►
From YouTube: 2020 10 13 Database Team Meeting
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
All
right
welcome
to
this
version
of
the
database
team
meeting
so
jumping
right
into
it's
the
last
week
the
milestone.
So
let's
look
at
the
milestone
board
review
the
issues
make
sure
they're
all
up
to
date,
so
andreas
you're
first.
B
Yeah
there
is
an
approach,
I'm
trying
right
now
and
just
had
a
discussion
with
giannis
about
it
and
I'm
going
to
work
on
this
a
little
bit
more
this
week.
A
Okay,
this
one
is
not
assigned
you
all
assigned
it
to
me,
so
I
think
we've
identified
some
common
themes
and
this
one
will
carry
over.
C
Yeah
so
there's
a
couple,
mrs
for
this
out
there
now,
the
first
of
which,
once
I
tested
it
on
the
database
labs
it's
not
going
to
work
in
production
because
there
has
to
rebuild
the
indexes.
It
just
is
not
going
to
work,
so
I
think
really
like
after
kind
of
building
that
spreadsheet
and
seeing
the
changes
that
this
is
going
to
be
like
a
significantly
larger
to
address
in
than
I
thought
initially
there's
a
lot
of
changes
here.
C
I
don't
know
how
realistic
it
even
is
to
maybe
make
some
of
these
changes
or
the
migrations
might
be
fairly
involved.
So
it's
probably
worth
having
a
conversation
about
about
that.
Maybe,
but
it's
not
gonna
be
straightforward
like
to
just
have
a
couple
migrations
and
that's
gonna
solve
it
and
then
everything's
better.
A
All
right,
yeah,
I
think
pick
it
up
asynchronously
and
if,
if
we
need
to,
we
can
pick
it
up
on
thursday.
C
Yeah
this
one
so
that,
mr
to
add,
the
helpers
is
in
review
right
now
and
I'm
gonna
start
working
this
week
on.
There
was
one
piece
that
we
sort
of
overlooked
or
didn't
really
know,
was
initially
signed,
a
nose
issue
which
is
adding
the
indexes
to
the
partition
table,
which
we
need
to
have
a
new
helper
for
so
I'm
gonna
be
working
on
that
this
week
and
we'll
see
how
it
goes.
Hopefully,
I
can
get
that
you
know
done
today
or
tomorrow
and
we
can
get
it
through.
C
A
A
That
one
same
thing
planning
the
next
one
giannis
europe.
D
B
And
this
one
has
been
under
here
for
one
and
a
half
weeks
now
so,
and
hopefully
it's
gonna
merge.
Today,
it's
been
just
a
long
back
and
forth
review
process.
Most
of
the
time
all
right
crown
job
is
finished
from
our
side.
It
should
have
been
installed
to
staging,
but
that
doesn't
or
didn't
work
out
as
expected,
and
I'm
waiting
for
feedback
from
the
sre
team
on
that
didn't
want
to
mess
with
the
playlist
myself.
D
Yep
this
one
depends
on
the
previous
one.
The
moment
the
other
one
is
merged
to
master.
I
will
have
to
change
some
specs,
because
this
those
new
migration
helpers
depend
on
the
other
ones
and
then
so
yeah.
It
will
be
released
this
this
week.
Okay,
that
one
I
haven't,
really
started
with.
B
I
wanted
to
pick
it
yesterday,
but
didn't
didn't
actually
start
it.
I
hope
to
have
some
time
this
week
to
do
that.
Actually,
okay,
so
I
would
rather
keep
it
right.
A
A
D
A
Okay
sounds
good
and
then
sounds
like
these.
If
anybody
has
time
we'll
just
pick
them
up
all
right
is
there
anything
that
is
not
showing
up
in
here
that
folks
are
working
on.
A
I
will
clean
that
up
later,
but
planning
for
13.6
the
major
initiatives
shouldn't
be
any
surprise
so
got
the
events
table
partitioning.
This
is
probably
not
the
best
title
we
can
rename
it.
It's
topic
we'll
talk
about
next
testing
migrations
before
they
get
to
staging
by
providing
developers
access
with
data
in
various
different
ways,
and
then
these
were
the
two
topics
that
jana
suggested
we
bring
in
the
thirteen
six.
B
There
is
one
more
topic
is
the
re-indexing
that
we
should
still
make
a
little
bit
more
smaller.
A
Right
is
that
issue
already
in
1306.
E
E
A
That
seems
like
we
have
an
awful
lot
in
13
6.
Those
probably
be
some
things
that
need
to
be
moved
out,
and
we
can
talk
about
that
asynchronously
and
I
will
just
I'll
update
the
planning
issue
with
suggestions
and
if
anybody
else
has
any
suggestions
feel
free
to
throw
them
in
there.
D
13.6,
so
it's
tracking
the
same
objects
that
we
track
for
self-hosted
instances,
but
for
gitlab.com
so
that
we
can
compare
and
have
in
the
same
dashboard
the
two
of
them.
D
A
A
A
We
threw
in
some
ideas,
but
did
you
have
some
thoughts
on
that.
B
B
A
follow-up
to
our
conversation
from
last
week,
and
but
it
sounds
like
we
would,
we
would
break
things
down.
But
what
is
that
something
that
we
should
also
spend
time
on?
Or
you
know?
B
Last
week
we
were
wondering
if
we
wanted
to
spend
like
a
couple
of
days
trying
that
out
to
get
a
better,
better
understanding
of
the
efforts
or
the
limitations
of
that
is
that.
A
A
I
think
it's
yeah,
it's
something
we
want
to
do
it's
a
high
enough
priority
item
that
we
need
to
solve
and
I
don't
think,
there's
an
easy
short
path
to
solving
that
migration
problem.
So
I
think
we're
going
to
be
the
ones
I
don't
think
I
know
we're
the
ones
that
should
be
leading
the
charge
on
that
one.
So
if
there's
some
ideas
down
below,
I
know,
we've
talked
about
starting
with
manual
migrations.
A
While
we
continue
to
work
with
nick
on
a
more
automated
process,
nick
or
something
else
like
if
we
ultimately
find
a
competitor,
a
piece
of
software
that'll
do
the
same
thing,
but
I
added
my
thoughts
down
below.
So
I'm
thinking,
maybe
we
as
a
team
figure
out
how
to
test
migrations
manually
against
database
labs
in
a
safe
way,
roll
it
out
to
the
maintainers
and
continue
to
update
docs
steps,
automate
whatever
we
need
to
on
the
initial
manual
steps,
so
that
developers
have
an
idea
on
how
to
do
it.
A
If
and
when
we
get
to
the
point
where
we
can
share
licensing
on
database
labs,
see
I've
written
up
the
issue
to
purchase
licenses,
it
doesn't
necessarily
mean
it's
going
to
go
through.
I'm
not
sure
how
the
pricing
is
going
to
be
taken
by
upper
management,
but
there
is
a
big
focus
on
making
sure
we
don't
have
these
migration
problems
anymore.
A
So
I
have
to
think
if
we
provide
some
kind
of
solution
that
it
will
get
approved
so-
and
this
is
our
we
already
have
the
software
now
we
know
we
can
do
the
manual
tests.
We
just
know
that
there's
some
things
we
need
to
be
aware
of.
So
this
seems
like
the
best
first
iteration.
C
A
Would
it
be
faster
to
use
the
use,
postgres,
ai
and
the
thin
cloning
tools
that
they
have?
I
mean
we
already
have
that
tool
available
to
us
right
now.
All
right!
Sorry,
I
didn't.
I
didn't
notice
the
distinction
of
what
you
were
saying.
I
was
thinking
it
was
the
tools
we
already
have
for
spinning
up
the
thin
clone
quickly.
B
Was
that
the
postgres
ai
solution
is
going
to
come
to
an
end,
if
we
don't
manage
to
extend
the
license
right,
yeah
and
everyone
wants
that
functionality.
So
there's
two
concerns,
or
this
is
one
concern
like
how
do
we
continue
to
provide
this
thin
coding
ability?
E
A
So
I'm
gonna
make
an
assumption
here.
It
would
be
faster
if
we
could
start
with
the
assumption
that
we
can
continue
to
use
postgres
ai,
regardless
of
the
license
wall
on
november
27th.
Right
so
I'd
say
we
start
with
that
and
let
me
figure
out
the
licensing
we're
no
worse
off
than
you
know.
If
a
month
from
now,
we
decide
we're,
not
gonna
move
forward
with
that,
then
we
can
start
testing
with
the
open
source.
A
I'd
say
work
with
the
tools
that
we
have
for
now,
because
even
then
that'll
give
us
some
ideas
on
making
sure
that
we
have
a
safe
environment
if
we
have
to
revert
back
to
the
open
source,
correct
we're
going
to
encounter
the
same
problems
anyway.
So
let's
proceed
with
the
assumption
that
we're
going
to
get
the
licensing
in
six
weeks
or
two
months,
whatever
the
timeline
is
so
that
we
have,
we
can
start
with
what
we
have
now,
instead
of
taking
a
step
back.
B
B
Instance
in
our
tcp
project
for
the
database
group
that
we
would
configure
in
a
way
that
we
are
more
confident
that
running
database
migrations
in
that
environment
is
safe.
Kind
of.
B
And
then
we
would
basically
want
to
give
out
access
to
that
instance
to
database
materials
at
some
point
and
we
might
build
some
automation
around
using
or
hooking
up
your
2dk
environment
or
something
like
that
where
you
can
run
the
migrations.
I'm
hooking
that
up
with
the
existing
postcards
ai
instance.
A
E
A
D
It
has
a
mailer,
it
has
access
to
the
outside
world.
So
most
probably
we
will
want
to
set
up
an
instance
without
smtp
running
or
access
to
the
outside
world
other
than
the
connection
just
be.
A
D
A
That
is
there
like
an
access
request,
issue
or
something
you
need
for
that,
or
is
that
just
an
email
to
alberto
and
team
or
sorry
a
certain
team
say
hey?
We
need
a
new
gcp
instance.
B
So
we
can,
we
can
start
tcp
instance,
ourselves,
that's
not
a
problem.
We
have
a
dedicated
project
for
us,
okay,
it
would
be
kind
of
experimental
or
you
know
it
would
basically
connect
to
the
to
the
other
gcp
instance.
B
D
D
E
D
B
B
Aspects
in
it,
so
if
we
you
know,
if
I'm
thinking
it
would
make
a
lot
of
sense
if
we
had
some
kind
of
collaboration
on
that
from
from
help
from
from
the
sra
team.
If
we
are.
B
B
B
And,
for
example,
data,
if
we,
if
we
wanted
to
set
up
database
labs
open
source
edition,
then
this
instance
has
to
have
access
to
the
object,
storage
from
production,
where
we
can
read
and
replicate
the
production
database
and,
and
that
is
kind
of
the
critical
resource.
That
means
auditing.
All
that
right.
A
I
was
just
wondering
what
we
could
start
actually
working
on,
so
we
understand
what
we
don't
know
right
and
start
writing
it
down,
so
that
we,
when
we
actually
go
to
to
request
the
I
guess,
the
long
term
gcp
project
for
auditing,
then
we'd
better
understand
what
we're
doing
so.
A
We're
better
set
up
for
success
down
the
road
but
sounds
like
we
need
to
have
that
audit.
First
for
doing
things
the
right
way.
I.
B
Think
if
we
wanted
to
have
database
labs
community
edition-
and
I
would
rather
approach
it
that
way
together
with
sre
figure
out
how
to
how
to
set
that
up
for
creating
that
isolated
instance,
I
mean
it's
probably
not
much
different
than
if.
E
A
B
A
A
What's
the
difference,
so
I
guess
database
labs
is
just
to
get
access
to
the
data.
Are
we
not
going
to
use
that
same
thing,
for
the
isolated
instance?
Is
that
we're
not
going
to
build
it
up
into
that
isolated.
B
B
I
think
if
so,
just
from
the
many
things
I
don't
know,
I
think
creating
an
isolated
instance
and
using
that
as
a
development
environment
is
probably
fine
right.
It's.
B
B
Can
do
replicating.
B
E
D
D
E
D
D
E
D
Will
have
to
as
a
database
team
to
figure
out
how
we
are
going
to
give
access
to
to
that
gcpa
instance.
If
we
are
going
to
give
access
to
others
or
not
or
if
we
are
going
to,
let's
say
something:
crazy:
build
an
automated
job
that
runs
the
test
or
something
like
that.
But
those
are
the
two
steps
at
the
moment.
Okay,.
D
E
E
B
If
we
can
run
database
migrations
on
on
that
data,
how
can
we
figure
that
out?
Is
there
a
team
that
we
would
approach
or.
A
Let
me
figure
that
one
out
I'm
thinking
I
mean
this-
is
something
that
we
do
every
step
of
the
way
anyway,
so
this
isn't
outside
what
we'd
normally
do
on
a
deployment.
I
don't
see
it
as
testing
against
production
data.
It's
we're
applying
the
steps
to
the
development
process,
so
let
me
figure
it
out.
D
And
maybe
we
missed
the
make
the
server
safe
there
together
with
install
jdk?
What
we
were
discussing
previously,
so
it's
like
install
jdk
and
make
yeah
makes
ensure
that
the
server
is
safe,
safe
with
respect
to,
if
we
run
a
migration
that
the
sensor
nails
emails
are
not
sent
or
updates
something
in
the
third
third
party
service.
Am
I
right.
D
Yeah,
it
should
not
be
that
tricky.
D
Are
afraid
about
in
their
own
local
dev
environments
like
running
the
migration
starting
sending
mails
to
our
customers
for.