
expensive.
My
mice
teammates
are
driving
hundreds
of
thousands
of
consumption
dollars
because
they
are
which.
B
A
B
A
So
if
we
look
at
alerts,
I
think
that
that's
catering
to
the
premium
and
ultimate
use
case,
and
then,
of
course,
if
we
think
about
production
environments,
most
of
our
customers
are
going
to
try
to
use
production
with
protected
environments.
Because
that
way
they
can
at
least
restrict
you
can
deploy
to
it
yeah.
But
if
we
add
that
reporter
plus
approval
to
deploy
permission
set,
that
makes
this
a
little
bit
more
accessible
to
production
environments,
not
necessarily
having
to
be
protected,
because
it
could
just
be
that
the
foyer
only
has
permissions
to
that
particular
environment.
A
B
A
B
B
A
B
And
just
also
thinking
just
you
know
the
dashboard
concept,
and
so
again,
maybe
not
an
analyst
in
viously
that
we
also
consider
okay,
you
know
causing
that
environment
or
you
know
restarting,
you
know
having
buttons
to
do
things
like
restart
the
environment
or
you
know
redeploy
it
all.
You
know
things
like
that.
You
wouldn't
normally
maybe
initiated
through
a
pipeline,
but
you
know
you
could
also
perhaps
get
to
from
the
dashboard.
So.
A
C
B
That's
something
that
who
wrote
when
I
first
use
that
that
feature
in
Heroku
I
was
floored.
This
was
so
easy
and
it
works
so
well
and
I've
used
it
many
times
right,
you
just
go.
Bang.
Take
me
back
and
that's
fine
from
migrations
can
be
a
bit
tricky,
but
everything
else
it
can
just
roll
that
instantly,
which
is
pretty
incredible.
Actually,
it's
a
great
production
feature.
A
A
E
A
B
D
A
different
question
about
a
romance:
is
it
possible
to
pin
them
to
a
location
so
basically
using
geolocation,
and
we
could
maybe
like
when
you
have
a
distributed
view
on
the
environments,
use
geographically
map
just
to
have
maybe
a
heat
map
on
the
world
map.
I
often
see
that
in
the
monitoring
space
that
people
just
want
to
see
that,
based
on
a
location
on
the
data
center
on
the
robot,
and
maybe
we
could
just
use
that
as
a
dashboard
or
dashlet
for
the
feuding
I.
A
D
Gu
is
actually
for
the
high
availability
of
github
itself.
I,
don't
know
if
that
directly
applies
to
environments
and
boy
if
we
can
just
used
it.
My
thinking
is
basically
to
have,
since,
since
we're
talking
about
larger
systems
and
larger
customer
systems,
to
have
a
get
over
episode
that
you
may
be
at
the
location
to
an
environment
is
a
variable
or
setting,
maybe
even
as
a
CI
extension,
and
then.
A
B
A
Yeah
I
think
MVC
would
probably
be
specified.
You
know,
add
an
environment
attribute
to
specify
the
location
of
target
environment,
and
then
we
can
kind
of
look
at
if
there's
rules
that
we
can
look
at
the
URLs
specified.
So,
for
example,
if
it
does
have
like
a
region
indicated
in
it,
I
will
need
to
validate.
A
If
this
is
something
that
would
move
the
needle
for
people
who
are
monitoring
and
get
lapped
today
or
not
so,
for
example,
if
location
isn't
as
relevant
because
they're
leaving
gitlab
to
go
to
AWS
or
GCP
to
look
at
the
the
region,
then
that's
a
great
conversion
opportunity
to
keep
people
and
get
lab
right.
So
it'll
incentivize
them
because
we're
offering
this
feature
in
get
lab.
But
if
people
are
not
leveraging
get
lab
to
even
deploy
to
multiple
regions,
then
that
might
be
a
feature
that
could
could
not
make.
B
It
I
think
I'm
gonna,
take
nickel
standpoint,
just
thinking
how
we
would
do
it.
It
probably
wouldn't
be
that
complicated
right.
We
just
have
a
map
and
then
we'd
somehow
label
the
environments,
whether
it's
done
by
the
data
center
or
some
other
way,
and
then
you
know
we
just
plot
them
on
the
map
and
it
would
it
would
look
pretty
you
know
it
would
have
the
the
bells
and
whistles
in
terms
of
showing
it
to
customers
and
so
I
definitely
take
your
point.
Jackie
like
it
would
only
be
super
large
customers.
A
I
have
two
or
three
in
my
mind
already
that
will
be
worth
talking
about
this
with
so,
for
example,
the
ones
who
want
to
manage
environments
that
scale
to
begin
with
the
filtering
that
they've
immediately
talked
about
are
just
by
target
URL
by
stage
and
by
project.
So
those
were
like
the
top
three
things
that
they've
indicated
as
wanting
if
I
was
to
suggest
region.
That
might
be
something
that
they
would
all
jump
at
right,
but
it
might
be
like
a
delight
feature,
so
we
may
prioritize.
A
You
know:
project
filtering
over
location,
mainly
because
today
I
think
we
would
need
a
user
to
specify
the
region
or
location
of
that
environment
and
any
time
where
you
add
a
step
or
add
a
click
or
add
an
action
on
the
user
adoption
decreases.
So
we
want
to
make
things
as
seamless
or
as
auto-detected
as
possible
and
get
lab
to
make
that
usable,
but
I
agree.
This
would
be
something
that
I've
already
added
to
my
agendas.
For
my
my
calls
to
speak
with
customers
that
I'll
bring
up
good
ideas,
Michael
look
at
you.
E
Here
I
remember:
we
had
similar
investigation
when
I
was
at
zebra
labs,
and
that
was
a
highly
mobile
likeable
thing.
So
I
agree
with
you
or
Jackie,
it's
I
remember
it
wasn't
like
the
first
need,
but
I
would
definitely
create
like
a
wow
effect
during
the
demo,
when
we
would
add
that
to
your
like
topology,
you
know
that
what
we
called
it
I
think
so
that
sounds.
A
I
think
about
the
audience
for
that.
It
really
is
like
the
the
monitoring
team,
the
system
admins
and
then
the
buyers
that
want
to
show
their
distribution
powerhouse
when
I
think
about
the
usability
of
the
release
manager,
they're
gonna
care
more
about,
like
a
calendar
view
or
toggling
between
calendar
view,
with
staged
view
like
how
many
environments
are
in
my
dev
test
for
production.
A
Yes,
Oh,
like
that
different
resource
for
you
could
all
be
in
one
plane
and
then
their
different
tabs,
like
I
own
and
I
kind
of
talked
about
this,
like
being
able
to
toggle
between
typography
and
calendar
and
stage
views
would
be
very
compelling
for
the
release
manager,
but
also
the
buyer
and
the
system.
Admin
we'd
be
able
to
cater
to
all
these
users.
In
one
view,.
E
Definitely
and
actually
I
had
the
questions
here
also
and
poured
on
me
if
that's
like
a
little
bit
of
a
weird
question,
but
let
me
ask
that
I
remember
when,
when
I
was
like
part
of
Philips,
we've
been
talking
about
two
different
types
of
like
two
different
ways
of
how
organizations
stat
hopping
is
their
environment.
E
Let
us
see
if
you
want
to
call
that
we've
been
using
more
of
this,
like
I,
think
Graham
environments
like
they
put
managed
manually
something-
and
you
know
it
would
probably
have
like
this
old
way
of
like
old
persona,
again
I'm
like
engineer
an
environment
manager,
something
like
that
helping
to
like
take
care
of
those
like
environments
who
are
usually
are
unlimited
amount.
Comparing
to
like
you
know
that
you
have
the
possibility
to
be
those
across
cloud.
Are
we
talking
about
they
speed?
Here?
A
Think
managing
environments
at
scale
could
cater
to
you
any
sort
of
cloud
target,
but
if
we're
to
consider
like
any
VM
or
server
or
on-prem
device,
that
could
be
even
larger
at
scale.
So
when
we
think
about
a
person
that
has
a
sass
offering
and
they
have
a
thousand
customers
each
with
their
own
instance,
they
would
need
to
track
the
version
of
there's
a
software
solution
or
their
solution
deployed
to
that,
on
instance,
of
the
customer.
More
like
single
tenant
applications
consuming
software,
for
example.
A
That
could
be
thousands
and
thousands
of
different
environments
that
you
would
be
monitoring
as
a
part
of
your
deployment
chain.
A
really
good
example
of
this
when
I
was
a
product
manager
at
a
previous
company.
We
delivered
a
substitute
teacher
management
system
and
it
was
a
part
of
frontline,
was
our
biggest
competitor,
miss
teacher
over
there
Lori,
ok,.
C
E
E
D
A
2,000
customers
k12
school
districts
using
this
solution
and
they
each
had
their
own
set
of
servers
and
duplication
servers.
So
our
instances
were
like
three
or
four
different
ec2
containers,
as
well
as
servers
to
create
backups
to
these
customer
hosted
platforms,
but
we
were
deploying
our
solution
onto
these
servers
and
we
were
monitoring
each
individual.
You
know
container
as
well
as
server
that
a
customer
had
so
that
was
a
you
know
like
a
hybrid
deployment
model,
yeah
and
I.
Think
that
this
would
this
view
we
would
want
customers
to
be
able
to
see.
A
E
Yes,
while
I'm
also
asking
about
that,
because
I
remember
when
we
would
be
talking
to
like
war
like
the
legacy
type
of
organizations
they
for
that
environment
manager
persona,
they
would
literally
need
a
calendar
to
see.
Ok
like
I
have
that
that
physical
environment
busy
until
at
that
time
and
then
we
can
clean
it
up
and
use
it
for
something
else.
So
there
would
be
a
lot
of
questions
around
like
the
availability
resolution
type
of
tasks,
whether
I
think,
if
you're
managing,
if
you're
managing
that
cloud
is
kind
of
like
easier.
E
If
you
need
more
just
spin
up
more
speed
down,
you
know
like
just
based
on
how
much
capacity
you
pay
for
so
that
I
remembered
just
that
was
the
biggest
like
difference
for
us.
When
we've
been
talking
to
those
two
different
sets
of
personas,
that
the
manual
management
would
need
more,
a
much
more
clear
way
like
to
see
okay.
What
do
I
have
today
available
because
it's
almost
they
would
be
all
like
after
working
with
some
like
limited
capacity,
I
would
call
it.
A
That
is
really
that's.
What
I
have
a
couple
of
different
screen
shots
in
our
agenda?
One
of
them
is
a
release
calendar
from
flute
aura,
which
highlights
like
all
the
different
target
environments
and
their
status,
and
then
we
also
have
a
lot
electric
clouds
view,
which
shows
you
the
stage
and
the
different
new
various
attributes
of
that,
whether
it's
like
certain
metrics
or
certain
resources
that
you
would
want
to
pull
in
I
could
see
that
all
of
these
would
be
relevant
for
both
multi
cloud,
but
also
server
on
Prem
deployments.
Yeah.
E
And
my
actually
second
question:
I,
don't
know
if
you
people
are
discussing
this
in
these
things
big,
but
I'm
curious
like
what
like
what
are
the
ways
we
have
today
and
get
love
to
like
to
help
people
manage
their
environments
like
I
know
that
we
have
the
environment
least
view
almost.
We
have.
The
alert
page
gives
you
warnings,
although
under
any
other
places
like
I,
know
any
port
or
anything
like
it
does.
It
sounds
like
we
have.
This
information
is
gathered
over
all
of
the
places
like
yeah
do.
E
A
I
can
tell
you
that
we
have
an
environments
dashboard
that
is
super
restrictive.
We
have
seven
projects
and
three
environments
per
project,
so
it
allows
you
to
see
a
very
limited
subset
of
cross-group,
though
so.
That's
nice,
like
all
of
your
projects
within
a
group,
can
be
pulled
on
here.
If
you
have
seven
projects
with
only
three
environments
per,
but
that
doesn't
give
you
like
the
global
view
of
if
you're
running
a
hundred
target
deployment
perspective.
A
B
A
B
B
A
B
A
B
A
C
No
I'm
not
playing
because
I'm
just
absorbing
all
the
information
and
I
was
actually
thinking
about
like
that
was
a
great
question
from
Nadia.
What
are
the
ways
for
us
to
find
the
information
and
everything's
all
scatter,
and
every
time
we
look
back
at
base
environment,
dashboard
and
I
know
it's.
It
leaves
like
a
higher.
E
C
A
C
C
B
Something
else
they
hire
is,
you
know
you
have
a
project
and
then
they
have
multiple
stages
and
they
actually
they
call
them
a
pipeline,
they
sort
of
say:
oh,
they
have
actually
introduced
a
CI,
but
that's
relatively
new
and
the
submit
that
they
have
this
concept
of
pipeline,
which
is
which
represents
your
your
development
environments
right.
So
you
had
like
a
staging
environment
and
maybe
a
UAT,
and
then
you
have
finally
a
production.
You
know
and
and
that's
where
this
this
button
is
to
do
the
the
rollback.
B
A
That
simplicity,
that
usability
and
nimbleness
is
really
what
we
need
to
give
to
these
super
complex
and
a
scale
organizations
so
I
love
that
we
keep
talking
about
like
Heroku
and
make
and
how
easy
it
is
like.
We
need
to
recreate
that
easiness
for
enterprises,
because
there
isn't
a
good
solution
in
the
market.
Today,
that's
been
a
currency.
A
To
use
z-bo
lab,
do
you
have
to
do
a
bunch
of
setup
and
it's
hard
to
set
up
I?
Had
it
I
had
to
use
a
Raspberry
Pi
to
install
I
use
my
ZB
labs
instance,
because
it
wouldn't
set
up
on
my
Mac.
It
was
ridiculous,
so
I
I
I
think
that
if
we
can
do
this
and
get
lab
comm,
our
biggest
customers
are
currently
on
get
lab.
Comm
would
drop
three
or
four
other
solutions
for
this
week.
A
A
huge
existing
customer
seed
expansion,
not
to
mention
the
ease
of
scalability
here
like
a
prospect,
would
be
floored
to
know
that
they
can
come
in
here
and
see
the
visibility
that
Jenkins
gave
them
before
and
be
able
to
plan
and
meet
the
visibility
and
transparency
for
the
other
other
personas.
It
can
be
really
helpful.
B
Yeah,
because
what
one
thing
to
get
live
does
really
well
is
any
tea
lights.
Use
I
mean
we
have.
You
know
kubernetes.
We
interact
directly
with
Google
cloud
and
although
it's
in
your
back
you're
basically
insulated
from
their
stuff,
because
it's
so
complex
and
you
know,
and
and
so
if
we
can
add,
even
as
I
say
in
Jackie,
even
another
layer
of
simplicity
over
the
top,
along
with
some
awesome
design,
because
I
think
that's.
That
is
something
that
he
Riku
has
right.
B
That,
for
example,
earlier
beers
does
not
have
right,
which
is
design
just
you
know,
and
and
I
think
that
does
make
a
difference.
You
know
I
think
I
think
we're
in
the
middle
of
those
two
of
those
two
things.
You
know
we're
certainly
superior
to
something
like
AWS,
but
you
know
I
think
some
of
the
other
products
are
so
polished
and
they
just
look
so
incredible.
You
know
you
just
want
to
look
at
the
dashboard
because
it
just
looks
so
beautiful.
A
Completely
agree
with
that
awesome
and
I
can't
tell
you.
I've
I've
been
on
six
calls
or
so
about
customers
who
have
showed
me
their
zbl
ABS
value
stream,
POC
and
I'm
like
Dee.
You
all
like
the
POC
is
not
going
to
get
you
like
where
you
need
to
be,
but
once
they
see
that
we're
looking
at
building
the
CI
CD
dashboard
like
have
it
be
natively
just
supported
in
their
gitlab
infrastructure.
A
They
automatically
like
just
stop
talking
about
CB
lives
so
like
giving
them
compelling
metrics
that
just
capture
what
their
team
is
doing
is
a
gap
that
we
need
to
solve,
and
we
need
to
solve
it
quickly
because
the
urgency
in
the
market
is
it's
coming
up
with
a
Z.
Bia
knows
that
were
a
threat
to
their
market
and,
as
a
result
like
any
of
these
deals
that
were
going
head-to-head
against
I've
seen
so
many
copy
pastes
from
community
forum.
People
complaining
about
how
they
can't
can't
manage
things
at
scale.
A
Copy
pasting
comments
from
other
analysts
telling
telling
the
market
that
gitlab
is
not
for
legacy
style
deployments
that
we're
only
for
cloud
native
style
deployments
and
those
kinds
of
these
kinds
of
fit
and
finish
that
you're
talking
about
Shaun
will
really
help
us
be
competitive
against
these
tools
that
have
been
serving
this
legacy
market
for
ten
years
now,
right,
they're,
really
the
laggards
but
they've
been
a
leader
for
so
long
that
it's
hard
to
uproot
that
culture
in
thinking
that
we
actually
can
offer
something
to
the
legacy
market.
Yeah.
B
And
if
you
have
it
beautiful
I
mean
you
know,
listen
what's
at
the
bottom
I
understand,
but
if
you
have
a
beautiful
dashboard,
their
content,
not
only
as
beautiful
but
does
all
this
stuff
I
mean
that's
almost
the
only
thing
you
need
to
show
you
know
like
you,
don't
need
to
show
people
the
am
all
right.
It's
like!
Okay,
that's
the
oh!
You
know
it
works
right
that
you
know.
Look
at
this
dashboard.
You
can
see
everything
that's
going
on
and
you
know
just
to
even
thinking
further
ahead.
B
B
A
A
A
Exactly
awesome
well
for
our
next
think.
Bag
will
probably
dive
into
a
calendar.
View
of
this.
I
will
hopefully
have
issues
created
about
us
being
more
competitive
with
her
aku's
usability
and
experience,
and
then
we
can
start
breaking
those
concepts
down
and
hayano.
We
can
revisit
our
CI
CD,
lo-fi,
mocks
and
kind
of
think
about.
How
do
we
want
to
consolidate
that
loaf,
I
mock
view
into
the
environments
dashboard
view,
and
maybe
that
becomes
like
a
CI
CD
dashboard
instead
and
that's
where
we
land
at
all.