►
From YouTube: Geo Scheduling Call 2020-07-28
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
A
All
right,
yeah
another
week
with
a
good
amount
of
items
in
the
closed
column,
so
nice
work,
everyone
will
go
through
these
first
one
is
mine,
moving
the
geo
deployment
from
rep
manager
to
petroni
on
the
primary,
so
yeah.
I
just
put
this
in
the
closed
column
today
got
a
documentation,
mr
merged,
to
our
validation
test,
section,
just
documenting
some
of
the
findings
in
this
issue.
We
created
some
follow-up
issues
and
there
was
also
a
demo
video
created
by
the
distribution
team,
highlighting
this
this
issue
as
well.
A
So
we're
going
to
shift
our
focus
to
supporting,
to
trying
to
implement
a
standby
cluster
using
petronia
on
the
secondary,
but
then
we'll
probably
revisit
some
of
the
issues
that
came
up
from
this
test
later
on.
B
A
That's
awesome,
yeah
great
work
and
congratulations.
A
And
then
next
one.
B
A
Cool
yeah
fun
part.
Indeed,
let's
see
here
and
then
yeah.
I
think
this
one
you're
not
assigned
to
it,
but
I
think
that's.
A
Cool
yeah,
thank
you
and,
I
think,
kind
of
related
to.
I
think
it
was
in
the
same
customer
issue
here
that
you
did
this
work
as
well.
Is
that
right.
C
Not
exactly,
but
I
think
these
both
issues
are
related
and
one
of
them
customer
haven't
seen,
but
I
think
he
would
see
the
second
issue
anyway
later,
so
I
just
fixed
it.
You
know
advance.
A
Yeah-
that's
that's
great.
It's
always
good
to
to
get
a
to
get
a
bonus
fix
in
with
with
what
you're
working
on
all
right
next
mic,
excluding
remote
stored
attachments.
D
Yeah,
actually
sorry,
I
wasn't
quite
quite
ready
earlier,
but
I
could
share
my
screen,
so
you
can
feel
it.
I
mean
be
more
able
to
take
notes
if
you
like.
Do
you
want
to
do
that.
D
Exclude,
okay,
so
yeah,
so
13
13.2
got
the
actual
fix.
Oh
wait!
Didn't
it
all
right.
Did
we
say?
Okay,
no
yeah,
13
2
got
the
actual
fix.
Well,
you
can
see
it
right
there.
That's
the
result
on
the
attachments,
progress
bar.
D
And
then
I
did
a
follow-up,
mr
to
add
tests
for
that,
because
I
basically
didn't
add
tests
for
it.
This
is
kind
of
going
to
be
relevant.
I
I
don't
already
called
it
out
for
the
framework
stuff,
because
everything
everything
is
now
like
going
through
replicas
replicables
for
a
geo-node
scope,
so
that
should
be
well
tested
and
yeah.
I
don't
know
if
we're
going
to
actually
be
able
to
dry
up
very
much
between
models,
because
the
relationship
to
project
and
namespace
is
different
for
everything
but
yeah.
D
So
so
this
got
closed
because
the
tests,
the
test
changes
got
merged
so
that
one's.
D
Good
okay,
valerie.
C
A
A
Awesome
and
that's
okay!
So
that's
specifically
s3
when
you
have
object
store
when
you
have
replicating
from
object,
storage
enabled
on
the
secondary
is
that
right
or
when
you're
when
you're,
when
you
don't
have
object,
storage
enabled
on
the
secondary.
But
you
do
on
the
primary.
A
Look,
I
think,
zach
had
to
hop
off
the
call.
Okay,
no
problem.
D
Oh
okay,
I
know
about
this
so
yeah,
so
this
is
kind
of
this
is
kind
of
what
happened
in
the
merge
request.
D
Which
zac
zach
helpfully
took
upon
himself?
You
know
back
end
work
here
so
now
this
this,
this
info
is
available
to
the
front
end
and
it
kind
of
like
aligns
with
the
data
that
is
going
to
be
available
from
replicator
classes,
so
that
kind
of
unblocked
his
other
work,
which
there's
more
discussion
about
which
I
haven't
gotten
to
reply
to,
but
I
saw
that
at
green
tea.
C
Yeah,
so
it
was
pretty
cool
to
us,
because
mike
has
helped
me
to
to
find
some
tricky
things
about
it,
but
so
we
almost
together
fixed
it.
So
thank
you
mike
and
I
oh
yeah,
I'm.
E
D
Yeah
yeah
so
geostatus
and
this
other
task
basically
didn't
take
into
account
the
framework
models.
So
it's
not
just
for
package
files
now,
if
a
new
replicator
is
enabled
they
should
be
taken
into
account
automatically
in
these
tasks,
so
I
think
yeah
that
worked
out
worked
out
well.
B
B
A
Can't
see
no,
I
she's
out
the
next
a
few
days
or
a
couple
days:
okay,.
D
Okay,
well,
it
looks
like
she
fixed
the
bug
in
the
qa
test.
D
That's
cool
zach
or
valerie
valerie.
D
F
F
Should
actually
have
been
in
verification,
because
the
main
merge
quest
is
merged,
which
is
great.
F
And
thanks
to
everybody
on
pitch
again,
so
I've
just
moved
it
to
verification,
because
that's
what
I'm
working
on
today,
it's
on
staging
already
and
I'm
just
going
to
check
if
it
introduces
any
regressions
because,
right
now,
this
replication
is
behind
a
feature
flag
and-
and
I
will
also
try
to
remove
it
and
check
if
it
works,
but
I
think
I
can't
see
it
in
the
ui.
So
that's
a
different
thing.
There
were
a
couple
of
other
mods
related
to
this
as
well
so
for
alex.
F
You
might
want
to
rebase
that
I
should
make
things
easier
and
there
is
some
documentation
update
coming
up
as
well.
I'm
just
waiting
for
the
final
review
from
the
docs
team
about
the
terraform
state
replication.
There
is
still
some
work
to
finish
off
this
main
big,
mr
I'm
not
sure,
is
alex
taking
it
over.
I
can
also
help,
after
I'm
done
with
some
more
vulnerability
work.
G
Merge
request
diffs
yeah
yeah.
My
plan
was
to
finish
up
the
the
current
work
on
terraform
state
and
then
see
kind
of
where
things
are
at
tone,
and
I
went
over
this
just
before
he
left
and
just
kind
of
talked
about
where
things
are
at,
which
is
to
say,
I
think
it's
pretty
close.
It
probably
also
needs
a
rebase
on
top
of
your
branch
and
there
it
needs
a
little
bit
of
work
on
its
selective
sync
code.
I
think-
and
otherwise
I
think
it's
pretty
close.
G
D
Yeah,
so
I
know
the
the
vulnerability
export
replication
doesn't
have
selective
sync
implemented,
yet
I
don't
recall,
there's
not
that
many
records
total
on
gitlab.com,
so
there
probably
aren't
on
staging.
So
if
you
do
enable
it,
it's
probably
fine
yeah.
I
do
have
one
other
thing
in
the
it's.
Well,
it's
kind
of
weird
because
it's
in
review,
but
what
I
did
was
earlier.
D
I
had
gotten
the
this
part
of
it
merged
with
this
one,
but
this
part
is
not
merged
and
I'm
kind
of
developing
it
right
now.
I
guess
I'm
almost
done
with
it,
so
it
will
be
actually
back
in
review
shortly,
but
I
did
want
to
mention
to
akriti
and
alex
that
it's
this
is
directly
relevant
to.
D
Oh
yeah,
I'm
not
sure,
okay.
Well,
basically,
I
think
I
made
a
mistake
with
the
excluded
by
selective
sync
method:
it's
just
not
comprehensive
enough,
and
so
the
fix
that
I'm
doing
for
this
is
I'm
checking.
Is
it
in
replicables
for
geonode?
D
So
that's
another
fix
that
I'm
doing
right
now
for
this.
So
you
know
that
parent
project
id
and
all
that
that's
going
away.
Sorry
about
that.
F
Okay,
do
you
have
an
mr
open
for
it
already.
D
I
don't
I
was,
I
no
problem.
D
Should
be
coming
today,
no
problem
and
I
can
helping
you.
D
E
D
Probably
I
don't
know
if
anybody
wants
to
speak
to
that
one
too
familiar
with
it.
A
I
can
yeah
eat
to
this
one
yeah
it's!
This
is
just
related
to
to
the
initial
testing
with
petroni,
and
we
should
be
able
to
close
this
one
out
too.
I
just
pinged,
larissa
and
hossein
on
there
to
see
if
they
were
good.
With
closing
this
one
out
since
we've
merged
that
the
docs
update
linking
to
the
testing
issue
and
we've
we've
created
follow-up
issues
out
of
this
one.
So
I
think
we
can
close
it
just
kind
of
waiting
for
for
the
okay.
F
Awesome
and
just
by
the
way,
zach
is
not
coming
back
to
the
meeting
because
he
had
a
fridge
disaster
and
he
needs
to
clean
up
all
day.
F
It
reminds
me
I
have
an
upcoming
disaster
to
take
care
of
in
my
fridge
as
well
before
that
happens,.
B
E
A
Yeah,
I
don't
have
really
any
updates
either.
Jenny
did
help
add
specifically
for
this
one
for
the
for
the
postgres
aj
on
the
secondary
epic.
She
did
go
through
and
add
quite
a
bit
to
the
descriptions
for
some
of
those
testing
issues.
A
I
think
there's
one
where
we're
trying
to
where
she
presented
a
couple
of
options
for
how
we
could
set
up
the
environment,
whether
like
we
could
either
use
orchestrator
and
then
kind
of,
and
then
do
some
manual
work
to
get
petrone,
enabled
or
potentially
use
the
performance
environment
builder
tool
that
that
quality
uses
to
load
test
reference
architecture.
So
we're
still
waiting
for
a,
I
guess,
a
firm
decision
on
which
way
we
want
to
go
but
yeah.
A
I
think
that's
the
update
for
for
quad
planning
on
on
this
one
at
least.
F
F
Your
update,
maybe
I'll,
also
look
into
your
workflow.
D
B
This
is
the
luggage
issue
to
remove
the
ftw
related
code.
I
split
this
into
three.
Mrs.
The
first
one
was
just
removed
the
feature
flag
and
was
measured
yesterday,
the
other
ones,
to
remove
a
bunch
of
finders
that
were
used
for
projects
and
weeks
and
it's
under
mundane
review
and
the
next
one
that
I'm
working
on
is
to
remove
the
find
that
wrapper
models
for
projects.
D
C
Yeah,
it's
tricky
one,
so
I'm
working
on
on
it
now
yeah
they
think
is
that
the
query
goes
through
the
workhorse
of
italy,
workforce
and
italy
again.
So
there
is
pretty
large
things
that
can
be
broken,
so
I'm
I'm
looking
into
it.
So
nothing
to
say
yet:
okay,.
D
Okay-
and
you
said
something
about
you've
kind
of
asked
for
help
in
chat
or
something
is
that
right.
B
D
Well
yeah,
maybe
maybe
we
could
look
at
it.
D
Together
at
some
point,
okay.
A
Yeah
he
provided
an
update
at
the
bottom
he's
been,
I
guess
just
like
learning
more
about
patrony
and
reading
up
on
on
docs.
He
said
that
in
the
next
couple
of
days
he
should
be
able
to,
I
guess,
recommend
an
approach,
but
I
think
what
what's
going
to
happen
here
is
is
whatever
we
come
up
with,
will
kind
of
be
an
experimental,
maybe
proof
of
concept
option
to
to
enable
this
on
on
the
secondary.
A
But
I
think
at
least
coming
out
of
13
3
will
the
the
goal
is
to
have
something
that
that
someone
can
ex
can
experiment
with
or
can
enable
an
omnibus
and
experiment
with
and
hopefully
will
have
any
considerations
any
manual
intervention
steps,
those
sorts
of
things
well
documented,
but
the
plan
is
to
have
a
better
sense
of
of
what
we'll
be
able
to
to
provide
in
terms
of
something
that's
available
in
omnibus
in
the
next
couple
days.
A
Yeah,
I
think
it's
a
little
tricky.
I
mean
the
yeah,
the
sort
the
there
are
just
some
unknowns
and
and
what
we're
starting
with
is
the
petroni
standby
cluster,
but
they're
just
some
unknowns
about
like
what
happens.
If
you
have
say
a
failover
in
the
primary
cluster,
you
know.
Does
the
standby
cluster
automatically
know
like?
A
Is
it
going
to
be
easy
for
it
to
automatically
know
to
follow
a
new
leader,
or
do
we
need
to
maybe
write
some
some
code
for
that
or
maybe
document
steps
that
you
need
to
take,
and
I
think
that's
the
the
sort
of
stuff
that
yeah
we're
trying
to
investigate
right
now.
A
D
All
right,
that's
it
for
the
in
dev
on
column.
Did
you
want
nick?
Do
you
want
to
take
over
or
would
you
prefer
if
I
just
keep
sharing
and.
A
Sure
yeah
I
can
take
over
real
quick
did
we
have,
did
we
go
over?
I
know
zack
had
we
talked
about
a
little
bit,
but
he
had
an
outstanding
question
on
the
progress
bars
and
so
just
wanted
to
highlight
that
issue.
Oh,
maybe
it's
not
even
on
the
board.
D
Okay,
so
yeah
he
kind
of
had
a
there's,
an
issue
where
we
call
we
only
use
projects
count
instead
of
repositories,
count
and
wiki's
count
here
so
yeah.
I
didn't
get
a
chance
to
get
back
to
that,
but.
F
Wondering
why
we
prefer
products
count
over
replicoms
and
wiki
cons
and
does
it
matter
so
much
if
we
have
all
three
from
the
api.
D
F
F
E
A
Yeah,
okay,
sure
cool
yeah.
I
think
that's
yeah.
I
just
wanted
to
make
sure
we
raised
some
awareness
around
that
and
had
a
either
just
confirmed
agreement
on
on
that
or
if
anyone
has
thoughts
on
it,
that
we
respond
to
him
there,
but
sounds
like
we're
we're
on
generally
on
board.
With
that
proposal,
cool
all
right,
thanks
mike
yeah
I'll
share.
My
screen
now
go
over
the
ready
for
development
column.
Yeah
not
too
much
has
changed
this
week.
The
themes
are
still
the
same.
A
We're
we're
moving
the
work
along
on
the
additional
data
types.
It's
awesome
to
see
that
progress
and,
and
then
the
backfill
work
of
course,
is
moving
along
and
and
it's
exciting
to
see
foreign
data
wrapper
code
being
removed.
Oh
am
I
sharing
my
screen.
Can
everyone
see
yep
see
my
the
billboard,
okay
yeah
and
then
it's
also
just
great
seeing
these
seeing
these
kind
of
one-off
bug
fixes
getting
getting
moved
along.
So
thanks
valerie
for
that,
so
ready
for
development.
A
Jenny
moved
testing
getaway
cluster
back
into
ready
for
development,
just
because
she
won't
be
able
to
get
to
it
for
a
couple
days
so
wanted
to
put
it
there
in
case
anybody
else
had
capacity.
So
that's
that's
a
still
a
high
priority
for
this.
For
this
release.
A
A
That's
the
other
big
13
3
planning
priority
that
we
wanted
to
get
started
on
the
release
and
with
the
idea
that
that'll
likely
span
a
couple
of
milestones
but
we'd
like
to
get
started
on
it
sometime
in
13-3
and
and
then
I
this
one
is
new
to
the
column,
which
is
enabling
usage
ping
on
geosecondaries,
and
this
is
part
of
our
northstar
metric
or,
I
think,
we're
calling
them
something
like
key
performance
indicators
now
or
product
performance
indicators,
which
is
to
measure
the
external
operations
performed
on
the
secondary.
A
I
think
that's
actually
getting
updated
a
little
bit
to
be
the
number.
Let's
see
what
josh
suggested
here
yeah.
So
instead
we're
going
to
try
to
measure
the
number
of
unique
performer,
unique
users
performing
any
external
operations
on
a
secondary,
which
I
think
aligns
with
some
of
the
the
gmo
that
gross
monthly
active
users,
some
sort
of
monthly
active
users
metrics
that
are
that
we're
trying
to
that
we're
trying
to
track
across
different
parts
of
the
product.
A
So,
as
a
first
step
to
that,
we
have
to
just
enable
usage
ping
on
secondaries,
and
so
I
added
an
issue
for
that
to
the
ready
for
development
column.
A
And
after
that
we
have
more
issues
for
the
additional
data
types
work.
I
didn't
really
kind
of
what
we
talked
about
last
week.
I
didn't
necessarily
prioritize
them
or
put
them
in
any
specific
order,
but
they're
here
and
I
think
the
owners
of
those
different
epics
have
have
a
good
sense
of
what
needs
to
happen
to
move
them
along.
A
So
I'm
I'm
just
kind
of
trusting
ocratee
and
alex
and
tone
to
manage
how
those
how
those
issues
get
moved
along
the
board
with
the
overall
idea
that
just
wrapping
up
yeah
finishing
those
data
type
epics
are
are
a
high
priority.
A
One
question
I
did
yeah
so
then,
so
some
of
those
are
also
in
the
open
column.
We
won't
worry
about
that
too
much,
but
one
that
I
did
have
a
question
about.
Was
this
one?
Improving
the
query
performance
when
object?
Storage
is
not
synced,
and
I
guess
I'm
just
wondering
what
yeah
what
we
need
to
do
here.
D
Yeah,
it's
it's
a
blocker
for
releasing
merge,
request.
Diffs
replication
like
we
could
certainly
merge,
merge
requests
just
replication,
but
we
we
just
can't
release
it
until
this
one
is
resolved.
A
D
So
yeah
tone
kind
of
like
laid
out
a
plan
for
that,
and
I
think
it
sounds
good
to
me
and
douglas,
but
there
is
a
bunch
of
work,
that's
required
for
that.
I
we
should
probably
just
modify
the
weight
right
now:
okay,
yeah,
it's
gonna
be
more
than
a
three.
I.
D
A
E
A
D
A
Okay,
yeah:
let's
do
that?
Let's
I
guess.
A
Maybe
we
can
put
a
10
on
there
just
to
signify
that
it
should
be
broken
up
some
more
and
and
then
I
will
move
it
to
to
the
top
of
near
the
top
here
since,
as
you
said,
it's
a
blocker
for
mr
external,
mr
diffs
and
yeah,
and
after
the
call
I
can
make
a
note
just
saying
that
we
talked
about
it
in
this
call
and
that
you
suggested
it
would
be
best
for
somebody
to
just
take
over
the
issue
and
and
figure
out
what
the
smaller
issues
and
order
of
work
should
be
because
it
sounds
like.
A
Maybe
we
can
focus
on
on
doing
all
this
for
external,
mr
diffs,
but
that
for
the
other
data
types
it
might
not
be
as
as
high
of
a
priority
or
as
much.
A
A
Yeah
yeah
that
sounds
good
cool,
okay,
yeah.
We
can
handle
that
off
the
call,
but
thanks
for
the
the
explanation
there.
A
Yeah
aside
from
those
items,
I
I
think
yeah
we're
yeah,
making
good
progress
on
additional
data
types
and
on
our
priorities
for
the
milestone.
Yeah
got
got
a
good
amount
of
items
in
the
ready
for
development
column,
continuing
to
move
things
along
at
a
good
pace.
So
yeah,
I
just
want
to
say
thanks
everybody
and
and
awesome
job.
Just
continuing
to
yeah
keep
keep
things
moving
along
the
board
yeah.
Does
anyone
have
anything
else?
They
wanted
to
talk
about.