►
From YouTube: Filecoin Core Devs Biweekly #21
Description
Recording for: https://github.com/filecoin-project/tpm/issues/49
For more information on Filecoin
- visit the project website: https://filecoin.io/
- or follow Filecoin on Twitter: https://twitter.com/Filecoin
Get Filecoin community news and announcements in your inbox, monthly: http://eepurl.com/gbfn1n
A
Okay,
all
right
good
morning,
good
afternoon,
everyone
welcome
to
the
21st
falcon
core
devs
by
weekly
meeting.
It
is
thursday
july
1st,
also
known
as
canada
day
here
in
canada.
I
think
we
have
reduced
attendance
today
for
probably
various
reasons
because
of
the
holiday
in
canada,
because
the
hyper
drive
network
update
was
last
week,
so
this
might
be
a
somewhat
quicker
meeting.
Currently
we
don't
have
representatives
from
forest
in
venus.
A
I
think,
but
that's
okay
cool,
so
we
can
kick
it
off
with
kind
of
the
big
book
the
the
file
coin
network
upgrade,
which
is
that
the
hyperdrive
network
upgrade
was
about
18
hours
ago,
went
successfully.
A
Folks
have
already
started,
taking
advantage
of
the
new
fips
and
the
technology
provided
by
snark
back.
So
we've
definitely
seen
some
pre-commit
batches
and
some
proof
commit
aggregate
methods
already
landing
on
chain,
which
is
great
yeah.
A
I
think
in
general,
the
the
the
upgrade
went
completely
smoothly
and
the
critical
partners
were
able
to
were
able
to
get
over
the
upgrade
and
are
and
are
thinking
well,
so
that's
very
exciting,
where
I
think
we
will
see
it's
a
little
early
to
see
to
see
numbers
based
on.
You
know
how
how
quickly
the
rate
of
onboarding
new
storage
has
increased
as
a
result
of
hyperdrive.
A
We'll
probably
get
a
better
picture
of
that
in
the
next
couple
weeks
to
months,
I
think,
as
as
people
start
to
as
miners
in
particular
start
to
explore
the
new
systems
and
see
see
how
that's
going
but
yeah.
Obviously
thank
you
to
everyone
here
who
is
a
part
of
the
upgrade?
I
think
venus
nodes
went
over
the
upgrade
with
us,
and
forest
and
fujan
nodes
are
shortly
behind,
I
think,
which
is
great
cool
in
terms
of
lotus
itself.
A
Yes,
obviously,
we
spent
a
lot
of
past
two
weeks,
preparing
and
testing
for
the
upgrade
and
we're
happy
to
see
it
go
successfully.
We're
planning
a
110
one
release
fairly
soon.
This
will
basically
just
be
fixes
to
the
lotus
miner
for
the
most
part
so
that
they
better
they
better
take
advantage
of
the
new
pre-commit,
batching
and
proof
camera
aggregates.
There's
a
lot
of
small
edge
cases
that
we
need
to
that.
A
We
need
to
fix,
and
so,
as
bug
reports
come
in,
we'll
continue
to
kind
of
iterate
on
that
fixing
them
along
the
way.
If
you
notice
something
odd,
while,
while
running
a
lotus
miner
and
trying
to
you
know,
batch
pre-commits
or
aggregate
proof
commits
please
report
them
we're
happy
to
to
get
them
fixed
as
quickly
as
possible.
A
We
also
cut
the
first
release
candidate
for
lotus
one
eleven
zero,
which
is
basically
about
two
to
three
months
worth
of
dev
work.
It's
mostly
bug
fixes
and
smaller
features,
we'll
be
testing
that,
because
there
was
a
little
bit
of
a
backlog
due
to
all
the
network
upgrades
we'll.
Probably
we
have
an
extended
testing
window
for
that.
We're
hoping
one
eleven
zero
can
come
out
can
come
up
sometime
in
the
next
couple
weeks.
A
Note
that,
based
on
the
lotus
release
scheme,
it's
an
odd
numbered
release,
111,
and
so
it
is
an
optional
release
and
so
folks
can
upgrade
to
it
whenever
they
want
to,
but
but
there's
some
there's
some
good
stuff
in
it.
I
think
that's
about
everything
from
the
from
the
lotus
team
updates.
B
Hi
today
maxine
was
absent,
so
it's
me
reporting.
So
as
for
the
mainnet
we
haven't
yet
we
updated
our
code.
Oh
well,
yeah
we
updated
the
to
the
hyper
drive,
spec
actors
upgrade
also.
B
We
have
detected
the
most
memory
consuming
parts
of
our
code,
so
we
stabilized
memory
consumption,
so
the
node
doesn't
consume
so
much
memory
and
the
process
gets
killed
so
yeah.
It
leaves
about
four
to
five
days
now
in
syncing
mode
and
but
there
are
still
some
bugs
somewhere
which
resulted
in
the
phones
we
haven't
caught
them
all.
B
Yet
also
maxim
is
currently
working
with
our
security
officer
to
this
scope
and
prepare
all
necessary
resources
for
the
project
for
the
code,
and
we
are
syncing
with
interrupt
nets,
so
it
seems
yep,
that's
our
dates.
A
Great.
Thank
you
very
much.
One
quick
question.
When
you
say
the
node
lives
for
four
to
five
days,
kind
of
what
what
happens
at
after
four
or
five
days
is
is.
Does
memory
can
is
memory
still
like
kind
of
ever
increasing
or
are
there.
B
Six
folds,
so
the
process
just
dice.
Okay,.
A
Interesting
that
sounds
difficult,
yeah,
well,
good,
good
job,
plugging
those
memory
leaks,
though
I
think
I
think
that's
great
cool
any
other
questions.
Sorry,
I
also
didn't
ask
if
there
were
questions
for
the
lotus
update.
If
anyone
has
any
I'm
gonna
take
them.
A
All
right,
sorry,
let's
have
an
update
from
kind
of
the
community
side
of
things
towards
that.
I
do
want
to
introduce
longfe
to
these
calls.
Long
faith
joined
the
filecoin
team
as
a
community
advocate
from
china,
so
lava.
If
you
want
to
introduce
yourself
real,
quick.
C
C
A
Thank
you
welcome
welcome
to
the
fun
squad
and
thanks
for
joining
these
meetings,
jennifer
do
you
have
a
community
facing
update.
D
Yes,
so,
like
you
said
last
night,
have
a
drive
happened
and
most
of
my
community
are
just
like
you
know
and
trying
the
trying
the
new
features
like
turn
their
like
configuration
to
make
sure
it's
the
best
for
their
money
set
up,
and
also
there
are
some
confusion
on
the
crypto
econ
set
of
the
scenes,
especially
on
the
batch
balancer
batch
discount
set
of
things
like
when
to
batch
is
the
most
and
the
most
economical
beneficial
for
managed
to
do
that
so
make
sure
if
you're,
a
minor
make
sure
you
follow
failed
announcement
in
file
coin
slack
closely,
we
will
share
like
the
most
up-to-date
updates
there
and
for
the
other
part
of
the
community,
like
clients,
developer
or
whoever
you
can
see
the
chain
get
less
congested
and
the
basically
is
really
low.
D
So
the
messages
the
gas
cost
supposed
to
be
like
and
cheaper
for,
like
making
deals
and
stuff
like
you
said,
the
full
impact
should
be
observed
in
the
over.
You
know
next,
like
few
weeks
so
stay
tuned,
we
will
have
more
updates
for
you
guys
later.
A
Thank
you
very
much
any
questions
about
those
updates.
E
Our
main
update
is
that
this
month
we
will
be.
I
can't
believe
it's
july
we
are
launching
the
mining
incubation
center
with
ipfs
force,
so
we
will
have
announcements
coming
up
soon,
but
very,
very,
very
thrilled
for
that,
and
then
dudley
and
sonya
can
jump
in
on
other
updates
from
the
security
and
advocacy.
E
B
B
B
Yes,
for
interrupt
net,
it
doesn't
take
the
gas
for
a
new
method
which
is
aggregate
seal
verification.
Also,
we
wanted
to
ask
whether
there
were
any
messages
which
called
the
aggregate
verification
method,
yeah.
A
So
two
things:
I
think
I
think
that
that
just
sounds
like
a
bug
in
in
lotus.
If,
if
it's,
if
it's
using
the,
if
it's
using
the
old
gas
price
list
for
interrupt
net-
and
it
wouldn't
surprise
me
too
much
because
interrupt
net
is
the
first
time
we
started
a
network
directly
from
a
version
of
actors-
it's
not
v0
in
this
case
b5
directly
or
I
forget
if
it
was
b4
or
v5.
So
so
that
sounds
like
a
bug
that
we
need
to
look
into.
A
It
might
unfortunately
mean
that
the
blocks
on
interrupt
net
aren't
like
it's.
Not
it's
not
correct
right
now,
but
that's
okay.
We
can
reset
that
network
easily.
To
the
second
question,
I
don't
think
any
pre-commit
batch
or
proof
commit
aggregate
methods
landed
on
interrupt
net
just
because
calibration
net
was
reset
at
kind
of
the
same
time.
So
most
of
the
testing
we
just
did
directly
on
calibration
net.
A
So
I
think,
if
you
want
to
check
your
cur,
the
correctness
against
a
test
network
calibration
is
probably
the
like
the
most
up-to-date
one
to
use
right
now.
A
Yep
sounds
good
yeah
thanks
thanks
for
reporting
that,
because
it
sounds
like
a
bug,
I'll
open,
a
lotus
issue
for
it
without.
A
Confirming
cool
any
other
questions
or
updates
or
just
points
of
discussion.
A
All
right,
I
think
we
can
wrap
it
up
here.
It's
been,
it's
been
a
busy
busy
couple
weeks,
certainly
leading
up
to
hyperdrive,
so
it's
nice
to
nice
to
slow
down
a
bit.
A
It's
too
early
to
start
thinking
about,
like
it
will
take
a
little
bit
of
time
to
chill
before
we
start
to
think
about,
like
the
next
network
upgrade
v14,
but
if
folks
have
ideas
for
like
stuff
that
should
be
getting
into
it,
whether
it's
tips
or
bug
fixes
or
anything
else,
we
can
bring
them
up
with
a
slack
channel
and
we'll
talk
about
them.
Probably
in
the
next
coordinates
call
once
once
the
picture
starts
to
emerge
a
little
bit
more
right
now.
A
I
think
our
focus
needs
to
be
on
making
sure
the
post
hyperdrive
network
is
stable.
So
if
we
can
all
do
that,
I
think
we'll
be
good
yeah.
I
think
we
can.
We
can
leave
it
there
thanks
everyone.