►
From YouTube: GitLab 16.0 Kickoff - Enablement:Gitaly
Description
Gitaly Kickoff issue - https://gitlab.com/gitlab-org/gitaly-planning/release-planning/-/issues/10
A
Hello,
everyone
and
welcome
to
the
getaway
16.0
planning
issue
for
gitlab
16.0.
We're
really
excited
about
this
upcoming
major
release
Milestone
and
have
some
pretty
exciting
things
planned
over
the
next
month
month
and
a
half
and
are
excited
to
share
them
with
you.
As
always,
this
planning
issue
will
be
linked
in
the
YouTube
video
Below
in
the
comments.
So
if
you
have
feedback
questions
or
comments,
we'd
love
to
hear
them,
I
will
also
from
this
feedback
issue.
You
can
get
to
all
the
issues
linked
and
again.
We
welcome
feedback
on
those
as
well.
A
There
are
no
capacity
limitations
during
this
Milestone
that
we
know
of
as
of
right
now.
So
we
plan
on
having
full
capacity
throughout
I,
do
have
a
direction
page
update
underway
and
an
update
was
just
done
so
feel
free
to
go
to
our
Direction
page
to
see
objectives
and
themes,
especially
as
we
finalize
setting
the
okrs
for
this
quarter.
I
will
be
updating
the
direction
page
again,
so
the
features
we're
looking
to
do
here
are
Implement
right
ahead.
Logging.
A
This
is
an
ongoing
part
of
an
ongoing
shift
towards
the
decentralized
architecture
for
cluster,
as
well
as
the
disaster
recovery
working
groups
need
for
improved
backups,
we're
continuing
the
effort
to
implement
right
ahead.
Logging
we've
been
working
on
this
for
the
last
month
or
two,
and
we
continue
to
make
progress.
We
don't
expect
it
to
be
fully
integrated
into
the
product
yet,
but
we
have
been
making
substantial
work
and
we
believe
that
the
effort
is
on
track.
A
We
do
update
this
status
weekly
to
report
back
on
how
things
are
going,
and
you
can
read
about
sort
of
the
overview
of
what
we're
intending
to
do
here.
If
you
have
additional
information,
you'd
like
to
add
or
questions
again,
this
is
an
ongoing
effort
and
we
welcome
all
sorts
of
feedback
on
it.
A
We're
also
focusing
on
paying
off
some
technical
debt.
There's
a
couple.
Things
have
been
lingering
for
a
while
one
was
we
had
a
ruby
sidecar,
which
is
essentially
a
bunch
of
Ruby
code
that
we
had
in
gitlab
itself
that
worked
in
conjunction
with
Italy.
We
have
finalized
removing
all
of
those
dependencies
and
we
expect
to
get
rid
of
Ruby
in
giddling
15.11,
which
will
have
ended
by
the
time
this
Milestone
starts.
However,
some
of
that
tail
end
work
May
reach
into
60
known.
We
don't
want
to
lose
focus
on
that.
A
We
do
want
to
finalize
the
entirety
of
this
Tech
debt
so
that
we
don't
have
it
looming
over
us
anymore.
So
we
want
to
make
sure
that
is
finalized
in
160..
They
should
have
no
customer
facing
impact,
but
it
is
important
to
our
team.
The
other
thing
we're
looking
to
do
is
live
get
to
deprecation.
A
We
are
working
on
the
deprecation
efforts
here.
We
do
have
a
lot
of
commands
that
are
being
replaced
by
standard
git
commands
and
we
are
working
through
them
with
the
get
mailing
list
and
Upstream
get.
This
will
take
multiple
iterations
before
we're
complete,
but
it
is
also
top
of
Mind
as
it's
more
technical.
That
we'd
like
to
remove
we're,
also
very
excited
to
start
taking
ownership
of
object,
pools
object,
pools
have
been
a
long-standing
issue
within
giddling
cluster
they've
never
been
quite
right,
and
we
really
really
aim
to
fix
that.
A
We
want
to
make
them
a
lovable
feature
within
giddly
and
make
them
really
work.
Well.
We
also
need
them
to
have
write
a
logging
work
correctly.
So
the
first
step
to
do
in
the
ownership
and
moving
object,
pools
from
rails
to
Italy,
was
to
create
a
blueprint.
We
have
a
blueprint
issue
out
there
where
we
are
actually
creating
this
blueprint.
There's
been
some
amazing
feedback
from
our
internal
and
external
Community
I
believe
we
are
working
through
this
feedback.
As
you
can
see
here,
it's
quite
a
long
issue.
A
There
are
31
current
unresolved
threads
on
it.
We
are
enjoying
all
the
feedback
and
trying
to
find
the
best
blueprint
possible.
So
we
can
properly
move
forward
with
that
in
a
way
that
benefits
both
us
and
our
users.
A
Finally,
one
of
the
big
features
we're
looking
to
implement
is
teaching
Italy
to
apply
back
pressure
to
traffic
searches.
It's
impractical
to
allocate
enough
resources
to
get
away
at
all
times,
so
there's
never
contention
and
that's
not
really
an
ideal
situation
either.
Having
servers
that
do
nothing
and
processing
power,
that's
doing
nothing,
99
of
the
time
is,
is
Impractical
and
expensive.
A
Therefore,
we're
looking
to
have
giddly
apply
the
back
pressure
when
there's
a
need
which
will
allow
this.
You
know
on
this
to
run
nicely
on
Hardware.
That
is
not
so
great
as
to
be
unresource
constrained,
and
this
will
allow
us
to
correctly
apply
the
back
pressure
such
that
we're
limiting
traffic
in
a
meter
way,
but
not
impacting
performance
or
usability
in
any
great
way.
So
this
is
something
that's
very
important
to
us.
A
Finally,
our
stable
counterparts
in
testing
have
identified
some
issues.
We're
working
through
here
from
the
backlog
of
severity
to
issues.
I
think
we've
triaged
a
few
of
these
and
closed
them
because
they
are
no
longer
relevant,
the
others.
We
do
plan
on
working
to
reduce
the
list
and
clear
the
list
of
severity,
two
issues
which
is
really
important
to
us
again.
If
there's
any
questions,
let
us
know
we're
excited
if
we
get
another
great
release
and
the
release
of
gitlab
16.0.