►
From YouTube: 2022-03-09 AMA about GitLab releases
Description
Monthly AMA with the Delivery Team
A
B
A
C
Yep,
thank
you.
I
was
actually
searching
for
the
links,
but
I
can
add
them
later,
so
there
are
two
things
that
I
am
excited
about.
The
first
one
is
the
introduction,
the
official
introduction
of
the
staging
canary
environment
that
will
allow
us
to
reorder
the
coordinated
pipeline
so
long
story
short.
First,
we
will
deploy
to
the
canaries
and
then
we
will
deploy
to
the
main
stages,
almost
at
the
same
time,
which
will
allow
us
to
account
for
mixed
deployment
testing,
and
it
also
has
some
nice
consequences
of
probably
reducing
the
coordinated
pipeline
duration.
C
So
we
can
reduce
the
mttp
the
mean
timed
production
with
that,
so
that
is
one
and
the
other
one
is
to
remove
the
post
migrations
from
the
coordinated
pipeline.
So
we
all
know
that
post
migrations
are
a
bit
painful.
They
are
not
the
most
easy
migrations
to
handle
and
they
actually
contribute
to
the
most
time
of
the
deployment.
C
Here's
one
of
the
elements
that
do
that.
That
does
that.
So
if
we
remove
that,
we
can
also
have
more
control
about
the
coordinated
pipeline
and
its
execution,
and
it
doesn't
mean
that
we
are
not
going
to
execute
it
anymore.
It
just
means
that
we
are
going
to
execute
it
in
a
more
controlled
environment
and
I
think
well,
I'm
also
excited.
C
I
am
excited
about
a
lot
of
things,
but
another
thing
that
would
be
automated
deployments
because
currently
their
deployments
to
production
depend
of
someone
being
there,
like
current
release
manager
being
there
to
actually
trigger
the
deployment
and
with
the
automated
deployments.
Well,
we
will
remove
that
human
dependency,
so
that
would
be
nice.
A
A
Yeah,
okay!
Well,
I
think
a
few
other
things
that
we
are
we
have
kind
of
coming
up
this
year
that
are
going
to
be
exciting
is
so
we
are
working.
We
are
on
the
kind
of
next
phase
of
our
kubernetes
migration,
which
is
the
stateful
services
and
with
someone
actually
mind
like
mario,
would
you
mind
just
making
some
notes?
Whilst
I
talk
awesome
thanks,
so
we
are
working
with
scalability
right
now
to
migrate
the
rate
limiting
redis.
So
that's
really
exciting.
A
It's
certainly
a
new
one
for
us,
so
it's
quite
exciting
sort
of
figuring
out
like
how
we
can
migrate
that
how
we
can
make
that
safe
and
then
how
we
can
repeat
that
over
hopefully
most
possibly
not
all,
but
hopefully
at
least
most
of
our
other
redis
instances,
and
then
we're
also
in
talks
about
like
what
will
come
following
that.
So
we
can
bring
some
more
things
into
the
kubernetes
customers
as
well
and
then,
as
a
lot
of
these
changes
come
together.
A
What
will
be
exciting
sort
of
later
in
the
year
is
how
we
can
start
getting
other
like
stage
groups
more
involved.
So
we're
hearing
that
you
know
there
are
lots
of
people
who
would
actually
like
a
lot
more
independence,
a
lot
more
freedom
with
deployments,
so
it's
gonna
be
really
exciting
to
actually
figure
out.
How
do
we
hand
these
this
sort
of
controls
back?
B
A
Yeah,
I
mean
it's
definitely
something
to
be
determined
like
we'd
like
to
give
stage
groups
as
much
control
as
we
can,
whilst
still
making
it
so
that
it's
it
is,
you
know
safe
and
reliable.
We
have
people
who
can
respond
to
problems
and
that
the
actual
you
know
deployments
or
releases
still
continue.
So
it
will
just
depend
a
little
bit.
We
still
need
to
kind
of
figure
out
exactly
what
would
be
involved
or
what
tools
we'd
need
to
to
sort
of
put
in
place.
For
that.
A
But
I
mean
the
idea
would
be
complete
controlled,
but
I
think
we're
a
little
bit
of
a
way
away
from
that.
We
certainly
need
to
probably
shift
some
things
around
in
kubernetes
and
hook
it
a
little
bit
more
with
error
budgets
and
as
well
as
sort
of
make
the
tooling
a
bit
more
friendly
for
for
someone
outside
of
delivery
to
be
able
to
to
use.
A
Okay,
well
one
thing
I'd
like
to
just
kind
of
mention,
so
myra
talked
a
little
bit
about
the
official
introduction
of
the
staging
canary
environment,
so
just
as
I
mentioned,
we'll
be
announcing
this
more
widely
in
the
next
few
days.
But
what
we've
been
working
on
for
the
last
few
months
is:
we've
introduced
a
staging
canary
environment.
A
At
the
moment
this
is
kind
of
duplicating
our
staging
environment,
so
we
deploy
packages
onto
stage
and
canary
we're
running
tests,
but
we
also
follow
up
with
the
deployment
under
staging
and
we
run
tests
and
we
are
very
close
to
being
able
to
do
the
pipeline
reorder,
which
is
kind
of
the
the
end
sort
of
piece
for
this
project,
and
that
gives
us
the
mixed
deployment
testing
setup
that
myra
mentioned.
So
this
will
change
the
deployment
pipeline
order.
A
It
groups
our
canaries,
together
ahead
of
our
main
fleets,
so
we'll
be
deploying
to
staging
canary
and
production
canary
and
then
staging
and
production.
So
it
should
be
mostly
invisible
to
people
but
we'll
be
using
the
canary
cookies.
So
that's
coming
quite
soon,
we'll
be
putting
out
more
announcements
with
details
that
what
this
means
and
how
people
can
make
sure
they
kind
of
know
what's
happening
and
also
let
us
know
if
anything,
doesn't
look
quite
right,
but
that
should
be
happening
very
soon.