►
From YouTube: 2021 01 05 Database Team Weekly
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
happy
new
year:
this
is
the
first
database
team
meeting
of
the
year,
so
jump
right
into
it
at
the
end
of
last
year,
two
weeks
ago
we
talked
about
setting
aside
a
time
in
these
meetings,
either
tuesday
or
thursday,
to
review
the
database
issues
on
the
infradev
board
and
one
to
sync
up
with
jose
and
figure
out
what
time,
what
day
works
best
for
you,
so
that
we
can
make
sure
it's
a
consistent
time
for
you.
A
So
if
you
need
to
go
because
you
have
some
other
high
priority
issues
to
work
on
that,
we
can,
you
can
go,
and
we
have
a
reliable
time
for
you,
instead
of
like
having
it
at
the
end
of
every
meeting
and
having
it
moved
so
to
jose.
What's
the
best
time
and
day
of
the
week
to
review
these,
is
it
tuesday,
the
thursday?
Do
you
want
it
at
the
beginning
of
the
meeting?
Do
you
want
it
a
half
hour
from
now
we'll
work
it
around
your
schedule,
whatever
works
best
for
you.
A
B
C
A
A
What,
when
we're
looking
at
these
issues,
how
do
we
look
at
them
and
know
that
they've
been
triaged
like
if
they're
in
a
milestone?
Can
we
effectively
ignore
them,
so
we're
not
triaging
them
again
or
how
do
we
know
that
they've
already
been
triaged?
What
kind
of
agreement
should
we
come
to
you?
There.
D
Why
so
I
can,
I
can
say
what
my
experience
is
in
geo
with
triaging
those.
So
the
first
thing
that
usually
happens
is
everybody
agrees
on
a
priority
in
severity
and
they
have
defined
meanings
as
to
when
these
are
actually
supposed
to
be
addressed
right.
If
it's
a
p1,
it's
essentially
do
it
now,
if
it's
a
p2,
it's
this
like
within
the
next
one
or
two
releases
right.
So
what
we
usually
do
is
we
leave
a
comment
in
the
issue
saying
this
is
the
severity
and
priority
we
understand?
D
You
know
what
this
is
and
from
that
is
essentially
from
that
point
is
a
commitment
as
to
when
we
will
address
them.
I
think
when
that
is
clear,
we
can
assign
a
milestone
to
them
and
say:
okay,
this
p2
is
going
to
be
in
that
milestone
and
maybe
for
craig
we
can
apply
a
deliverable
to
them,
because
then
we
are
essentially
saying
we're
going
to
do
this.
D
I
think
the
important
thing
there
for
me
is
sometimes
things
are
labeled
in
a
specific
way,
but
you
don't
quite
know
if
it's
really
that
bad
or
not
right.
So
there
may
be
a
little
bit
investigation
up
up
front
needed
to
actually
say
this.
Is
you
know
this
is
the
correct
labels
and
I
think
that
is
part
of
the
the
triage
that
we
are
committing
to
to
doing
essentially
weekly
or
whenever
so,
but
I
would.
D
A
I
think
to
that
point
I
agree
with
that,
but
the
way
we
can
further
filter
this
down
is
by
milestones.
So
once
everybody
agrees
and
adds
comments
and
we
come
up
with
severity,
then
I
think
the
next
step
would
be
to
add
a
milestone
to
it,
whether
it's
backlog,
because
it
turns
out
it's
not
that
big
of
a
deal
or
like
the
next
milestone,
so
that
then
we
can
further
filter
this
down
to
only
see
ones,
that
what
is
it
milestone
equals
none
right.
A
D
I
think
the
only
thing
I'm
going
to
say
is-
and
that's
maybe
just
me-
I
am
unwilling
to
put
milestones
out
there
that
are
like
four
milestones
off
for
me.
Milestones
are
at
most
the
next
milestone,
or
maybe
the
one
after
because
other
than
that
it
gets
really.
You
know
like
it
looks
as
if
something
is
going
to
happen,
but
reality
usually
differs
right.
A
Yeah
is
there,
I
know
in
some
projects
there's
an
awaiting
further
demand
milestone,
so
I'm
trying
to
figure
out
a
way
to
make
sure
that
some
of
these
issues
don't
fall
into
gaps
right
so
that
we
can
look
past
the
ones
that
have
already
been
triaged
and
then
figure
out
what
needs
to
be
prioritized,
and
maybe
it
just
means
that
there's
two
different
boards-
that
we
need
to
create
so
one
for
triaging
and
one
for
scheduling.
C
I
wonder
how
that
sort
of
relates
to
the
infradev
process
that
we
also
have,
because
what
we're
talking
now
is
that
we
as
a
group,
look
at
these
and
sort
of
help,
finding
the
priority,
severity
solutions
and
all
that
would
they
still
go
to
the
intro
death
meeting
and
being
like
further
triage
there.
Or
do
we
stop
that
from
happening
or
how
does
it
go.
A
Together,
if
I
recall
from
the
meeting
they
they
have
yet
another
board
that
they
look
at
and
if
they're
in
the
left
column,
where
they
don't
have
like
a
workflow
breakdown
to
them,
then
there
will
be
further
triage
like
they
want
to
make
sure
that
it's
scheduled
and
assigned
to
a
team
and
in
some
kind
of
workflow
validation.
So
if
it
doesn't
have
a
workflow
label
to
it,
then
yeah
it
will
be
triaged
yet
again,
or
at
least
it
will
be
for
assignment.
C
C
They
go
into
that
process
or
add
those
details
and
make
maybe
schedule
that
for
our
team
maybe
find
another
team,
and
if
we
don't
manage
to
do
that,
it
goes
to
the
introduction
process:
yep,
okay,
it's
actually
similar
to
janice,
and
I
discussed-
and
it's
on
the
agenda
on
the
fourth
point
for
because
we
often
have
these
issues
like,
for
example,
say
you
often
find
like
index
suggestions
and
stuff
like
that,
how
do
we
treat
those,
and
then
we
also
find
to
find
issues
where
that
are
kind
of
low
hanging,
sometimes
sometimes
not,
and
we
were
wondering
if
we
should
have
a
sort
of
a
separate
label
for
that,
although
they
should
also
go
to
the
intradepth
for
this
process?
A
E
Issues
like
address
found
a
page
that
was
taking
six
seconds
that
was
taking
six
seconds.
This
is
something
that
someone
has
to
address.
D
E
Have
to
figure
out:
are
we
going
to
address
it,
or
are
we
going
to
send
it
to
the
feature
team
that
is
responsible
for
issues,
for
example?
So
we
have
to
to
gather
those
all
those
inbound
requests
somewhere
so
that
we
can.
B
E
The
infrared
board,
can
I
ask
a
question
sorry
for
taking
us
back
here:
should
we
check
for
database
or
for
group
database,
because
a
lot
of
things
are
database
related,
but
are
not
group
database
anymore?
So
so,
for
example,
there
we
have
already
said
there
are
already
two
issues
that
other
groups
have
taking
over,
so
they
are
already
asked
and
assigned
to
other
groups.
So
we
are
checking
group
database.
E
A
C
B
A
Yeah
they're
different
projects
too,
which
is
the
other
problem,
so
we
have
different
projects,
different
labels,
different
criteria
for
looking
at
all
these
things.
So
I
think
ultimately,
we're
gonna
have
multiple
boards
but
narrowing
down
how
we
view
these
boards
and
in
what
order,
is
what
we're
trying
to
do
here.
C
And
something
that
we
had
had
recently
was,
I
think
I
was
creating
an
issue.
I
was
adding
the
infradef
label
to
it
and
then
it
was
sort
of
removed
later
on
with
the
comments
saying
that
well,
this
is
if
this
is
something
that
we
encounter
on
the
development
side.
Then
this
doesn't
qualify
for
introductive,
so
that
made
me
wonder
like
if
we
were
if
we
wanted
to
use
that
label
more.
B
E
If
we
want
to
triage
there
are
three
important
requests
for
maybe
it's
a
request
for
help.
Maybe
it's
a
request
for
anything
and
that's
the
question
there.
Should
we
have
a
database
three
hours
label
that
will
help
us.
You
know
have.
E
Move
them
in
our
workflow
forward
have
one
development
or
send
them
to
another
team
or
schedule
them.
A
I
think
adding
a
specific
database
triage
label
might
be
redundant
with
the
current
infradev
process.
I
think
if
it
has
the
infradev
label
on
it
and
database
label
on
it,
it
needs
to
be
triaged
and
that's
how
it'll
be
treated
in
the
meeting.
If
again,
if
it
doesn't
have
that
workflow
breakdown
or
team
assignment
to
it,
they
will
assume
that
it
needs
to
be
triaged
and
assigned
to
someone.
E
I
agree
with
craig,
but
those
are
two
different
discussions.
So
the
the
one
is
the
infrared
discussion
where
we
are
very
okay:
we're
going
to
check
all
informative
and
database
and
those
are
inbound
requests
from
the
infrastructure
team
and
then
there
are
all
the
other
requests
that
may
also
come
from
development
teams
and
feature
teams,
and
that's
the
discussion
about.
I
think
the
database
treehouse
also
where
we
don't.
We
cannot
assign
an
initiative
right.
A
Yeah,
typically
a
lot
of
times
those
will
come
through
with
just
the
database
label
right
so
you're,
proposing
that
we
add
there's
only
a
thousand
of
them,
proposing
that
we
add
another
label
on
there.
That
says
like
database
triage
and
then
is
that
what
you're
saying
so
that
we
prioritize
those
to
be
triaged.
E
D
So,
for
some
context,
what
I've
been
doing
we're
doing
here
is
a
euphemism
for
creating
the
label
and
looking
occasionally
at
it
right
is,
I
have
a
scoped
label
named
geo
triage
right,
so
that's
something
that
I
often
apply
to
things
and
I'm
like.
I
don't
really
know
what
to
do
with
this
thing.
We
need
to
think
about
it
right
before
we
do
anything
further
with
it.
D
That
is
mainly
something
internal
to
geo
right,
and
I
think
here
what
maybe
what
andreas
is
proposing
is
saying,
for
you
know
any
any
issue
right
where
the
database
team
needs
to
actually
look
at
it
and
decide
what
to
do
with
it
right.
Is
it
a
bug?
Is
it
a
problem?
Is
it
something
we
need
to
do
whatever
that
triage
means?
We
should
have
that
label
and
so
the
other
advantage.
This
has
for
me-
and
this
is
sort
of
a
wider
discussion
that
we
don't
need
to
go
into.
D
Is
I
also
have
a
geoactive
label?
That's
also
scoped,
and
those
are
the
the
things
that
are
actually
being
worked
on
by
the
team
at
this
moment
in
time.
That
allows
me
to
have
a
kanban
style
board
for
geo.
You
know
where
you
can
go
and
you
can
see
what
is
all
the
stuff
that
the
team
is
actually
working
on
and
then
they
have
workflow
labels.
D
That's
a
that's
a
little
bit
of
a
hack
because
I
it
avoids
having
a
board
with
900
open
issues
in
the
open
column
right,
because
I
find
that
very
disruptive,
but
you
have
like
9
000
in
the
closed
column,
900
in
the
open
column
and
stuff
in
between
it's
super
messy,
and
so
I
use
these
scoped
geo
labels
for
that
purpose.
That's
maybe
something
that
the
database
team
could
do.
D
D
A
I'd
say:
let's
take
this
to
an
mr
that
maybe
we
throw
up
on
our
database
team
page
and
we
can
discuss
the
mechanics
on
how
we
want
to
do
this
so
that
we
can
get
back
to
the
actual
issues
and
why
jose
is
here
to
see
if
there's
any
burning
issues
that
we
need
to
be
working
on
now
and
then,
as
a
team,
we
can
figure
out
how
we
function
on
on
triaging
incoming
issues
from
every
possible
source.
So
if
nobody
else
gets
to
it
I'll
throw
something
together
later
on
today,.
A
B
No,
I
did
just
one
investigation
about
a
peak
yesterday
that
I
think
was
a
select
for
update
that
we
already
was
in
our
radar.
I
guess
I
was
investigating
a
bit
today
or
talking
the
last
days
with
janis
about
the
marginalia,
because
we
are
like
somehow
aggregating
it,
so
it's
harder
for
us
to
trace
I'm
investigating
what
can
we
do
that
to
get
more
information
from
which
host
from
where
it
comes
these
kind
of
requests?
B
A
So
I
think
the
informative
portion
of
the
meeting
is
covered.
Then,
if
they're,
if
you
don't
have
anything
else
jose,
you
feel
free
to
leave
any
time.
A
All
right,
I'm
andre-
I
simply
this
is
your
topic
number
two
here.
C
Yeah,
probably
everybody
has
seen
that
already
right,
so
that
there
is
a
proposal
for
how
we
can
do
database
migration,
testing
or
database
testing
general
I've
linked
these
things
again,
if
you
haven't
seen
so
there's
a
a
code,
change,
basically
vmr
and
also
the
recording
from
yesterday,
and
the
readme
also
has
some
details,
and
this
is
kind
of
so
before
we
had
that
we
had
this
single
server
setup
that
we
were
using
and
it
had
like
network
isolation
to
some
extent
and
that
was
sort
of
a
manual
approach
to
this
and
works
fine
for
us
supposedly.
C
But
this
sort
of
takes
it
one
step
further
and
would
allow
us
to
run
this
on
any
ci
pipeline
without
handing
out
access
to
the
to
the
database.
So
that's
that's
the
benefit
and
I
think
we
can
do
a
lot
more
things
there
like
if
you
saw
that
what
gets
back
to
the
mr
as
a
feedback
currently
just
has
the
timings
for
the
migrations,
but
we
can
get.
We
can
add
so
much
more
detail
useful
details
to
that.
C
C
A
C
This
is
something
that
everybody
can
run
so
as
a
developer.
You
come
and
you
push
your
migration
into
ci
and
then
just
kicks
off
automatically.
C
So
basically
there
is,
this
doesn't
run
in
the
regular
ci
pipeline,
so
in
regular
ci
there
is
only
a
job
that
sort
of
kicks
off
another
pipeline
on
the
ops
instance.
So
this
is
a.
This
is
a
separate
private
instance,
and
this
is
the
one
that
we
care
about
from
a
security
perspective.
This
also
executes
the
code.
It
has
network
limitation
on
the
document,
docker
level
and
the
important
thing
is
kind
of
we.
C
C
A
B
E
C
And
it's
sort
of
overtaking
the
idea
of
of
working
on
the
anonymization
part
right,
which
is
a
really
large
topic.
If
you
think
about
it,
yeah.
D
I'm
very
pleased
with
that
specifically
because
I
think
the
whole,
let's
anonymize
our
production
data.
You
know
that
is
something
in
my
experience.
You
really
really
really
want
to
avoid
unless
you
really
have
to
because
it's
complicated
and
it
involves
lawyers,
and
that
usually
is
not
something
that
you
you
want
to
do.
D
Yeah-
and
I
had
a
couple
of
high-level
questions
on
this,
so
the
first
one
is-
I'm
quite
excited
about
this
because
it
feels
like
if
our
intention
and
goal
was
to
provide
people
better
information
on
their
migrations
on
production-ready
data.
This
feels
like
more
automated
way
of
doing
that.
D
Somehow
right,
I
know
this
uses
postgres
ai
right,
there's
a
lot
of
unknowns
here,
but
I
can
see
this
being
very
compelling,
because
this
is
a
problem
that
I
think
many
organizations
have
right.
So
I
think
this
is
maybe
something
we
should
think
about
and
say
like
if
this
is
something
that
works
well
for
us,
what
can
we
do
or
could
we
even
you
know,
maybe
it's
impossible,
but
is
this
something
that
others
our
customers
can
benefit
from
as
well?
D
D
This
has
enough
value
maybe
to
to
to
be
shipped,
but
I
don't
know
how
that
would
how
that
would
work
right,
but
it's
it's
sort
of
my
mind
immediately.
I
saw
that
I
was
just
like.
I
think
this.
This
is
really
an
interesting
feature
right
and
currently
we're
posting
this
as
a
comment
right
in
the
in
the
discussion
on
the
mr,
but
imagine
you
know
you
had
a
sort
of
a
little
tab
in
the
mr
that
says,
like
okay,
migration
report,
right
bam,
bam,
bam.
Everything
passed,
you
know.
C
Totally
agree
with
you
yeah
janice,
and
I
we
were
just
talking
about
similar
things.
For
example,
wasn't
there
a
didn't?
We
also
talk
sometime
about
dog,
fooding
database
reviews
and
what
we
can
do
so
yeah.
D
Exactly
that
was
the
discussion.
I
think
honestly,
like
at
the
end
of
last
year,
is
to
say
like
maybe
we
like,
I
think
this
is
also
interesting.
It's
like
these
are
features
right.
There
actual
features
for
developers
right
and
for
people
who
work
on
on
other
things,
so
maybe
database
reviewing
and
doing
these
kinds
of
things
is
a
category,
because
this
is
a
thing
that
I
could
see
being
quite
compelling.
Maybe
there
are
other
things
as
well
right,
but
something
like
that
is
probably
quite
valuable.
C
Yeah
and
it's
not
only
about
migrations,
for
example,
we
will
be
currently.
C
We
want
developers
to
come
back
with
query
examples
with
plans
and
all
that-
and
this
is
this-
is
similarly
manual
like
testing
migrations
and
what
if,
for
example,
you
had
a
tab
in
your
mr,
where
you
can
add
those
query
examples,
and
then
the
automation
kicks
in
grabs.
Those
runs
on
the
production
database
comes
back
with
the
create
plans,
and
that
would
be
super
cool,
yeah.
D
I
think
I
think
so
too,
and
this
is
the
kind
of
stuff
that
I'm
quite
excited
about
personally,
because
I
think
this
is
you
know
these
are
things
that
I
think
are
high
value
for
us,
because
they
increase
developer
productivity
right.
If
people
are
not
busy
copy
pasting
code,
they
are
busy
actually
doing
their
jobs,
which
is
developing
features
mostly,
and
so
this
stuff
is
quite
valuable
and
you
could
you
can
easily
see
how
this
scales
as
a
as
a
feature
category
from
free
to
premium
right.
D
You
can
easily
see
sort
of
people
starting
off
right
and
you
you
get.
Maybe
your
explain,
statements
right
and
that's
already
pretty
valuable
and
then
all
of
a
sudden
you,
you
start
seeing
like
well,
but
there
are
all
of
these
other
things.
I
want
to
do
right
until
you
arrive
at
maybe
something
like
this,
which
is
quite
valuable.
So
I
think
it's
a
it's
a
really
cool
category
idea.
E
But
yeah,
those
are
amazing
ideas
and
you
know
we
are
the
customers
for
the
for
this
and
we
know
the
value.
So
we
have,
we
are
feeling
the
pain.
This
is
one
of
those
cases
where
we
know
the
pain
of
having
to
do
that
and
how
useful
it
is.
I
don't
know
how
many
companies
need
that,
but
from
my
experience
everything
needs
something
like
that.
The
moment
they
have
a
they
are
working
with
databases,
so
yeah.
E
C
Sort
of
general
that
is
like
it
makes
sense
for
a
database
topic
or
a
database
review
category
in
expense,
but
is
there
even
a
way
to
generalize
this
more
like?
C
Are
there
other
types
of
sort
of
reviews
that
you
would
want
to
do
where
you
have
those
things
like
artifacts,
like
a
courier,
a
migration
that
you
want
to
feed
into
a
separate
system,
and
that
gives
you
feedback
on
the?
Mr?
I
wonder
if
sort
of
this
can
even
be
generalized
to
different
areas.
D
I
think
immediately
memory
comes
to
mind
right,
like
performance
concerns
right,
it's
like
we.
You
know,
I
think
we
talked
about
that
earlier
this
week,
where,
on
a
general
level
right
now,
it's
very
hard
to
know
what
feature
and
what
change
in
code
contributes,
how
much
to
specific
memory
usage
and
how
that's
impacted-
and
you
know
you
can-
you-
can
see-
I
think,
different
dimensions-
of
what
a
change
actually
does
right
to
your
system,
whatever
that
is
right
and
getting
these
these
reports
out,
you
know
and
also
sort
of
automatically
passing
certain
parameters.
E
Like
we
do
for
security
profiling,
so
maybe
it's
a
memory
profiling,
but
yeah
one
step
at
a
time.
So,
for
example,
we
have
to
I.
I
think
that
this,
if
we
could
never
do
something
like
that,
it
would
be
an
mvc
with
postgres
a
lot
of
people
use
postgresql,
because
this
is
specific
to
postgres
and
then
we
would.
We
would
have
to
think
about
yeah,
but
then.
D
Yeah-
and
I
think
that
there
are
obviously
many
questions
here
as
well
in
terms
of
security,
and
you
know
I
think
this
requires
like
postgres,
ai
and
thin
cloning
capabilities,
and
how
would
you
even
provision
something
like
that
and
so
on
and
so
forth?
I
understand
all
of
that,
but
I
think
if
we
think
big
a
little
bit
right,
what
we
just
discussed
is
quite
appealing
right,
and
I
think
it
also.
D
D
Cool
sorry,
I
didn't
want
to
subject
that,
but
I
think
it's
very
exciting.
A
Yeah,
I
think,
and
it
actually
leads
into
making
that
go
away
number
three.
If
we
have
that
epic
and
the
ideas
of
feature
development
that
we
want
to
work
on,
then
we
can
link
to
that
doc
to
say,
if,
given
more
time,
we
can
work
on
these
features
which
will
lead
to
you
know
potential
sales,
I
suppose
what
they're
looking
for
or
given
our
current
scale
and
level
of
interruptions
we'll
never
get
to
these
things.
However,
we
need
to
put
it
to
pose
it
more
positively
for
that
staffing
pitch.
So
yep.
D
No-
and
I
think
this
is
I
yeah
I've
revisited
this
already
janice
gibson
feedback.
You
can,
you
can
look
at
it
in
in
essence,
this
is
a
high-level
request
for
some
additional
team
members
for
databases.
D
I've
revisited
this
a
little
bit
in
in
the
morning
by
also
highlighting
sort
of
the
types
of
things
that
have
been
done
in
2021
already
right
and
why
they're
valuable
and
then
saying,
okay,
but
at
the
scale
that
we
are
right
now
right
and
gitlab's
growth
trajectory
we're
kind
of
blocked
right.
We
cannot
continue
doing
some
of
those
things
right
and
if
we
have
concrete
examples
like
this
one
in
an
epic
you
can
say
like
hey,
we
can't
actually
do
stuff
like
that,
but
it's
really
cool
right.
D
A
Cool
all
right:
next
one
did
we
cover
all
of
this.
E
Yeah,
but
this
is
this
was
not
about
the
infra
dev
only.
It
was
right.
E
Thing
and
then
what
I
think
I
think
that
fabian,
what
fabian
said
about
what
they're
doing
in
jio
sounds
pretty
interesting.
So
it's
something
similar
to
that,
so
not
only
for
refrigerator
so
yeah
we
can
think
about
it.
We
have
those
issues,
for
example,
and
we
have
to
think
you
know
how.
How
fast
can
we
decide,
respond
and
move
them
forward.
A
So
we
actually
have
this
issue
to
discuss
our
overall
team
processes,
and
one
of
them
is
actually
the
proposal
that
fabian
had
put
forward
about
how
geo
does
their
kanban
style
boards
so
prior
to
actually
having
an
mr.
If
we
want
to
discuss
asynchronously,
I
would
just
say
add
comments
to
this
topic,
so
we
can.
D
Continue
yeah
and
I
think
to
me-
and
I've
said
the
several
time.
I
don't
care
that
much
about
a
specific
process.
I
care
about
us
being
comfortable
with
it
and
getting
value
out
of
it.
I've
worked
in
different
processes
in
different
positions
in
gitlab.
I
find
the
the
geoprocess
that
we
use
pretty
helpful
for
developers
in
particular,
because
it
means
there
is
a
single
board
and
that
single
board
has
a
static
link.
It
does
not
change
right.
D
You
can
always
go
to
that
board
and
you
see
what's
happening,
and
that
was
one
of
the
biggest
issues
I
had
at
gitlab
is.
I
didn't
know
where
all
of
the
information
was
with
milestones
and
whatnot,
and
I
like
the
workflow
labels,
because
they
give
a
more
detailed
overview
of
where
you
are
at
in
the
in
the
cycle
of
actually
building
something
so
yeah.
A
Okay
and
then
database.
C
Relaxing
before
we
get
into
that
sorry
to
circle
back,
but
I
was
wondering
how
do
we
get
the
security
review
going?
You
have
any.
Is
there
somebody
dedicated
to
this?
I
saw
somebody
was
looking
into
that
topic
already.
A
So
he's
copied
on
all
of
our
planning
issues,
so
he
he
is
the
security
counterpart
for
enablement.
C
Should
we,
or
should
I
reach
out
to
him
about
that
topic,
yeah,
okay,
cool.
C
Why
don't
we
wait
for
some
initial
feedback
on
the
security
side,
because
that's
kind
of
what
makes
this
fly
or
not
right,
yeah.
E
C
C
C
So
basically
they've
been
a
cup
there's
a
bit
of
discussion
going
on
right
now,
because
the
database
size
is
growing
and
we're
basically
going
to
have
enough
free
space
anymore
and
that's
why
we're
kind
of
interested
in
doing
more
database
reindexing
and
I'm
suggesting
that
instead
of
doing
that
manually,
we
sort
of
spend
the
time
getting
the
automation
out
so
that
this
can
run
continuously.
Basically-
and
this
is
the
last
issue
I
linked
in
there,
this
is
currently
waiting
for
the
sre
team
about
the
graphana
integration.
C
That
needs
a
couple
of
secrets
to
be
added
to
the
chef
configuration.
That's
what
we're
waiting
for
there,
I've
pink
alberto
on
that,
but
once
that
is
done,
we
can
basically
discuss
the
schedule
for
the
re-indexing.
So
when
this
should
should
run,
I
remember
a
discussion
where
we
were
talking
about
running
this
on
the
weekends
or
running
this
during
during
the
day,
and
there
is
benefits
and
drawbacks
for
both.
C
Basically,
so
that's
what
we
would
have
to
discuss
and
then
likely
what
is
missing
is
a
run
book,
basically
for
the
encore.
How
do
you
interact
with
that?
How
do
you
stop
things
if
things
go
wrong
on
stuff
like
that,
but
other
than
that?
It's
basically
ready
and
waiting
for
srv
at
the
moment.
A
What
you
mentioned
earlier
that
we're
running
out
of
database
space
to
run
the
re-index,
what
size
are
we
at
now.
C
A
C
And
but
also
the
re-indexing,
I
mean
this
is
going
to
help
a
lot.
If
we
manage
to
remove
bloat
continuously,
then
that's
a
lot.
That
is,
I
think,
a
terabyte
is
so
that
we
can
shape
off
overall,
but
sometimes
we
have
to
increase
the
disk
size.
I
guess
yeah
in
any
case
and.
D
D
Oh
something
that
comes
to
mind.
We're
done
on
that
topic
is
that
correct,
yep,
okay,
just
popped
into
my
my
head.
Have
we
addressed
all
of
the
integer
overflow
issues?
D
E
Helpers
and
one
of
the
the
next
days
we
will
start
the
migration
for
the
biggest
table.
The
the.
E
It
will
take
one
release
and
then
next
release
and
during
the
time
we'll
maybe
have
to
add
some
additional
migration
helpers.
We
have
to
think
about
it,
so
this
is
done
and
we're
waiting
for
for
it
to
go
to
production.
So
then
we
are
going
to
release
the
step
one
we
will,
which
will
start
the
the
migration
and
also
this
milestone.
We
should
add
any
other
remaining
migration
helpers.
That
will
enable
step
to.
D
E
We
have
a
dashboard
after
addressed
updates,
so
we
can
also
check
live
how
things
are
working.
D
You
can
tell
I'm
totally
on
top
of
things.
You
know
after
the
vacation,
to
give
you
an
example
of
that
I
made
coffee
in
my
in
the
kitchen
like
before
the
meeting.
They
poured
all
of
the
coffee
grounds
into
my
cup
rather
than
into
the
coffee
maker
right.
So
I'm
I'm
doing
well.
D
No,
what
happens
to
me-
or
it
has
happened
to
me
at
least
twice-
is
I
have
coffee
with
milk,
but
no
sugar
and
instead
of
actually
pouring
milk
into
my
coffee
in
the
morning,
I
I
put
in
orange
juice
and
then
I
get
to
sit,
but
I
was
just
like
this
is
terrible,
just
like.
Why
is
it
happening
it's
like,
and
if
your
brain
does
not
process
the
difference
between
you
know
a
milk
carton
and
the
translucent
yellow,
shiny
orange
carton.
You
know
that
you
shouldn't
make
any
important
decisions
on
that
day
right.
That's.
E
A
E
A
local
greek
type
of
like
a
very
distinct
wine.
D
Yeah,
well,
I'm
I'm
glad
you
you
spit
it
out
and
I
hope
it
was
by
accident.