►
From YouTube: GitLab 14.3 Kickoff - Enablement:Database
Description
Kickoff for the Database Group for the GitLab 14.3 release
Planning issue: https://gitlab.com/gitlab-org/database-team/team-tasks/-/issues/188
A
A
There
are
broadly
three
areas
that
I'm
going
to
address:
infra
death
issues,
the
work
we're
doing
to
address
the
risk
of
a
primary
flow
overall.
A
Excuse
me
the
work
we're
doing
to
address
the
primary
key
overflow
risks
for
tables
with
an
integer
primary
key
and
then
some
known
issues
for
our
ongoing
sharding
or
more
specifically,
decomposition
efforts
of
the
database.
A
So
this
may
result
on
more
load
on
the
primary
or
postgres
cluster
and
then
some
other
issues
with
active
record.
So
these
are
highly
technical
database
related
issues
that
we
need
to
fix
in
order
to
ensure
that
gitlab.com
runs
smoothly
and
they
take
priority
over
some
of
the
more
long-term
longer-term
work.
A
The
second
thing
here
is
the
ongoing
work
to
address
the
primary
key
overflow
risks
for
a
table
with
an
integer.
So
essentially,
these
tables
have
grown
so
much,
but
at
some
point
they
run
out
of
primary
keys
because
they
are
simple,
integers,
so
we're
working
to
actually
migrate
that
those
two
begins
so
that
they
don't
overflow-
and
this
has
been
ongoing
for
a
few
releases
already
in
14.2-
we're
swapping
out
the
relevant
columns
and
then
in
14.3
the
upcoming
release.
A
That
may
otherwise
be
impacted,
as
you
can
see,
there's
a
lot
of
activity
going
on
in
14.2
and
a
lot
of
these
items
here
in
14
point
will
be
in
14.3,
and
this
is
mainly
about
finalizing
this
and
dropping
columns
and
we're
all
looking
forward
to
having
this
this
resolved.
A
A
So
I'm
not
going
to
go
into
the
details
for
all
of
those,
but
to
give
you
one
example:
here
we
have
migration
helpers,
for
example,
and
these
migration
helpers
are
not
aware
that
there
is
more
than
one
database,
and
so
we
need
to
now
actually
fix
them
and
ensure
that
it
becomes
a
way
of
having
at
least
two
potentially
more
in
the
future,
and
this
is
important
so
we're
able
to
continue
with
that
work
and
the
database
group
is
working
very
closely
with
the
sharding
group
on
delivering
that
kind
of
work.
A
A
To
avoid
that
we
get
a
hard
block
on
databases
and
can
grow
them,
but
this
is
now
in
the
finishing
steps
of
actually
swapping
and
dropping
columns,
so
excited
to
get
that
done
and
that's
it.
There
is
a
lot
going
on.
I'm
very
grateful
for
working
with
such
a
television
team
and
I'll
talk
to
you
next
time.
Bye.