►
From YouTube: Instant Finality Implementation, Peer Discovery Code, Deferred Transactions & More! August 9th
Description
The EOS Node Operator Roundtable meets every Wednesday at 14:00 UTC to share and discuss feedback to improve the Antelope protocol.
00:00:00 Intro
00:02:14 Instant Finality Implementation
00:03:30 Peer Discovery Code & Read-only Transactions
00:7:07 Replay Test code
00:12:36 Planning and Timeline
00:19:05 Kylin and Jungle Testnet
00:33:53 Communication and Milestones
00:41:46 Deferred Transactions
The EOS Network Foundation coordinates financial and non-financial support to encourage the growth and development of the EOS Network. We’re harnessing the power of decentralization to chart a coordinated future for the EOS Network as a force for positive global change.
// EOS Network Foundation //
https://www.eosnetwork.com
https://twitter.com/EosNFoundation
https://www.linkedin.com/company/eos-network-foundation
https://t.me/EOSNetworkFoundation
https://discord.gg/eos-network
info@eosnetwork.com
#EOS #Web3
B
B
A
B
Various
steps
of
of
the
protocol
upgrade
sort
of
coordination
process
that
we
need
nude
operator
and
BP
participation
on
it
is
a
draft.
Things
may
need
to
pushed
out
and
also
there's
the
possibility
that
we're
missing
important
steps
in
here
from
like
a
community
perspective,
and
so
that's
that's,
where
I'll
take
yeah
feedback
from
folks
and
then
again
in
the
in
the
future,
I
I
expect
the
community
will
actually
be
leading
this
discussion.
B
But
for
now
let
me
let
me
just
walk
through
what
we
have
here:
I'm
going
to
come
back
to
these
first
two,
which
are
about
deferred
transactions,
because
there's
a
bit
more
of
a
maybe
lengthy
discussion
to
be
had
around
this,
so
I'm
going
to
skip
those
for
now
and
go
straight
into
the
remainder
of
the
plan.
So
these
first
four,
this
group
of
the
first
four,
are
really
the
the
major
functionalities
that
are
intended
to
be
in
5.0,
which
are
being
supplied
by
multiple
parties.
B
B
However,
there
was
discussion
about
possibly
bringing
that
date
in
and
I
I
have
a
I,
have
an
action
item
to
to
sync
with
the
origin
team
and
and
get
an
up-to-date
thing.
So
this
is
the
biggest
item
that
might
push
out
this
plan
that
you're,
seeing
here
we've
made
some
assumptions
about
how
it
can
be
pulled
in
in
the
future.
It'd
be
good
to
get
the
origin
team's
direct
involvement
here
or
we'll
continue
to
proxy
for
them
if
they're
unable
to
attend
so.
B
Really
this
and
replay
testing
are
the
two
we
we
would
alter
the
plan
for
these
two
in
the
middle
here.
If,
if
they
run
late
and
these
other
items
aren't
running
late,
we
would
likely
explore
cutting
them.
However,
those
items
are
peer,
Discovery
improvements
which
are
currently
owned
by
futan
I,
say
currently,
they
are
owned
by
food
hand
and
read-only
transaction
improvements
for
increased
stabilization.
B
Those
really
transactions
and
really
the
the
headline
here-
is
just
the
the
furthest
out
date,
which
for
replay
testing
is
9,
15
and
again,
possibly
as
late
as
929
for
instant
finality,
but
we
are
making
the
assumption
that
it
will
pull
in
to
the
15th
for
this
current
plan
and
I
will
follow
up
with
updates
in
a
future
call
if
we
find
that
that
isn't
possible
so
before
I
move
beyond
that,
any
any
questions
or
thoughts
or
concerns
or
comments
on
these.
This
first
group
of
items.
C
B
B
There
are
somewhat
old,
PRS
and
I'm,
making
the
assumption
that
those
PRS
do
not
Encompass
everything.
That's
in
this
patches,
which
may
or
may
not
be
true
so
assuming
that
there's
additional
changes
from
the
PRS
which
are
about
four
months
old,
but
we
need
a
updated
PR
from
that
team
and
we're
coordinating
that
now.
But
as
soon
as
we
have
a
PR
cabin
will
do
review
of
of
that
item
and
and
that's
basically,
the
the
coordination
will
happen
through
code
review
on
the
GitHub
PR.
D
Perfect,
it's
currently
a
separate
branch
that
only
works
against
borrow,
and
so
what
we
want
to
do
is
make
it
work
and
all
those
branches,
so
they
have
some,
they
I
talked
to
them.
They
they
know
they
have
some
changes.
Some
updates
that
we
Converge
All
That
code
together
and
then
we'll
get
it
out.
The
right
release.
B
B
All
right
so
moving
on
so
so
this
first
group
was
the
functionality
itself
being
you
know
done
this
next
group
is
about
you
know,
testing
and
actually
cutting
a
release
candidate,
and
so
this
is
taking
place
over
a
very
short
period.
But
you
know
we
will
have
completed
the
replay
from
Genesis
testing.
You
know
code
and
now
we'll
actually
be
running
those
that
replay
test.
The
purpose
of
this
is
just
to
ensure
that
this
integrated
functionality
has
not
broken
our
ability
to
replay
from
Genesis
right
and.
D
C
And
the
reason
I
mentioned,
that
is
I'm
pretty
sure
in
foro
now
at
least
latest
three
or
four.
We
found
a
few
quirks
that,
because
nobody
replays
EOS
from
Genesis,
except
for
dumbasses
USA,
to
regenerate
State
history
files,
it
also
had
to
do
with
syncing
and
it
being
a
state
history.
So
some
of
these
kind
of
you
know
specific
cases,
but
it
found
some
regression
problems
that
couldn't
make
it
pass.
So
we
fixed
those.
C
C
So
I
was
curious
about
the
rethinking,
especially
when
we're
doing
peer
changes,
and
all
of
that
I
can
tell
you
it's
taken
about
a
month
with
a
local
Pier
running
at
as
fast
as
it
can
validate
on
an
I9
it's
taken
a
month
to
get
to
roughly
the
end
of
2020
or
the
Q3
of
2020..
So
I
mean
it's
going
to
take
another
month
or
two
to
actually
do
it.
C
D
C
Wasn't
out
it
was
the
actual
node
OS
shut
down
and
corrupted
the
logs
and
everything,
including
sync,
the
chain.
It
just
happened
to
be
in
whatever
process
the
state
history
was
processing
it.
That's
where
it
was
blowing
up
something
about
traces
or
something,
but
the
main
thing
was
when
we're
only
doing
a
snapshot
and
a
quick
replay
of
whatever
that
it
is
right
now.
C
D
Yeah
we
definitely
we
definitely
wanted
in
the
future,
use
the
replay
test
as
a
way
to
exercise
a
lot
more.
The
features
and
functionality,
because
there's
a
variety
of
these
transactions
and
that
I
just
want
to
clarify
that
we're
we're
really
focused
on
this
blocks,
log
replay
and
making
sure
that
we
can
do
that
in
this
line
item
here,
and
so
it
may
not
get.
D
B
E
B
Doing
this
to
be
sure
that
we
have
preserved
ability
to
replay
from
Genesis
yeah
yeah
cool
all
right.
So
then
you
know
that
yeah
I
should
have
probably
had
a
calendar
upset
in
front
of
me,
so
that
I
can
be
referencing
the
days
of
the
week.
I
wonder
if
I
should
just
format
these
dates
differently.
Whoops.
E
F
E
B
B
So
that's
something
that
we're
that
will
provide
us
some
some
test
cases
that
you
know
will
ask
for
help
with
running
those,
but
yeah
so
we'll
make
sure,
because
at
this
stage
instant
finality
won't
actually
be
enabled
on
Jungle
it'll
just
be
the
5.0
code
that
includes
instant
finality,
but
off
right,
so
we'll
test
those
features.
This
is
basically
to
establish
that
5.0
without
the
feature
activated
is
like
working.
B
You
know
as
an
anticipated
on
October
2nd
the
contracts
that
there
are
some
some
but
core
system
contract
updates
related
to
instant
finality,
so
those
would
be
released
and
we
would
of
course,
kind
of
give
the
thumbs
up
like
yeah.
This
lead
five
5.0
rc1
is
good
on
Jungle
we'd,
activate
the
protocol
feature,
and
then
there
would
be
basically
a
you
know.
A
period
of
almost
a
week
for
system
contract
upgrade
and
to
occur
and
finalize
our
keys
to
be
registered.
B
Finalizer
keys
are
just
you
know,
something
that
instant
finality
needs
notes
to
have
so
and
then
we
would
get
instant
finality
enabled
on
Jungle
and
that's
basically
the
setup
for
testing
of
instant
finality
now
you'll
notice
this
this
is,
you
know
a
longer
period
here
right,
so
we'll
have
some
test
cases,
but
also
the
just.
The
kind
of
ongoing
operation
of
the
network
in
itself
is
a
test
right
like
we
get.
We
get
to
see
if
anything's
blowing
up,
because
instant
finality
is,
is
touches
everything
kind
of
foundationally
right.
B
B
And
then,
basically
from
then
until
November
29th
is
the
period
for
EOS
block
producer
nodes,
in
particular
to
you
know
to
to
get
upgraded.
So
this
is
when
coordination
of
you
know
with
across
all
all
the
other,
all
the
top
BPS
and
and
really
all
all
standby,
DPS
and
notes
that
are
you
know
critical
needs
to
happen,
including
exchanges
or
you
know,
across
both
sort
of
the
East
and
West
right.
B
This
we
need
to
pick
a
date
for
this,
but
you
know
basically,
there
will
be
a
sometime
close
to
this
date,
where
the
deployment
has
happened.
It'll
be
a
go
no-go
call
right
for
for
Community
BPS
to
you
know,
make
the
decision
and
coordinate
on
exactly
when
the
protocol
activation
will
occur
right
now
in
the
plan,
we've
tentatively
just
put
a
date
in
of
the
29th.
So
you
know
at
the
end
of
this
period,
which
is
a
Wednesday
and
then
again
there's
about
a
about
a
week
period.
B
I
think
if
I'm
doing
the
math
rate,
it
might
be
a
week
and
a
half
or
something
but
or
no
I
guess
it
would
have
to
be
one
or
two
weeks
anyway
for
you'll
notice.
This
kind
of
mirrors
that
what
happens
on
Jungle
right
so
upgrade
the
system,
contract
register,
finalizer
keys
and
then
finally
enable
instant
finality
on
EOS
on
the
6th
of
December.
B
E
You
know
you
know
what
a
good
test
is
where
there
is
a
global
distribution
of
top
20
white,
which
is
surely
that's
going
to
be
one
of
the
main
things
in
regards
to
this
instant
finality
when
you've
got
like
latency
between
everybody
and
I
know
on
kylan,
a
lot
of
us
have
got.
You
know
four
BPS
running
it
kind
of
defeats.
What
we're
trying
to
do
so,
where
we
can
get
the
best
distribution
is
probably
the
best
right.
I.
C
Would
assume
jungle
has
probably
the
best
distribution
minus
all
the
animals
that
the
crypto
Lions
runs,
as
you
know,
drunken
bad
or
whatever,
but
I
think
the
idea
is
yes,
Island
isn't
nearly
as
distributed
than
a
I
think
it's
got
Forks
everywhere,
every
whatever,
but
what
it
is
designed
to
do
is
not
so
much
test.
Yeah
like
the
latency
and
and
the
infrastructure,
and
all
of
that,
but
more
provide
the
chain
environment
for
the
dads
we've
already
tested.
The
LA
is
is
what
it
was.
C
I
mean
last
time
we
did
it
and
I
think
it's
a
general
design
doesn't
mean
it
has
to
be
I,
don't
know
how
many
people
really
still
use
it,
but
it
was.
We
definitely
fixed
it
for
the
3-0
upgrade
to
be
able
to
do
this
same
thing,
and
it's
easier
to
you
know.
I
think
Nations
and
I
have
a
m-sig
control
that
we
can
just
kind
of
push
things
to
get
them
done
because
we've
already
tested
them
in
Jungle
in
theory,
so
a
valid
point,
but
not
so
much
on
the
distribution
and
the
emperor
side.
C
G
A
G
C
C
But
like
I
say
this
is
also
when
you
would
say
the
exchanges
now
probably
have
their
exposure
to
a
5-0
platform
that
they
would
start
testing
on.
They
could
use
jungles,
some
may,
but
I
think
that's
what
you
know
is
kind
of
a
really
designed.
So
that's
where
they're
going
to
start
being
able
to
whatever.
So
if
you
only
do
it
for
a
week
in
theory,
you're
you're
kind
of
saying
well,
we've
tested
all
of
our
stuff
for
a
month.
We
feel
good
about
it.
A
Exchange
often
it's
like
you
know,
they
got
a
million
different
chains.
They're
supporting
is
eos
a
priority
that
they
can,
that
they
can
prioritize
in
that
one
week
window.
That's
that's
going
to
be
a
challenge
so
yeah
we
found
in
the
past
like
giving
a
good
two
months
from
final
release
and
correct
me
for
wrong.
Matthew,
like
we've
seen
a
lot
of
the
exchanges,
don't
touch
anything
until
there
is
a
final
release:
they're
not
interested
in
testing
your
pre-releases
they're
like
give
me
the
final
and
then
we'll
start
testing.
A
C
I
agree,
I
will
say,
though
you
could
probably
sacrifice
I,
don't
know
a
week,
maybe
on
Jungle,
because
I
mean
we're
we're
pretty
responsive
I
mean
you
know
that
that's
more
you're
organized
testers
and
people
that
are
involved
on
this.
Do
you
know
what
I
mean
that
maybe
you
could
trim
a
week
off
of
that
and
you.
C
27,
that's
three
weeks
yeah.
So
that's
a
three
week
Gap
in
between
I
think
so
yeah
anyway,
point
is
that
you
could
probably
lean
a
little
harder
on
Jungle
BPS
to
and
line
them
up
and
knock
them
down
and
I.
Don't
know
what
you
know:
y'all's
testing
would
need
to
be
on
that,
but
I
mean
obviously
we're
testing,
Ross's,
latency
and
and
the
distribution
and
all
of
that
very
by
the
point
boys.
Oh.
E
Go
ahead:
boss,
I
just
had
a
check,
there's
14
of
us
like
real
BPS
on
Jungle,
so
we've
already
probably
need
another
seven
to
ten.
We
might
be
able
to
do
that.
Just
by
reaching
out
to
other
chains,
I
mean
it's
all
Antelope
software
right,
so
we
could
probably
pull
some
some
other
guys
and
to
say
come
join
the
join
the
test
with
the
new
five.
You
might
be
interested.
E
E
H
H
G
A
C
B
B
H
B
H
B
That,
maybe
that
doesn't
work
anyway.
I
don't
have
to
do
this
entirely
live,
but
does
the
the
general
premise
work
here
like
if
we're
doing
Kyla
and
we
would
delay
we
would
the
adoption
can
kind
of
happening
concurrently
with
jungle
over
a
two-week
period?
We
would
do
the
activation
on
Jungle
first
and
then
follow
by
activating
on
Highland.
D
A
Well,
there's
going
to
be
I,
guess
some
communication
milestones.
We
might
want
to
track
here,
but
a
blog
post
or
two,
and
we
usually
get
the
the
EO
support
team
to
help
distribute
that
I'm,
not
sure
if
I
hope
the
enfs
distribution
mechanism
is
like
has
it
evolved
since
then,
maybe
there's,
actually
that's
something
you
guys
can
can
be
getting
on
the
horn
too.
A
A
A
A
C
C
I'll
also
say
you
know,
and
we've
talked
about
it
before-
is
that
Nations
at
least
had
the
article
last
time
that
gave
the
heads
up
about
deprecated
parameters
when
you
launch
it.
If
you
just
use
the
same
config,
it
won't
even
launch
you
gotta
turn
some
stuff
off
and
then
even
recommended
parameters,
for
you
know
like
the
read
window
and
Max
transaction
time.
Otherwise
it
goes
walking,
but
that
may
not
be
an
ens.
It
may
not
be
an
official
whatever,
but
I
know.
We've
talked
about
this
in
general.
C
Some
way
in
that
announcement
to
kind
of
give
that
that
Geek
Squad
a
heads
up
on
if
you're
a
BP.
You
need
to
do
this.
If
you're,
you
know
like
the
subjective
billing
methodology,
even
changed.
That
may
not
be
a
required
setting,
but
understanding
it
I
had
to
meet
with
Kevin
who
might
I
say
deserves
a
raise.
A
C
Say
it's
probably
safe
to
assume
that
they're
going
to
be
upgrading
from
an
older
three
dot,
X
version,
it
would
have
been
the
last
time
they
had
to.
You
know
there
were
all
sorts
of
features
and
quirks
and
whatever
so
we
that
have
gone
through
the
4-0
upgrade
forget.
Oh
that's
right!
It
doesn't
support
1804
anymore,
the
Boost
versions,
change
I!
Don't
have
to
go
into
all
the
details,
but
because
of
that,
typically,
yes,
you
need
a
snapshot.
B
Well,
this,
like
I,
said
this
is
the
sort
of
the
first
draft
plan.
Format
and
whatnot
might
change,
but
you
know
we're
we're
just
using
this
to
collect
feedback
on
the
initial
plan.
This
will
evolve
as
we,
you
know,
better
understand
whether
things
are
going
to
come
in
on
the
dates
we
expect
or
not,
and
we'll
will
remain
participants.
B
E
Hey
Brian,
so
I
know
that
this
is
obviously
node
operator
called
Tech
tick,
whatever
it'll
also
be
I,
think
beneficial.
If
we
spoke
a
little
bit
about,
for
instance,
the
new
features
and
the
new
look
on
consensus,
and
even
though
this
isn't
the
the
brain's
Trust
of
the
guys
that
are
actually
building
the
code
and
all
the
kind
of
Economics
behind
other
that
you
may
net
would
could
be
in
a
few
years
time.
E
It
would
be
great
to
bring
us
up
to
speed
with
some
of
the
thinking
you
know
unless
everybody
else
in
this
pool
doesn't
agree
with
me
I'd.
You
know,
I
I
read
about
all
the
the
cool
stuff
and
articles
that
are
being
released,
and
sometimes
it's
not
to
get
some
info
and
let
it
you
know,
click
through
my
head,
you
know
foggy
with
Marco,
can't
see
let
somebody
else
speak.
You
know
that
kind
of
thing.
B
It's
teams
right
because
we've
got
got
multiple
parties
involved,
so
it
might
be
one
of
those
things
where,
especially
as
the
status
update
part
starts
to
get
briefer
right,
like
we're
kind
of
going
over
the
whole
plan
right
now,
but
at
some
point,
it'll
be
like
yep,
we're
so
green
on
that
you
know
and
that'll
be
the
call.
So
we
would.
We
should
have
space
for
that
sort
of.
B
E
B
Sorry,
let
me
back
pedal
and
show
you
don't
think
time,
because
I
forgot
I
skipped
something
at
the
beginning,
my
past
so
deferred
transactions.
I
wanted
to
talk
about
that
feature.
So
deferred
transactions
are
a
it's
a
feature
that
has
previously
been
deprecated
in
an
Ideal
World.
We
would
actually
get
them
removed
for
the
5.0
release.
B
F
B
And
especially
as
a
as
part
of
a
protocol
upgrade
because
they
would
require
a
protocol
feature
Activation
so
and
we
we
try
to
limit
that
to
once
a
year.
So
if
we
don't
get
it
in
5.0,
we're
gonna
have
to
wait
until
next
year.
Basically
to
get
it.
So
that's
why
it
would
be
ideal
to
get
the
removal,
but
there's
a
problem.
B
There's
kind
of
two
categories
of
things
that
depend
potentially
depend
on
deferred
transactions,
there's
stuff,
that's
sort
of
under
code
base
control
like
core
code
based
control
right,
so
there's
some
system
contract
stuff
that
currently
depends
on
deferred
transactions
that
you
know
we
can
just.
We
can
just
make
changes
there
and
you
know
they'd
be
part
of
the
contracts
that
would
get
adopted
as
part
of
this
overall
flow.
B
However,
there's
a
larger
sort
of
unknown,
which
is
smart
contracts
in
the
wild
that
made
it
depend
on
this
feature
right
now.
What
we've
discussed
is
like
ways
that
we
can
de-risk
the
removal
of
this.
You
know,
because
what
we
don't
want
to
do
is
remove
it
and
then
have
that
be
the
reason
why
this
whole
schedule
is
completely
blown
out
of
the
water
right,
because
we
weren't
well
prepared
for
some
Fallout
right.
B
There's
two
things
that
we've
kind
of
discussed.
One
is
that
if
we
sort
of
we
do
the
removal
changes
we
can.
Those
can
be
in
a
you
know,
protocol
feature
so
that
we
can
sort
of
independently
decide
whether
or
not
to
activate
removing
that
right,
and
so
that
way,
if
we
do
find
that
it's
a
problem,
we
could
you
know
we
could
sort
of
backpedal
and
say
yeah.
B
So
another
thing
there's
a
precedent
that
that
you
know
we've
seen
sometimes
in
these
sort
of
distributed
decentralized,
feature
dependency,
World,
a
call
back,
I
forget
what
year
this
was,
but
there
was.
There
was
a
thing
called
I
used
to
work
at
a
at
a
large
Cloud
hosting
company,
and
there
was
a
there
was
an
event
called
world
IPv6
day
right
anyway,
and
preparing
to
see
like
whether
weather
IPv6
support
had
become
ubiquitous
right.
B
And
then
you
know
it's
planned
to
be
temporary
right.
Just
to
you
know
we're
going
to
do
it
for
a
certain
period
of
time
and
when
it
comes
back
on
you
know,
we'll
we'll
sort
of
share
gossip
on
like
okay
did
did
the
Sky
Fall
did
people
where
there
are
lots
of
you
know,
broken
apps.
You
know
that
sort
of
thing
we're
considering
doing
something
like
that
for
deferred
transactions,
so
that
we
can
so
that
we
can
get
feedback
and
and
the
benefit
of
doing
that.
B
So
we
actually
don't
really
need
to
wait
for
any
code
changes
with
this.
We
would
just
be
able
to
coordinate.
You
know
disabling
the
Deferred
transaction
feature
for
a
temporary
period
of
time.
Now,
of
course,
the
it
would
require
a
multi-cig
right,
so
so
there's
some
Community
coordination.
That
would
be
required
to
do
that.
B
G
F
B
B
So
the
thought
there
is
that
we
can
get
knowledge
about
whether
this
is
sort
of
a
non-event
or
something
that
should
just
be
completely
avoided.
I
mean
in,
for
example,
we
may
learn
that
actually
removing
deferred
transactions
at
all
is
just
really
a
a
problem
right
or
that
we
need
a
very
robust
sort
of.
B
B
B
B
B
C
Out
other
than
my
mind,
the
one
that
pops
in
which
isn't
probably
a
concern
really
at
the
the
hold
up
level,
would
be
there's
still
a
lot
of
chains
that
are
just
still
kind
of
running
on
the
old
way.
You
know,
there's
just
some
contracts
and
refunds
and
stuff
piggyback
on
that
deferred.
They
just
haven't
had
any
issues
with
it
yet
and
really
haven't
upgraded,
but.
C
Well,
I
mean
it's
been
proven
that
you
know
you
can
obviously
deprecate
them
and
upgrade
to
the
current
system
contracts
that
in
the
last
like
three
years
so
I
mean
again.
That's
what
I
say.
You
know
you
need
to
push
them
off
a
cliff
eventually,
like
that's
really,
because
most
of
the
major
active
chains
have
already
once
you
get
bogged
down.
Deferred
transactions,
don't
really
work
anyway.
Right
I
mean
refunds,
don't
really
kick
out,
you
got
to
go
manually
trigger
them
just
because
the
Deferred
transaction
didn't
Trigger
or
whatever
you
know
or
things.
C
H
B
C
H
B
Okay,
well
then,
that's
that's
something
that
you
know.
We
can
discuss
further
next
time,
but
yeah,
we'll
we'll
coordinate
on
a
plan
for
requesting
help,
doing
sort
of
a
smoke
test
for
Fallout
of
disabling
that
because
of
course,
if
we
can
proactively
disable
it
and
there's
no
Fallout
really,
then
we
can
like
confidently
just
remove
the
feature
and
not
have
to
do
as
many
steps.
So.
D
C
H
B
I'm,
remembering
correctly,
that's
exactly
how
it
worked
for
the
the
IP
testing
I
was
talking
about
is,
like
you
know,
the
the
one
day
it
was
like
Hey,
everybody
turn
it
off
for
like
an
hour
and
then
it
was
increased
and
then
finally,
there
was
a
whole
day
like
a
24-hour
period
where
it
was
turned
off,
and
it's
like.
Okay,
if
the
internet
didn't
break
yeah.