►
From YouTube: Filecoin Core Devs Biweekly #11
Description
Recording for: https://github.com/filecoin-project/tpm/issues/24
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
Alrighty,
hello,
everyone
hope
everyone's
having
a
good
time.
Welcome
to
the
11th
cordova's
bi-weekly
meeting
today
is
thursday,
the
11th
of
february
couple
more
people
popping
in
yeah
usual
usual
agenda
today,
we'll
start
with
updates
from
the
various
folks
and
then
there's
a
couple
of
items
of
discussion.
Note
that
the
representatives
from
venus
aren't
joining
because
they're
celebrating
spring
festival
in
china,
so
we
wish
them
happy
holidays.
A
They
did
share
an
update
that
I'll
read
out
for
the
benefit
of
the
recording,
though
yeah
yeah
cool.
Let's
kick
it
off
with
maybe
forest
folks.
What
updates
do
you
have.
B
Hey
yeah,
so
I
mean
it's
definitely
an
exciting
time
in
forest
land
we
are
syncing
with
mainnet.
We
have
two
nodes
running
on
there
right
now
and
they've
been
running
for
the
last
couple
of
weeks.
I
think
we
had
a
couple
small
discrepancies,
since
the
last
call,
but
yeah
they've
been
fixed
and
and
things
are,
are
looking
great
and
I'm
not
sure
if
we
mentioned
this
last
call
or
not,
but
we
can
also
produce
blocks
on
like
a
devnet,
which
is
great.
We
obviously
have
not
done
that
on
main
net.
B
Yet
just
because
of
like
the
hardware
requirements,
but
we
are,
we
would
like
to
do
that
at
some
point,
so
yeah
just
figuring
out
the
best
way
to
actually
get
to
that
point.
We're
also
I
mean
like
because
we
got
to
get
got
to
this
point,
doing
like
a
bit
of
a
documentation.
Push
so
hopefully
you'll
see
some
more
stuff
in
terms
of
just
like
yeah
documentation
in
the
near
future
other
than
that
we're
also
working
on
updating
to
v3.
B
So
the
hampton
and
the
amp
are
done
right
now
and
I
think
eric
started
the
actors
upgrades
yesterday.
So,
just
working
on
that
and
also
we
are
updating
our
pc
to
just
make
sure
it's
at
least
somewhat
compatible
with
lotus,
at
least
that's
the
goal.
B
I
mean
we
talked
a
bit
about
that
in
our
call
last
week,
but
we
were
working
on
that
and
just
adding
http
support
as
well,
because
we
had
it
before,
but
then
we
removed
it
when
we
were
adding
web
socket
support,
but
now
yeah,
adding
it
back
in
yeah
trying
to
think.
If
there's
anything
else,
major,
we
are
hoping
to
have
an
audit
done
in
early
march.
B
So
yeah,
I
think
we're
going
to
be
doing
that
with
sigma
prime
deep
from
the
pl
team
is
helping
us
arrange
that
but
yeah,
I
think
I
think
that's
about
it.
We
have
some
personnel
changes.
I
think
this
is
going
to
be
austin's
last
core
devs
call,
I
don't
know
austin
if
you
want
to
just
say
a
few
words.
C
Yeah,
I
wasn't
really
planning
on
saying,
but
yeah
like,
as
my
last
day
is
tomorrow,
so
like
I'll
still
be,
I
guess
around,
but
I
won't
be
like
actively
developing
on
forest
yeah.
That's
pretty
much
it
and
for
the
for
the
update,
there's
just
one
small
point
is
just
that:
there's
only
one
small
consensus
issue,
which
was
a
week
ago,
so
the
node
had
been
running
for
a
week
and
then
there
was
a
consensus
issue
and
then
it's
been
running
since
so
yeah
everything's
pretty
much
good.
There.
D
A
Nice
very
nice
yeah,
it
isn't
technically
saving
time.
Six
six
plus
days
of
staying
in
sync
with
maintenance
is
pretty
cool,
with
no
issues
and
yeah
austin.
Obviously,
sorry
to
hear
that
you
will,
you
will
not
be
part
of
this
particular
squad
anymore,
and
thank
you
for
all
of
your
contributions.
It's
been
great
yeah
cool
with
the
fulham
folks
check
your
work.
E
Yes
sure
everyone,
so
for
us
it
was
also
quite
an
exciting
time.
We've
been
working
on
the
performance
improvements
and
in
the
last
call
I
mentioned
that
we
had
faced
an
issue
where
we
can
not
catch
up
with
the
main
net
due
to
the
performance
issues
in
synchronization
process.
E
So
we've
made
decent
efforts
in
improving
that
shout
out
to
austin
for
helping
us
out
to
point
at
the
features
that
could
help
us
in
improving
the
performance.
So
we've
been
able
to
improve
from
five
and
a
half
minutes
per
block
to
11
11
seconds
on
average
per
block.
So
now
we're
in
good
shape
to
continue
with
synchronization
and
currently
well.
We
have
restarted
our
node
today.
It
has
downloaded
all
the
data
and
now
it's
in
the
process
of
sync.
So
hopefully
no
no
issues
will
be
found.
E
Well,
probably,
there
was
some
will
be,
but
they
will
not
be
as
critical
and
we
will
see
how
interpretation
will
go
and
catch
up
and
think
of
the
note
now
after
afterwards,
so
we
are
still
working
on
the
wrapping
up
with
actors
we
do
we
have
we
need,
we
decided
to
need
some
more
or
less
big
refactoring
on
some
of
our
actors
due
to
well
changes
in
actors
we
0
to
v2,
so
we
decided
to
take
out
time
and
do
it
now,
when
we
have
a
time
for
that,
so
we
have
finished
with
more
or
less
everything
except
two
actors
which
are
storage,
actor
and
store.
E
E
And
we
are
going
forward
to
implement
them.
I
think
well,
I'm
not
sure
about
the
next
weekly
meeting,
but
probably
the
one
months
later
I
will
wrap
up
with
them
for
sure.
E
So
our
aim
is
also
to
request
for
the
security
audit
after
we'll
have
our
node
for
at
least
a
week
or
two
running
on
the
mainnet.
So.
E
If
everything
will
go
well,
probably
on
the
next
meeting,
I
will
announce
that
we'll
have
will
start
scheduling,
security,
yep
and
I
think
that's
mostly
it
from
from
my
site
yeah
and
we
are
going
to
start
miner,
probably
on
a
test
net.
I
think.
A
Yep,
that's
it
from
my
site.
Very
cool
thanks
yeah
sounds
like
a
lot
of
progress.
It's
really
cool
that
all
of
the
implementations
are
kind
of
kind
of
getting
there
getting
to
the
same
place.
At
the
same
time,
good
work
I'll
quickly
read
out
the
update
that
stephen
from
the
venus
team
shared
with
us.
A
He
says
the
team
is
working
on
a
venus
remote
wallet
for
some
new
features
like
encrypted
key
store
for
private
keys,
they've
created
venus
miner
and
venus
sealer
projects
to
make
it
work
and
to
support
a
distributed.
Mining
pool
down
the
road
which
is
great
and
they've,
also
been
working
on
documentation
and
guidance
for
developers
and
other
users,
so
the
documentation
push
kind
of
around
the
world,
which
is
great
all
right.
Well,
the
lotus
update
is
much
less
exciting
than
all
of
you.
A
Folks,
yeah,
we've
kind
of
had
quite
a
few
weeks
internally.
Pl
is
working
on
kind
of
re
redirecting
for
our
2021
mission.
So
for
the
most
part
load
aside,
we've
been
testing
out
the
1.5
release,
150.,
which
we'll
be
introducing
v3
actors.
We
were
hoping
to
get
it
out
before
a
chunk
of
east
asia
went
on
holidays
but
decided
not
to
rush
it.
So
we've
been
testing
that
in
the
time
since,
and
we're
now
at
that
point
where
we
have
the
confidence
to
put
it
out.
A
We're
thinking
last
week
of
february
first
week
of
march
is
when
the
actual
upgrade
will
happen
with
the
release
coming
out
somewhat
earlier
than
that
yeah
hit
some
bumps
in
the
testing
found
some
bugs
in
in
b3
across
code.
So
it
was
good
that
we
were
doing
that
I'll
quickly.
Mention
that
as
part
of
that
testing,
we
unfortunately
took
down
the
calibration
network
just
because
it's
a
calibration
network
has
a
lot
of
mining
power
distributed
across
the
various
miners.
A
By
this
point,
and
so
it's
hard
to
quickly
iterate
on
upgrades
there,
and
so
the
failed
upgrade
causes
a
bunch
of
miners
to
get
slashed,
and
then
everyone
else
is
not
operating
so
so
that
network
needs
to
be
reset
and
we're
working
on
that
and
also
working
on
figuring
how
we
can
stably
deploy
networks
again,
because
that's
kind
of
been
an
issue
lately
yeah.
I'm
pretty
sure.
That's
it
with
with
lotus
side
updates.
D
Cool,
I
have
a
question
there,
so
so
I
mean
you,
you
mentioned
that
there's
some
bugs
in
the
actors
code,
and
so
I
was
wondering
like
when
a
release
for
that
is
gonna
be
made,
because
I
mean
right
now,
I'm
targeting
the
v0
v3.0.0
tag.
I
I'm
guessing,
that's
probably
the
one
that
has
bugs
in
it.
D
D
A
Bugs
yeah
totally
yeah,
I
realize
we
haven't,
we
haven't
tagged
one,
yet
we
should.
So
if
you
like,
the
the
fixes
are
in
master
and
they
should
be
the
only
committed
master
between
them,
but
we
should
just
tag
a
fixed
release.
A
And
they're
all,
oh
so
yeah
the
fixes
are
one
line
in
the
in
the
v3
and
then
two
one
line
fixes
in
the
actual
v3
actors
themselves.
I
would
be
curious
to
see
if
folks
wound
up
kind
of
making
the
same
bugs
I'd,
be
I
I
wouldn't
expect
it
to
because
they
were
all
very
much
like
that's
the
wrong
variable
name.
You
use
there
or
something
like
that.
A
Yeah
yeah
thanks
that
question.
Any
other
questions.
A
A
Just
because,
like
that's
the
one
topic
where
we
really
want
input
from
everyone
to
be
fully
aligned,
so
maybe
let's,
let's
wait
a
couple
weeks
to
have
that
conversation,
if
only
because
we're
not
planning
the
next
network
up
which
shouldn't
be
for
a
while
for
1.5,
maybe
we'll
start
a
thread
async
to
to
get
preliminary
thoughts
on.
That,
though,
also
wanted
to
quickly
talk
about
fib
12.
Let
me
drop
the
link
in
here.
A
Jennifer
adds
to
let
her
know
if
you
need
any
help
with
the
community
comms
for
any
of
the
stuff,
that
of
all
the
exciting
stuff
that
y'all
have
going
on.
Jennifer
has
good
access
to
the
community.
So
if
there's
something
you'd
like
shout
it
out,
let
her
know
yeah
so
fib
12
is
a.
Is
a
nice
simple,
flip
open
from
our
this
kind
of
was
motivated
by
our
product
team?
A
So
right
now,
there's
a
bit
of
an
irritating
feature
in
the
verified
registry
actor,
which
is
persisting
right
into
b3,
which
is
the
the
one
line
version.
Is
that
a
verified
client
cannot
cannot
have
their
data
cap
topped
up
so
a
verifier
gives
a
verified
client
some
data
cap
and
unless
they
use
it
all
the
way
down
to
zero.
That
address
cannot
have
its
data
cap
topped
up.
A
This
was
implemented
primarily
as
a
security
thing
where
we
didn't
want
anything
to
be
exploitable
here,
and
so
we
just
said
use
a
new
address
every
time,
if
you,
if
you,
if
you
want
some
more
but
as
you
can
imagine,
this
isn't
great
user
experience
and
so
there's
a
fairly
strong
ask
from
the
various
partners
that
we're
working
with
to
allow
people
to
reuse
their
addresses
as
verified
clients
and
have
that
data
cap
be
topped
up.
A
So
that's
basically
the
entirety
of
this
of
this
feb
it's
from
the
early
stages
and
we're
not
probably
not
moving
on
it
and
implementing
it
soon,
not
on
the
v3
upgrade,
but
yeah
take
take
a
look
at
it.
Any
preliminary
thoughts.
A
Yeah
that
that's
the
hope,
as
always
the
thing
to
think
about
is
whether
there
is
a
somehow
whether
this
is
exploitable
for
an
attack.
Somehow
the
security
consideration
section
is
somewhat
fleshed
out,
but
yeah.
I
I
agree
that
that's
fairly
uncontroversial.
A
All
right,
cool
yeah,
as
always
the
discussion
can
continue
in
the
first
report,
but
I
wanted
to
bring
that
to
folks's
attention
I'll,
let
you
know
how,
when
we
decide
to
move
it
along
or
we'll
decide
when
to
move
it
along
collectively.
As
a
group,
I
don't
think
it
has
to
happen
just
yet
cool
any
other
questions
or
comments.
A
One
thing
I
did
want
to
bring
up
is:
yeah,
sorry
that
we
still
don't
have
stable.
We
I
feel
like
we.
We
made
negative
contributions
in
terms
of
providing
stable
test
networks
and
that
we
took
away
one
I'll
reiterate
that
it's
a
priority
to
like
we're
just
trying
to
make
this
work
properly,
which
is
why
we're
taking
our
time
with
it
and
making
sure
that
we're
not
going
to
be.
You
know
both
from
the
what
is
what's
wrong
right
now
in
terms
of.
A
Ideally,
we
have
one
that
we
can,
that
we
have
a
lot
of
control
over
and
then
there's
one
that
miners
are
testing
on,
because
that's
a
very
different
kind
of
requirement,
I'm
hoping
that
we
will.
These
networks
will
be
born
next
week.
Sometime.
I'm
sorry
again
that
it
has
a
target.
A
Yet
questions.
A
Nice
sounds
like
sounds
like
everyone's
just
like
sprinting
towards
the
finish
lines,
or
I
guess
they're
not
finish
lines
we're
just
sprinting
towards
the
goal
post
for
now,
which
is
great
jennifer
denver
says
that
the
community
suggested
we
have
a
test
net.
That
requires
a
sign
up
in
some
capacity
interesting.
What
that
would
look
like
that
sounds
like
a
good
idea.
Yeah.
F
So,
like
basically
the
idea
because
of
the
150
situation,
they
think
if
miners
can
extend
up
for
the
test
net,
so
we
may
have
like
more
participation
like
during
the
network
up
upgrades
on
a
test
net,
because
you
know
it's
not
like
incentive
or
anything
like
for
doing
it.
It's
not
necessary
for
them
to
do
it
as
well,
so
have
an
exclusive
like
list
sort
of
setting
may
help
there
it's
coming
from
the
community
and
seems
like
quite
a
few
personal
agree
to
that.
I'm
just
wondering
how
folks
think
about
this.
A
A
good
idea
for,
like
the
minor
focus
to
test
that
it
is
a
little
like
we're.
You
know
we're
controlling
it,
but
it's
a
test
network.
So
it's
fine
to
say
that
we're
in
charge
here.
I
think
I
think
it's
interesting
proposal.
D
Yeah,
I
mean
something
I
mean
I,
this
is
probably
too
much
but
like
something
almost
like
the
space
trace
network
and
keeping
that
alive
would
be
kind
of
nice.
I
don't
know
and
like
lowering
hardware
requirements
for
that
as
like
a
minor
test
net.
That
could
be
an
interesting
way
like
I
know.
A
few
other
networks
have,
like
you
know,
quote
unquote:
test
nets
but
they're
like
not
test
nets,
but
they
are
test
nets.
D
I
think
that's
what
kusama
on
polka
dot
is
that's
at
least
that's
what
I've
been
told.
I'm
not
too
familiar
with
it,
though,
but
yeah,
something
like
that
could
be
nice
as,
like
you
know,
yeah
like
a
betagram
or
something.
A
Yeah
yeah,
I
will
say
so
a
couple
things
there.
One
of
the
things
is
yeah.
The
lowering
hardware
requirements
is
kind
of
one
of
those
tricky
things
where
that
that's
kind
of
what
we
wanted.
Our
test
network
to
always
be
is
just
512
megabyte
sectors.
A
But
then
we
wind
up
not
test
not
necessarily
testing
32
gigabyte
sector
stuff,
which
which
gets
a
little
tricky,
and
if
you
have
even
a
couple
miners
testing
out
the
32
gigs,
then
your
512
gigabyte
miners
are
kind
of
automatically
dwarfed
which
gets
a
little
irritating
one.
A
One
definite
learning,
though,
is
that
the
the
new
test
net
will
probably
have
the
the
github
oauth
faucet
again,
because
if
not
testament
faucets
get
drained
real
quick
by
our
community
and
then
it's
a
really
bad
scene
where
folks
have
to
be
begging
each
other
for
fake
monopoly
money
like
that's,
not
what
you
want,
so
I'm
rich
yeah
so
I'll
be
bringing
that
back
for
sure.
A
Cool
sounds
good
yeah,
any
any
follow-up
questions
bring
up
in
the
film
planters
channel,
but
yeah.
Congratulations
on
all
the
work.
It
will
be
really
cool
to
have
other
implementations
on
mainnet
really
soon,
hey,
bye,
everyone,
bye,
bye.