►
From YouTube: 2021-04-12 Multi Large Working Group 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
B
Thank
you
yeah,
so
we're
continuing
working
on
the
api
service
migration
at
the
moment,
the
the
main
thing
that
we've
been
spending
our
time
on
is
continuing
on
this
investigation
about
the
we
renamed
it.
So
it's
the
the
issue
discussed
last
week,
but
it's
we've
now
narrowed
it
down
to
it
is
a
service
discovery,
sometimes
failing
inside
of
kubernetes,
so
graham's
continuing
investigating
that
various
pieces
around
console
and
probably
we
need
to
upgrade
it
and
you
know
sort
of
narrowing
in
on
what
the
problem
is
going
on
there.
B
So
at
the
moment,
this
is
the
main
blocker
we
have
before
we
can
move
beyond
canary
with
the
api
service.
So
it's
in
progress,
though,
but
hopefully
it's
a
simple
fix
game
has
an
idea.
It
might
be
a
simple
fix,
but
it
is
still
an
unknown
at
this
stage.
A
Yeah
I
was
following
this
issue
and
I
looked:
jason
was
working
with
the
infrastructure
people
together
to
work
on
this.
So
I
was
following
this
issue
looks
like
we.
We
do.
I
run
into
some
kind
of
a
blogger
here.
B
Yes
exactly
so
we
certainly
don't
want
to
put
the
api
service
over
there
until
we
understand
what's
going
on
here
so
yeah.
At
the
moment
the
impact
is
manageable,
but
won't
be
for
the
api,
so
so
yeah
we'll
continue
doing
our
continue
working
on
that
one.
But
I
think
at
the
moment
we
have
the
help
we
need
and
then
on
what's
happening
next,
like
alongside
that
stuff,
we
do
still
have
various
kind
of
more
standard
migration
work
which
is
underway.
B
A
B
Yeah
there's
two
blockers
on
the
I
just
put
the
link
in
rather
than
rewrite
them.
There's
two
blockers
at
the
moment.
One
is
we've
just
discussed
the
service
discovery,
one
and
the
other
one.
It's
also
in
progress,
which
is
jason
earlier
registry
db,
migrations,
which.
A
Oh
that
that
one
was
a
new
one
to
me.
I
didn't
notice
that
okay,
so
it
looks
like
jason,
is
helping
the
team
right
now.
C
Yeah,
the
short
answer
is
we
inadvertently
included
something
that
ended
up,
causing
a
hiccup.
We
didn't
know
because
it
was
an
edge
case.
We
actually
hadn't
tested
we've
sorted
that
out
for
the
migrations
job
itself.
C
We
do
have
further
items
in
regards
to
the
upcoming
work
within
the
registry.
Regarding
the
metadata
we'll
be
touching
back
on
that
in
terms
of
what
needs
to
change
in
other
configuration
items
within
the
next
milestone.
A
Thank
you,
okay,
move
on
to
discussions,
so
I
discussed
this
over
as
american
offline,
so
looks
like
our
this
working
group,
the
work
or
the
the
blockers.
The
appearance
of
the
blockers
is
kind
of
a
cyclic.
When
we
kick
off
a
new
migration,
I
mean
a
migration
of
a
new
cluster
that
is
generally
when
we
see
more
bloggers,
we
or
we
discuss
more
blockers,
but
when
we
are
passing
the
first
stage
and
basically
moving
the
nodes
and
also
testing
the
the
blockers
discovery,
pace
slows
down.
So
it's
kind
of
a
cyclic.
A
So
I
was
proposing
that
we
changed
the
cadence
of
this
meeting
to
bi-weekly
fortnightly
during
the
low
tides
and
then
we
increase
back
to
the
weekly
when
we
are
in
a
high
tide
period.
So
that's
my
suggestion.
I
wonder
how
people
feel
about
it.
B
Yeah,
I
think
for
me
that
makes
total
sense
like
at
the
moment
we're
very
much
just
moving
through
fairly
routine
migration
work.
So
yeah
there's
certainly
less
that
we
require
from
the
working
group,
but
as
we
go
into
the
the
next
stage,
the
web
stage
then
yeah.
I
expect
we'll
have
the
need
to
to
have
people
available
to
help
us
with
that.
So
yeah.
That
makes
sense
to
me.
A
Okay,
cool
so
I'll
adjust
this
meeting
to
fortnightly.
For
now,
once
we
see
a
spike
of
the
new
discoveries,
we
will
restore
this
meeting
to
the
weekly
cadence
and
meanwhile,
don't
hesitate
to
ask
for
new
meetings.
If
we
are
in
a
in
a
week,
there
is
no
meeting
or
we
do
async
anyway.
So
we
can
do
everything
we
can
we
need
here,
but
now
I
will
just
slow
down
the
pace
of
this
meeting.
To
give
us
give
everyone
30
minutes
back
every
other
week:
okay,
cool,
okay,
next
amy.
B
Yeah,
so
it's
just
following
up
from
last
week-
and
I
was
just
I
saw
the
the
issue
for
disabled
serving
pages
from
legacy.
Storage
was
from
the
comments
it
looks
complete,
but
from
the
checklist
in
the
description
it
doesn't
look
complete.
I
was
just
wondering
whether
we
were
actually
new.
Are
we
currently
expecting
that
to
be
any
nfs
traffic.
A
Yeah
I
looked
into
that
epics
and
the
issue
looks
like
we
are
still
seeing
some
traces
from
the
nfs,
but
more
investigation
is
necessary
to
find
where
they
come
from.
At
this
point,
I
think,
according
to
the
update
in
the
issues
we
nothing
is
going
to
happen
anytime
soon.
The
priorities
are
given
to
the
api
migration
and
also
some
other
works
like
the
record
actions.
A
So
how
how
urgent,
I
think
we
want
to
migrate
over,
I
will
cut
off
as
long
as
possible,
but
at
this
point
there
seems
to
be
work
on
both
the
infrastructure
side
and
the
dev
side.
Infrastructure
side
is
basically
to
find
out
where
the
mf
nfs
traffic
comes
from.
A
B
So
on
the
infrasight
is:
is
that
so
because
sort
of
just
want
to
make
sure
I'm
kind
of
clear
in
mind,
so
we
started
looking
at
moving
the
batch
of
sidekick
queues,
the
batch
seven
and
we
saw
there
was
still
some
traffic.
We
didn't
do
anything
with
it
from
that
point
because
we
were
we
saw.
There
was
still
this
issue
that
we
needed
to
disable
serving
pages.
We
were
waiting
on
that
so
I'm
are.
A
I
forgot
it's.
I
saw
one
comment
either
from
jarv
or
from
scarback
is
there
is
some
traffic?
There
is
some
nfs
traffic,
probably
you're
right
from
the
psychic?
A
So
I
wonder
if
that's
a
a
stopper
here
to
kind
of
address,
I
think
yeah
so.
B
We're
expecting
that,
once
this
work
has
been
completed,
that
traffic
should
just
go
away.
I
don't
think
there's
like
additional
work.
It's
more
that
it's
this
from
the
comments
I
saw.
I
think
it's
that
it
wants.
The
legacy
storage
like
once:
we've
disabled,
the
serving
of
pages
from
legacy
storage.
That
traffic
should
just
naturally
go
away.
A
B
I
think
that's
what
we
need
to
confirm,
but
I
don't
think
it's
it's.
It's
not
a
so
we're
waiting
to
migrate,
the
sidekick
cues.
So
at
some
point
they
will.
They
won't
have
any
nfs
calls
on
them,
and
at
that
point
we
can
do
the
migration,
but
I
believe
the
work
to
stop
those
calls
it
should
be
covered.
I
believe,
by
this
disabled
serving
pages
from
legacy
storage
task.
A
Okay,
do
you
have
any
idea
which
which
teams
should
be
working
on?
This
looks
like
there
is
a
vladimir
here.
I
can
actually
give
you
opinion.
His
last
update
is
like
he
may
not
come
back
to
this
issue
until
14.2
or
three
I'll,
just
ping
him
to
see
the
latest
update.
That
was
one
week
ago.
A
Okay,
our
follow-up
looks
like
his
last
comment
was
six
days
ago:
14.0,
okay,.
B
We
have
other
things
to
do,
but
it
just
want
to
make
sure
we're
not
all
waiting
on
each
other,
so
that'd
be
great
thanks
and
then
yeah
number
three.
I
think
I
think
jason
yeah
you've
answered
this
one,
which
was
really
I
was
just
following
up
on
a
few
weeks
ago,
we
had
a
discussion
about
the
product
requirements
for
migrating
italy
to
kubernetes
and
yeah
trying
to
keep
track
of
the
the
work
there.
B
We
will
be
starting
to
plan
out
the
sort
of
the
technical
side
which
will
answer
your
your
point.
There
number
eight,
which
we
will
add
in
the
issue
onto
that
epic,
one
one
two,
but
before
we
do
that
we'll
check
in
on
the
product
requirements.
C
Right,
I
want
to
toss
out
there
keep
in
mind
that
the
issue
there
is
just
about
italy.
We
still
need
to
be
concerned
about
prefect
itself
and
how
that
works.
We
do
already
have
people
going
well.
Can
we
run
prefect
in
and
then
giddily
outside
and
the
that
we
still
have?
The
issue
of
scaling
is
problematic,
because
while
we
have
the
ability
to
have
quorum
and
to
manually
scale,
we
do
not
have
the
ability
within
gideon
prefect
right
now
to
adaptively
scale
automatically.
C
B
C
C
You
know
we're
basically
throwing
stuff
at
the
wall
and
seeing
the
breaks.
So
it's
a
matter
of.
I
don't
have
enough
deep
knowledge
about
what's
going
on.
I
just
know
it
doesn't
work.
B
Okay-
and
I
know-
we've
talked
before
about
having
a
gitly
representation
in
this
meeting-
that
I
mean
at
some
point
we're
going
to
have
to
have
sort
of
that,
or
certainly
I
guess
at
some
point.
This
topic
will
switch
to
being
the
focus
of
this
meeting
as
we
as
we
get
closer
to
that
migration
rate.
So
maybe
it
just
naturally
occurs.
A
Yeah,
I
will
yeah.
I
already
gave
the
em
and
pm
heads
up.
I
think
they
were
here
for
once,
but
now
we
don't
have
any
topic
to
them,
so
they
probably
just
temporarily
dropped.
But
when
we
are
ready
to
move
the
kidney
nose,
I
I
will
just
give
them
a
heads
up
again
to
rejoin
this
meeting.
A
Yeah
cool,
that's
the
the
last
item.
Anything
else.