►
From YouTube: GitLab 15.2 Kickoff - Enablement:Distribution
Description
Planning Issue - https://gitlab.com/gitlab-org/distribution/team-tasks/-/issues/1021
A
So,
as
always,
we'll
start
off
with
our
scheduling
issue
and
similar
to
last
milestone.
If
you
didn't
watch,
we
are
bundling
the
15
1,
2
and
3
milestones
in
one
scheduling
issue.
Just
try
something
new
and
I've
updated
the
description
today
for
the
recent
developments
as
we
move
into
15
2..
So,
just
like
last
time,
we
are
going
to
continue
working
on
get
lab
operator
as
a
top
priority.
A
Another
consideration
that
we
had
is
related
to
fips,
which
I'm
going
to
talk
a
little
bit
about,
but
that
was
kind
of
the
big
sticking
point
was
we
have
to
consider
something:
that's
ubi
capable
and
also
will
work
for
fips
so
finishing
that
up
or
close
finishing
up
early
on
152
unlocks,
essentially
the
rest
of
these
issues.
You
know
ensuring
all
container
images
have
uvi
variants.
I
think
that's
pretty
straightforward
that
that'll
unlock
and
then
those
images
being
deployed
successfully
and
then
sort
of
the
final
certification.
So
that's
what's
gonna
wrap
up.
A
Moving
on
to
the
next
item,
which
is
new,
we
had
been
working
on
fips
for
a
while,
but
I
hadn't
listed
it
because
we
had
kind
of
finished
up
a
lot
of
the
work,
but
as
we
move
towards
the
next
audit,
which
is
going
to
help
land
just
after
15
2,
there's
a
couple
items
that
distribution's
going
to
be
picking
up
and
sort
of
company-wide.
A
This
is
an
important
issue,
so
it's
going
to
be
an
important
issue
for
us
in
15
2.,
there's
just
a
couple
issues
listed
that
are
listed
here
that
are
related
to
fips
that
we
want
to
that.
We
want
to
complete
in
enough
time
to
have
them
ready
for
15-2.
A
Moving
on
these
are
familiar
so
we're
still
working
on
or
making
a
priority
for
some
of
our
tech,
debt,
internal,
tooling,
work
related
to
you
know
our
long-term
build
efficiency
and
also
improving
our
team
pipelines.
So
both
of
these
hopefully
will
pay
dividends
down
the
road
to
make
the
features
and
tools
we
roll
out
in
the
future
easier
to
implement
and
we're
going
to
talk
about
a
couple
of
things
we're
working
on,
but
a
big
one,
that's
going
to
be
finishing
up
or
is
very
close.
A
A
lot
of
work
was
done
at
15.1
is
investigating
how
we
can
paralyze
omnibus
and
the
builds
related
omnibus
and
basically
improve
the
upgrade
diffs
for
for
omnibus,
and
this
is
going
to
lay
a
nice
foundation
for
our
long-term,
build
efficiency
and
the
other
issues
tied
to
it.
A
So,
moving
on
into
a
couple
of
issues
I
wanted
to
call
out,
the
first
ones
are
going
to
be
related
to
long-term,
build
efficiency
and
the
efficiency
work
related
to
our
pipelines.
I
won't
talk
about
any
issues
related
to
openshift
and
fips,
just
because
I've
talked
a
fair
amount
about
them
and
if
you'd
like
to
know
information,
please
click
these
two
epics,
but
moving
on
to
the
first
one.
A
Here
we
want
to
refactor
our
ci
pipelines
to
use
rules
essentially,
there's
there's
two
ways:
we
currently
do
it
and
we
like
to
use
a
consistent
approach
and
in
this
case,
we're
probably
going
to
use
rules
but
we're
going
to
kind
of
investigate
and
finish
up
this
pipeline
work,
which
is
again
related
to
improving
our
pipelines.
So
we're
excited
about
this
one
and
another
one
here
that
we're
excited
about.
We
actually
don't
have
it
tied
to
one
of
the
efficiency
epics,
but
it
is
certainly
related.
A
We
want
to
make
sure
that
any
changes
to
our
rails
app
won't
break
anything
on
any
of
our
os
builds.
So
right
now
it's
a
manual
process
for
each
of
the
builds,
but
we
want
to
make
sure
essentially
make
sure
this
is
fairly
automated.
So
this
will
be
a
nice
improvement
as
well
and
then
the
final
one
I'd
want
to
call
out
relate
to
some
of
our
efficiency
work
and
pipelines
is
right.
A
Now
we
want
to
make
sure
the
issue
bot
which
we're
implementing
will
trigger
slack
notifications,
so
this
is
kind
of
the
top
level
one
and
there's
issues
related
to
our
charts,
omnibus
and
operator
projects.
A
A
So
this
is
a
basically
a
bug,
improvement
that
we
want
to
implement
15-2
so
that
customers
can
still
use.
You
know
web
services
that
they'd
like
ssh
for
layer
four,
so
this
will
be
a
nice
one
as
well,
and
then
this
is
another
one
that
we've
been
investigating
for
a
while
and
we're
excited
to
implement,
which
is
multiple
pg
bouncer
instances
on
a
single
node
yeah.
A
So
this
is
just
a
nice
one
that
we
can
now
run
or
hopefully
we'll
be
able
to
run
multiple
instances
on
a
single
node
moving
forward
and
then
moving
on
to
sort
of
the
final
customer.
One
I
want
to
bring
out
is
console
being
able
to
watch
additional
postgres
services.
So
right
now,
it'll
fail
due
to
some
hard
codes
that
we
have,
but
if
we're
building
a
new
environment
with
two
high
availability,
postgres
setups
for
gitlab
and
prefect,
we
would
like
them
both
to
be
able
to.
A
You
know,
be
watched
and
right
now
that
it's
unable
to
so.
This
will
again
be
another
nice
improvement.
So,
finally,
a
fun
one
I
want
to
call
out
is
now
that
the
bullseye
os
is
out.
We
want
to
provide
that
package
for
raspberry
pi,
and
so
this
is
another
thing
we
work
on
15.2,
so
that
is
rounding
out.
A
Our
15.2
milestone
again
we're
going
to
be
working
on
the
operator
and
fips,
mainly
and
we'd
love
to
get
these
done
in
15.2,
or
at
least
these
phases
of
those
projects,
and
then
we're
going
to
continue
working
away
at
our
efficiency
work
to
make
the
live
of.
You
know
the
engineers
a
little
easier
as
we
roll
up
features
in
the
future.
So
thank
you
for
listening
and
have
a
great
rest.
Your
day.