►
From YouTube: Geo Scheduling Call - 2019-10-21
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
Okay,
I'm
going
to
share
my
screen
and
I'm
prepared
this
time
and
then
I
have
everything
in
the
right
window.
So
I
thought
this
time.
I
would
just
open
up
the
discussion
page
first
to
just
remind
us
of
where
we
were
last
week.
Yes
last
week,
so
at
the
end
of
last
week
we
saw
that
there
were
four
epochs
currently
in
progress
that
we
were
right,
close
to
finishing
being
automatically
choosing
the
best
mode
object,
storage,
simplifying
the
upgrade
process
and
the
replication
of
design
uploads.
B
And
then
we
were
going
to
concentrate
on
replication
and
with
validation
framework
followed
by
gian,
get
led,
become
and
the
promotion
demotion
and
fail
Allah.
So,
let's
see
how
we've
measured
up
to
that
over
the
past
week.
Okay,
just
refresh
this,
so
know
that
it's
the
latest
thing:
okay,
okay,
there's
eight
in
closed.
Ok,
let's
go
through
those,
so
it
able
replicating
this
in
project
storage.
If
I'm
correct
this
was
the
flag.
Yes
to
remove
the
feature
flag,
yeah
excellent!
So
that's
done!
That's.
A
Done
so
there
was
one
discussion
that
you
mean,
so
we
decided
to
release
this
as
a
beta
feature.
Okay,
because
of
some
scaling
concerns,
so
part
of
the
sort
of
generally
availability
is
that
this
is
something
that
can
be
applied
at
any
scale
and
given
that
we're
not
doing
this
CENTCOM,
there
was
some
concern.
A
A
A
B
A
A
B
A
A
This
is
by
this,
so
I
had
a
customer
call
during
the
week,
and
this
came
up
and
I
think
this
is.
This
is
going
to
make
a
huge
difference
for
some
of
our
our
customers,
because
it
essentially
means
they.
You
know,
have
a
fully
transparent,
EO
configuration
option
which
is
really
exciting.
So
I'm
really
pleased
that
this
is
so.
C
A
B
Explicitly
said,
node
name:
ok,
it's
just
the
documentation.
Any
changes
right.
Yes,
I
think
that
swords
yeah
our
sec
to
take
a
look
at
this.
C
B
B
A
B
C
A
Have
a
quick
discussion
about
this
because
I
think
he
like
worked
on
it
with
I
can't
quite
remember,
and
they
said
oh
actually.
This
is
a
relatively
large
issue.
It
has
size
of
five.
We
can
split
it
up
into
two
things
and
then
they
just
ask
for
confirmation.
I
told
them
that
that
was
the
way
to
go
so
I'm.
B
B
A
B
B
A
Stays
blocked
and
I
think
we
actually
I
have
to
pick
this
up
with
tone,
because
the
the
reason
for
the
failure
in
the
upgrade
was
not
that
the
documentation
that
we're
trying
to
write
is
wrong,
so
I
think
I
think
we
need
to
make
a
decision
and
I'll
pick
this
up.
You
know
if
actually
like
the
steps
that
we
performed
our
we're
all
right,
but
it
failed
for
other
reasons
that
has
nothing
to
do
with
the
documentation
and
the
failed
kind
of
right
at
the
end.
A
B
An
issue:
okay,
okay,
yeah,
then
I-
think.
B
B
A
B
B
B
B
A
Think
we
are,
we
are
in
the
same
boat
here,
because
I
also
need
to
do
some
reading
on
it.
But
it's
really
good
to
see
that
this
is
being
picked
up
and
there's
interest.
Some
information
I
had
a
customer
call
last
week
and
the
customer
is
trying
to
implicate
a
pretty
big
installation
with
several
h.a
installations
and
they
raise
that
not
having
sort
of
consistent
verification
is
an
issue
for
them.
Yeah
not
say
anything
other
than
that
I
fully
agree,
and
so
I
think
this
is
a
good
thing
that
we
are.
B
B
B
Okay
horse
be
picked
up
and
that's
working
progress,
so
it's
not
really
much
to
you.
I
mean
for
anyone
watching
I
will
dig
into
all
of
these
further
off
to
the
core,
but
it
doesn't
look
to
be
blocked.
It
seems
like
it's
working
progress.
I
didn't
see
this
in
progress
last
week,
so
partial
success,
okay.
So
this
is
the
one
we
were
talking
about
where
we
need
to
talk
on
the
issue
later
as
to
what
the
next
steps
up
actually
are
here,
yeah.
A
A
Seven
or
something
to
find
know
from
eleven
to
fourteen.
Yes,
nine
point
six
point
eleven
to
eleven
to
two
fourteen
that
cannot
be
disabled
by
setting
a
priest
on
upgrade
hospice,
because
that
only
applies
to
major
versions
and
essentially
after
upgrading,
the
entire
cluster
and
and
following
the
instructions.
At
the
end,
you
need
to
shoot
ain't
that
beside.
C
A
A
So
after
going
through
the
works,
essentially,
the
the
only
thing
that
was
missing
was
updated.
Foreign
data
wrapper
tables
which
fails
because
the
underlying
post
goes
up,
the
database
was
updated,
and
so
the
library
version
that
the
foreign
data
wrapper
big
task
actually
uses
is
different
to
what
is
running
on
the
system.
So
it
fails
and
then,
when
up
restarting
posters
on
the
secondary
that
didn't
resolve
it,
we
started
posters
on
the
primary,
essentially
Center
Buster
into
disarray.
A
So
concerning
so
a
that
means
to
be
followed
up,
because
I
would
like
to
understand
what
that
actually
means
for
for
geo
customers
and
B
I.
Think
Tony
I
need
to
make
a
determination
if
you
know
the
instructions
for
for
this,
we're
not
really
the
problem
right.
So
we
can
unblock
the
the
Geo
upgrade
instructions.
I
think.
B
We,
it
also
means
that
we
need
to
figure
out
how
to
include
this
as
part
of
the
testing
when
the
stuff
is
released
with
omnibus,
because
that
must
mean
that
it
was
I
mean.
Obviously,
distribution
is
testing,
the
omnibus
releases
that
it
must
mean
that
they're
not
testing
it
in
the
Geo
clustered
environment,
yes,
and
we
need
to
come
to
some
agreement
as
to
who's
responsible
for
the
testing
and
how
to
get
it
right.
I.
A
Think
there
is
no
way
anymore
that
we,
like
I,
don't
know
what
they
do
right.
So
that's,
maybe
number
one.
So
confirmation
is
rightful
to
do,
but
going
forward.
I
think
all
of
those
scenarios
should
be
automatically
tested
because
at
the
moment
we're
doing
this
manually,
which
is
time
intensive
and
in
its
own
right,
I,
think
not
the
best,
the
best
time
investment
for
humans.
It
could
be
automated,
and
so
that's
really
important
and
also
you
know,
the
impact
of
minor
Postgres
upgrades
and
major
post
press
updates
in
the
future
should
be
I.
A
B
But
also
this
is
a
huge
wait.
I,
don't
want
to
say
waste
of
time,
but
it
is
a
it's
a
massive
expense
of
time
when
we
could
be
doing
other
things
and
all
of
this
stuff
we're
doing
it
over
and
over
and
over
again
so
I
think
I
mean
we
did
have
a
target
for
ourselves
to
50%
automate
this
stuff
and
for
whatever
reason,
it's
not
happening
for
a
number
of
different
reasons.
There's.
B
A
B
B
Ash
is
confirmed
with
Devon
that
this
is
that
the
code
is
the
right
place
and
he
said
when
I've
chatted
with
him
earlier
that
he
was
going
to
take
a
look
at
this
this
week,
so
fingers
crossed
he
can
get
to
that.
Okay,
so
that's
nine
things
that
are
in
development
well,
age,
things,
because
the
one
thing
is
I'm
just
going
to
drag
this
down
to
the
bottom,
because
it's
sort
of
for
interest-
it's
not
active.
Well,
it's
not
active
in
G.
Oh
yes,.
A
So
I'm
not
100%
sure
if
Mike
is
already
working
on
this
or
not.
If
I
would
just
make
a
question,
you
know
if
this
is
actually
in
development
or
not
because
he
assigned
himself
regardless.
This
is
the
this
is
important
to
do,
because
it
has
to
it's
one
of
those
issues
in
our
our
epoch
here
which
we
have
been
released
for
12.4,
but
would
like
to
release
or
12.5,
and
so
in
line
with
our
sort
of
strategy
discussion
from
last
week.
A
A
A
A
Nine
point
six
point,
fourteen
because
we've
updated
it
to
that
actually
already,
so
we
may
need
to
update
the
issue
description,
but
let's
see
expose
an
API
for
design
repositories
for
geo
admin
area
and
then
oh
yeah.
These
are
the
two
issues
we
put
into
the
work
ready
for
development,
which
are
the
finishing
the
front
end
work
for
an
abling
synchronization
of
design
repositories.
So
those
are
also
correct,
because
if
we
actually
have
to
like
reorder
this
a
little
bit.
A
Yeah,
so
their
education
is
ongoing
from
Valerie
showing
the
sink
information
is
ongoing
on
these,
here
are
probably
the
two
things
that
Zac
is
going
to
to
pick
up
so
again
in
line
with
our
discussion
from
last
week.
We
want
to
close
this
out,
and
so
I'm
happy
I'm
happy
with
that.
Okay,
okay,
so
this
one
here
yeah,
it's
the
upgrade
past
continuation
continuation.
This
is
a
little
bit
more
work
on
the
on
the.
A
Dr
so
the
impact
of
upgrading
a
cluster
for
dr,
so
I
think
this
is
a
quick
documentation-
sticks
happy
to
do
this
and
then
those
two
items
here
already
to
do
with
the
next
steps
for
jo-ann
calm.
Someone
is
the
server
architecture
diagram
for
geo,
which
I
think
we
should
do
think
you've
done,
or
maybe
done,
sometimes
I.
A
And
document
every
step
needed
to
replicate
a
new
data
type
I
think
that's
still
worth
doing,
especially
if
we're
thinking
about
a
potentially
like
better
way
like
a
more
framework
e
approach
for
their
educational
application.
I
think
capturing
the
status
quo
is,
is
a
nice
next
step?
That
being
said,
if.
B
A
So
I
think
my
point
would
be
I.
Don't
know
how
clear
is
what
we
actually
need
to
do
right
now
in
order
to
really
accomplish
this
and
I
always
feel
like
capturing.
You
know,
clearly
we're
not
happy
with
the
status
quo,
so
I
think
it
may
be
worse.
I,
don't
I,
don't
think
it
needs
to
be
in
superheater
just
to
capture
this
is
the
current
state
right
so
that
we
then
actually
can
say
these.
B
B
A
Would
be
review
work
to
do
and
they're
still
enough
to
pick
up
also
for
Zach
and
this
year,
for
example,
to
keep
people
busy
and
I.
Think
then
I
will
use
the
the
week
to
sort
of
capture
the
verification
and
replication
discussion
and
synthesize
that
information.
Maybe
we
can
even
do
that
partially
together,
so
that
we
are
we
on
the
same
page,
okay,.
A
Yeah
and
I
think
also
I
believe.
Most
of
those
things
here
are
are
scheduled,
but
I
will
need
to
make
like
I
can
do
this
after
the
call,
because
I
still
would
like
this
to
be
closed
off
right.
We
discussed
it
last
week,
so
I
made
like
if
some
one
or
two
may
not
be
in
the
in
the
the
scheduling
phase
yet
make
sure
that
that
actually
happens.
A
B
A
A
A
That's
done
and
this
year
actually
moved
forward
and
I
think
we
we
will
see
some
movement,
so
that's
good.
The
design
uploads
are
also
on
the
way,
as
discussed
so
I
think
we're
making
good
progress
and
then
the
replication
of
your
education
framework
is
something
to
think
about
yeah
and
we
just
need
to
make
sure
that
we
accelerate
as
best
as
we
can
on
the
geo
comm
front.
A
B
A
B
A
So
essentially,
what
I'm
saying
is
this
is
happening
disaster
recovery
to
verify
all
replicated
data
I
think
we
probably
have
to
rephrase.
This
is
the
framework
work
we
are
hopefully
going
to
close
off
the
geo
upgrade
documentation
in
12.5,
which
is
mainly
the
AJ
upgrade
it's
more
like
the
more
robust
foreign
data
table
updates
on
the
way
object.
Storage
is
mostly
done.
Design
repositories
are
well
on
the
way
we
are
going
to
start
thinking
about
promotion,
demotion
and
failover,
which
is
great,
then
in
november/december,
I
think
with
a
new
product
designer
starting.
A
A
School,
okay,
very
rough
timelines
right
ad
unreplicated
data
types
is
in
here,
but
that
makes
not
that
much
sense
when
we're
rolling
it
out
to
to
staging
at
the
moment
it
may
take
longer
and
have
to
revisit
this
with
you
finding
the
architecture.
I
hope
that
gets
done
in
the
next
month,
and
then
you
know
we
have
to
roll
it
out
to
dr.
A
A
I
think
this
is
the
state
of
this
is
the
last
thing
I
wanted
to
say.
I
think
this
is
actually
also
for
me
to
to
review
I
created
these
ad
and
replicated
data
types
and
verify
all
replicated
data
ethics,
and
they
were
I
think
obviously
the
sort
of
never-ending
and
very
large,
and
then
they
got
blocked
broken
down
into
smaller.
A
Chunks
of
you
know
what
specific
data
type
and
you
know,
verify
X
35,
Y
and
if
we
like
I
spoke
about
that
with
Gabriel,
and
it
hasn't
really
worked
out
as
well
as
I
hoped
in
the
sense
that
you
know
when
I
started,
I
thought.
Oh,
we
are
not
replicating
those
things.
Let's
just
replicate
those
things
right
and
oh
very
fit.
We
are
not
verifying
those,
let's
just
verify
them,
but
I
think
through
the
throughout
sort
of
the
attempting
to
do
this
work.
A
I
think
actually,
even
though
we
didn't
accomplish
what
we
set
out
to
do,
I
think
for
me
this
is
sort
of
a
the
point
now
where
we
have
to
take
this
learning
and
say
like
hey.
Actually,
this
is
really
hard
and
we're
not
really
solving
the
food
problem
here,
and
maybe
you
know
rethinking
how
we
do.
Those
things
is
actually
the
right
thing
to
do
and
know.
B
And
that's
why
I'm
glad
that
the
conversations
are
starting
now
for
how
these,
how
we
move
each
of
those
items
forward
because
yeah
it's
hard
and
cumbersome
to
do
it,
they're
a
big
long
and
I'm?
You
know
they
doesn't
really
feel
like
that.
They'll
ever
end
and
it's
nice
to
it's
nice
to
know
that
us
we
can
it's
done
on.
How
are
we
going
to
do
this
in
the
more
yeah.
A
And
I
think
from
a
court
perspective
I
think
they
are
also
interesting
sort
of
questions
on
you
know
scalability
for
like
get
lap
in
two
or
three
years
right
and
I
think
the
recognition
at
where
this
my
view
is
sort
of
currently
that
the
model,
the
operating
model
of
of
geo
doesn't
scale
as,
and
you
know,
we
can't
be
the
gatekeepers
for
for
adding
new
things.
People
are
going
to
be
able
to
add
things
much
more
quickly
than
we
are
going
to
be
able
to
add
them
to
geo,
and
so.
B
Yeah
and
I
think
that
the
the
idea
of
having
it
that
people
can
do
it
themselves
as
grown.
So
if
the
self-service
model-
and
there
was
even
discussion
with
Gabriel
again
about
having
a
self-service
model
for
the
backup/restore
as
well.
So
that's
something.
We
also
need
to
start
talking
about
at
some
point.
Spencer
and.