►
From YouTube: 2021 04 27 Database Group Sync
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
B
Yeah
I
closed
that
one.
We
had
it
open
for
a
long
time.
There
was
good
feedback
on
the
issue.
We've
rolled
it
out
to
database
containers
and
more
people
can
actually
start
using
it.
We
haven't
yet
advertised
that,
but
I
just
wanted
to
close
that
issue.
For
now.
Okay,.
A
B
I
think
we
need
to
do
a
little
bit
more
work,
make
it
make
it
more
robust.
I
think
we
were
already
seeing
that
database
lab
is
pretty
busy
at
times,
so
we
might
want
to
be
a
bit
careful
before
we
find
it
out
more
okay,.
A
Okay,
andreas
are
you
closest
to
these.
B
Yeah
we
have
a
couple
of
those
basically
always
talking
about
step,
one
adding
new
columns
and
syncing
data.
This
is
needed
for
a
lot
of
tables,
and
that
is
one
of
them
and
I
think
we
are
approaching
nearly
having
shift
migrations
for
most
of
the
tables.
I
think
so
not
a
lot
is
missing.
B
So
that's
looking
pretty
good,
I
think,
and
the
step
two
is
gonna
like
once.
I
think,
once
we
have
added
the
migrations
for
those
tables,
we're
gonna,
look
at
the
next
step.
B
D
B
Yeah
exactly,
I
think
the
reason
is
that
we
basically
want
development
environments
to
run
data
migrations
conveniently
so
they
don't
need
to
care
about
that
at
all.
They
just
run
database
migrations,
but
we
don't
want
to
do
that
in
production,
and
now
we
stop
doing
that
in
testing.
Because
of
this,
because
the
data
migration
can
take
a
long
time,
we
still
have
it
in
development
for
now.
B
Eventually
that
might
even
go
away
and
we
make
it
more
explicit
so
that
you
basically
run
database
migrations
and
you
also
have
a
way
to
to
execute
pending
data
migration
separately
from
that.
But
right
now
it's
still
there
for
development.
A
B
A
Probably
don't
need
to
run
through
the
next
two
unless
there's
anything
interesting
to
cover
there
and
cross
finished
up
supporting
multiple
columns
in
a
single
migration.
That's
awesome.
B
That
was
great,
that
was
unblocking
us
to
to
ship
those
migrations
for
a
lot
of
tables,
including
ci
builds,
which
needed
the
multi-columns
support.
Yep.
B
That
you
can
actually
push
that
too
closed
that
has
been
deployed,
so
we
basically
swapped
web
book
logs
we're
now
using
the
partition
table.
There
is
still
triggers
in
place
so
that
we
still
maintain
the
non-partition
table.
So
the
last
step
is,
we
can
then
drop
the
non-petition
table
at
some
point.
B
B
They
need
to
access,
but
we
don't
manage
those
explicitly
somewhere,
so
this
is
basically
being
set
up
manually
and
when
you
swap
a
table
you're
not
like
getting
the
same
permissions
as
you
had
before,
and
that's
where
it
caused
persecutors
exporter
to
stop
working
and
they
needed
to
re-add
the
permissions
on
on
that
table.
B
B
B
B
So
the
the
rig,
mostly
thing
is
currently
on
hold,
but
I'll
pick
it
up
sometime
this
week
and
the
interesting
one
of
these
is
this
one,
where
we
basically
try
to
do
what
we've
done
manually,
try
to
do
that
automatically
so
that
we
take
advantage
of
weekends
or
low
traffic
times
still
on
very
early
like
implementation,
and
we
will
have
to
see
how
that
works.
A
Yeah
I
was
just
reading
through
that
mrs
very
cool
yeah
that
was
probably
not
interesting.
Adding
new
columns
giannis
did
ask
for
a
review
of
his
time
decay
blueprint.
He
sent
that
out
yesterday,
so
I
think
he
copied
most
of
us
on
that.
So
when
you
have
time
comment
on
that,
dev
heinrich
the
first
one.
E
A
E
One
yeah
this
this
is
the
admin
panel
and
I've
talked
about
it
a
bit
in
the
issue
like
I
could
probably
just
reuse
some
elements
that
we
already
use
in
the
admin
section,
because
we
use
tables
a
lot
and
most
likely.
This
would
be
something
like
tabular
right
and
yeah.
So
I
just
needed
something
from
my
other,
mr,
which
also
adds
the
failed
status
to
the
enum
for
migration.
E
Batch
migration
batch
migrations,
the
parent
row,
because
we
don't
have
a
failed
right
now,
and
I
thought
it
would
be
useful
in,
like
this
admin
panel
to
separate,
like
those
that
have
that
are
running
that
are
failed
and
those
that
have
completed
right,
because
those
failed
ones
are.
Those
that
are
running
are
the
more
interesting
ones
and
those
would
have
like
buttons
for
adjusting
or
for
retrying
or
all
that.
C
E
Yeah,
so
what
I'm
doing
is,
like
the
other,
mr
I'm
doing,
is
doing
some
retries
right
and
we
have
like
a
max
attempts,
and
so
if
all
the
jobs
are
not
retriable
anymore,
like
they're
over
their
max
attempts
and
we
still
see
failed
jobs
and
we
mark
the
parent,
that's
failed,
so
it
shows
up
in
the
admin
panel
has
failed
and
user
could
probably
click
retry
or
something
to
reset
the
attempts
or
something
like
that.
Yeah.
C
And
if
I
recall
correctly,
there
was
also
a
discussion
on
this
one
with
regarding
to
input
from
ux
design,
and
I
think
people
essentially
said
proceed
no
immediate
concerns,
but
I
think
these
people
are
now
in
the
loop.
If
you
need
anything
just
holler
and
I'll
organize
it
somehow.
A
Yeah-
and
I
posted
a
question
this
morning,
do
you
need
any
actual
ui
help?
I
think
the
only
front-end
developer
we
have
in
all
of
enablement
is
zach
on
which
team
is
he
on?
I
forgot.
E
Yeah
sure
I'll,
let
you
know
if
front
end
or
ux
support
is
needed.
Okay,.
B
D
D
D
It
is
now
I
don't
know
if
there's
anything
interesting
to
talk
about
on
it,
but
still
working
on
it,
though.
A
All
right
and
then
we're
going
to
go
through
the
ready
for
development.
A
Triage
board,
although
typically
it's
giannis
looking
at
that
I'll,
take
a
look
at
it
later.
A
And
then
I
put
a
note
about
the
migration
framework
beyond
maintainers,
so
I
will
follow
up
on
that
one
basic,
that's
it!
We
had
a
very
short
agenda
for
today.
Is
there
any
other
topics
folks
want
to
cover
today,
or
anybody
need
to
talk
about
something
they're
working.
D
A
Jen
asked
me
why
we
had
week
meetings
twice
a
week,
and
I
put
the
note
up
here
so
I
I
thought
I
had
updated
the
handbook
page
to
cover
this,
but
thursdays
are
optional
and
if
anybody
has
any
topics
they
want
to
cover,
sometimes
we
end
up
just
having
a
coffee
chat
about
random
stuff
on
thursdays.
So
it's
it's
not
meant
to
go
over
the
same
topics
we
go
over
on
tuesday.
I
have
since
updated
the
handbook
to
reflect
this.
One
thing
that
did
come
out
of
like
reviewing
our
team
processes
is.
A
We
do
typically
set
aside
the
first
thursday
after
a
milestone
close
to
do
the
retro.
If
it's
not
already
been
done,
because
the
feedback
from
the
team
was,
let's
do
this
synchronously,
sometimes
there's
it
doesn't
feel
as
meaningful.
If
we
don't
talk
about
it,
so
actually
this
thursday
would
be
if
we
haven't
done
the
retro,
which
I
don't
think
anybody
has
would
be
our
first
opportunity
to
do
it.
A
So
I'll
put
it
on
the
agenda
to
talk
about
the
retro
this
thursday,
but
it's
not
required
if
you've
done
the
retro
asynchronously
and
you
have
other
things
to
work
on.
You
do
not
have
to
come
to
thursday
meetings
so
and
then
I
don't
know
how
many
folks
know
about
the
engineering
allocation
effort,
but
as
part
of
that
work,
I'll
link
the
agenda
in
the
issues
here.
As
part
of
that,
I
need
to
label
issues
and
add
weights
to
them
so
that
we
can
dog,
food
or
road
map
see
this
one.
A
A
Ugly,
so
the
weight
gives
us
the
percent
complete
sorry
trying
to
show,
as
you
can
see
the
49
there,
and
then
the
label
just
makes
it.
So
we
can
utilize
this
roadmap.
A
B
Yeah
I
alex-
I
saw
your
comment
on
that
nice.
You
were
basically
asking
like
what
is
the
version,
one
that
we're
targeting
right
for
for
database
testing
or
perhaps
even
what
is
the
version
2w
that
we
want
to
do,
and
I
think
you're
you're
totally
right
on
that.
I
think
we
should
define
that
better
for
sure
and.
D
B
Yeah,
exactly,
I
wonder,
like
how
we,
how
we
do
it
and
is
fabian.
Is
that
something
that
you're
also
interested
in
helping
us?
Yes,.
C
I
mean
so
I
can.
I
can
tell
you
from
my
experience,
how
I
usually
do
it
and
I've
not
spent
time
on
this
here
yet
so
we
often
like
group
everything
in
an
epic
and
then
essentially
are
pretty
strict
on
like
what
we
add
or
remove
right
and
lay
out
in
the
epic
description
sort
of
what
is
the
minimal
functionality
that
we
expect
for
for
that
iteration,
right
and
kind
of
say,
okay,
you
know
this
is
how
we
expect
this
to
look
depending
on
on
that.
C
You
know
we
we
tried,
like
I
tried
to
control
the
scope
of
those
right,
because
there's
often
a
lot
of
scope
creep
right
over
time
where
it's
like.
Oh
you
start
doing
it
right
and
you
figure
out.
C
Oh,
we
should
really
think
about
this
and
think
about
that
and
ethics
get
longer
and
longer,
and
I'm
I'm
starting
to
be
a
little
bit
more
rigorous
with
saying:
okay,
the
first
thing
that
we
can
actually
roll
out
we'll
have
roughly
you
know
these
specific
functionalities
and
that's
it
and
anything
else
will
move
to
a
the
next
epic
for
the
next
iteration
and
that
that
means
that
there
is
a
definite
point
in
the
future
where
you
are
going
to
close
this
epic
and
you're
going
to
release
it
in
a
release
post
I
mean
here,
maybe
it's
it's
internal
and
then
you
get
celebrate
and
you
move
on
to
the
next,
the
next
iteration,
depending
on
how
large
these
things
are
right,
it
may
contain
several
sub
epics
that
you
can
that
you
can
also
close
right,
so
you
may
have
like
a
relatively
large
epic.
C
It's
like
okay,
if
we
are
able
to
do
all
of
this
sort
of
background
migration
stuff,
including
automatic
automatic
scaling,
that's
sort
of
the
first
thing
that
we
can
close
off
frame.
That's
like
33
percent
of
the
entire
epic,
and
there
are
like
a
couple
more
things.
That's
usually
how
how
I
do
that.
C
I,
like
I'm,
happy
to
help
you,
I
think
alex
actually
has
a
great
understanding
of
that
kind
of
process
from
from
geo
as
well.
So
maybe
we
can
get
together
and
hash
this
out.
B
C
And
that's
by
no
means
a
criticism
at
all
right.
It's
just
that's
kind
of
what
we
what
we
can
usually
do
and
I
think
that
helps
sort
of
define
when
when
is
it
good
enough
to
actually
roll
out
right?
And
what
does
that?
What
does
that
mean?
B
Yeah,
that's
great,
I
think
we're
in
terms
of
like
defining
what
we
what
we
want
and
how
that
should
look
like.
I
think,
we're
still
a
little
bit
in
that
sort
of
mvp
state.
We
have
something
working,
but
we've
never
really
talked
about
what
we
actually
want
in
the
end
or
how
that
should
look
like.
C
What
about
I
mean
my
my
this
week
is
busy,
but
maybe
like
I
don't
know
who
would
like
to
be
involved
in
that
discussion
alex
so
yeah.
We
we
have
to
push
it
into
the
afternoon,
which
is
impossible
for
me
this
week,
but
I
don't
think
it's
super
urgent.
C
What
I
can
offer
is
I'll
I'll
take
a
look
already
and
make
some
suggestions.
We
can
do
a
little
bit
of
async,
but
maybe
for
our
next
meeting.
It
may
make
sense
to
spend
30
minutes
on
sort
of
going
through
it
and
getting
some
some
thoughts.
C
Thank
you.
I
also
like
I'm
a
very
strong
proponent
of
using
these
templates
for
that
we
actually
have.
I
often
delete
things
that
are
not
necessary.
It's
like
my
typical
sort
of
problem
to
solve
right
proposal.
Success
criteria,
you
know
like
additional
stuff,
because
it
it
helps,
helps
me
like
to
put
it
into
sort
of
a
standardized
format.
B
C
Yeah,
so
the
issue
templates,
I'm
not
sure
we
actually
have
epic
templates,
so
I
copy
the
issue
template
over
and
because
I
have
so
many
epics.
I
now
just
copy
epics
from
that.
I
created
and
delete
the
content
and
just
keep
the
skeleton,
because
I'm
pragmatic
and
feel
very
little
shame.
D
C
B
It's
also
going
quite
well,
I
think
better
than
we
that
we
anticipated.
I
think
so.
That's
that's
good.
At
the
end
of
the
month,
we're
gonna
be
like
more
confident
with
timelines
and
not
having
that
overflow
happen.
Ultimately,
and
that's
gonna
be
good,
but
different
topic.
C
Yeah
actually,
like
I
spoke
with
a
friend
of
mine,
who
you
know
also
works
in
in
software
and
about
the
the
state
of
our
database
right
and
I
said
like
so.
If
you
look
at
the
like,
we
have
these
large
tables.
What
do
you
think
could
go
wrong?
They
said
like
well,
you
know
the
primary
key
could
overflow,
but
I
reckon
that
would
have
been
something
that
by
now
is
no
longer
an
issue
and
I'm
just
like.
Yes,
let's
like
continue
walking,
it's
fine.
D
The
this
like
civilization,
has
anyone
here
played
civilization,
so
in
the
first
civilization
game,
one
of
the
civilization
is
like
this:
it's
like
a
civ
builder,
so
you
have
like
everyone,
has
their
own
country
and
they're
ai
players,
and
each
person
gets
like
a
historical
figure
that
is
their
leader
right
and
in
the
first
version
of
civilization,
gandhi
always
went
straight
to
nukes,
and
people
were
like
whoa
whoa,
that's
really
intense,
because
what
happened
was
they
set
him?
They
set
his
passive
score
all
the
way
at
the
bottom.
A
A
So
you
know
the
the
video
and
the
song
I'm
talking
about
right,
well,
youtube
stored
their
view
count
in
an
integer
and
his
video
became
so
popular
that
it
overflowed
the
counter
and
they
had
to
quickly
fix
that.