►
From YouTube: 2021-04-07 meeting
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
B
A
A
B
B
B
B
C
It
has
been
in
a
draft
state
for
almost
two
months
now,
so
I
was
wondering
if
youtuber
will
continue
on
it,
because
we
will
need
it
for
nginx
and
httpd
someday.
B
Okay,
yeah,
so
not
always
there.
I
think
he
can
answer
answer
it.
Well,
then
I
mean
we
just
we
were
talking
about
the
jagger
exporter.
I
think
you
joined
at
the
right
time,
so
I
think
what's
the
current
status
and
I
just
wanted
to
understand.
D
D
For
for
the
jaguar
exporter,
I
will
give
an
update
this
week.
Yeah
I'm
still
completing
some
some
test
part
with
the
official
jagger.
I
think
that's
a
collector
or
agent,
I'm
still
testing
that.
So
I
will
push
an
update
this
week
and
I
think,
should
be
in
formal
state
now.
It
is
still
in
draft
stage.
B
Yeah,
I
think
that's
I
I
did
have
some
of
the
some
of
the
stuff
in
agenda,
but
I
think
probably
it
was
better
if
josh
and
some
other
people
would
have
been
here
to
discuss
that.
That
was
regarding
the
prs
which
josh
is
raising
for
trace,
contest,
multiple
processor
and
resource
integration,
but
I
think
probably
mark
it
for
next
week.
In
case
we
don't
have
him
in
the
meeting
today,
so
yeah
anything
anything
tom
you
want
to
discuss,
I
think
probably
it
would
be
the
shortest
meeting
we
have
in
that
case.
A
Okay,
now
so
so
the
group
is
there
sorry,
I
just
somehow
I
I
think
I've
missed
the
invitation,
but
maybe
this
is
just
my
github
settings.
I
I
am
in
the
group.
However,
I
have
not
received
any.
I
don't
think
I
received
any
emails
related
to
this,
but
I'm
in
the
group.
So
I'm
I'm
I'm
yes.
Thank
you.
Yeah.
B
B
B
Yeah-
and
we
can
see
if
by
adding
into
the
group,
it's
that
sufficient
to
create
a
release,
because
that's
what
sim
was
currently
blocked,
he
could
not
create
a
release
as
of
no
probably
something
hopefully
once
he
is
part
of
that
group.
He
should
be
able
to
create
a
release.
B
So
but
probably
I
think
thomas,
you
can
also
quickly
check
if
you
get,
you
see
the
release
button
in
in
the
context
or
not,
probably
that
will
that
will
make
it
much
clearer
that
by
adding
into
the
group,
that's
the
sufficient
permission
to
create
a
new
lease
or
not
so
I
mean
you
if
you
go
to
the
content
repo,
do
you
see
a
release
button
to
create
a
new
release.
C
C
A
Okay,
no,
I
I
don't
see.
Is
it
okay?
If
I
share
a
screen
for
a
second,
please
yeah
until
you
share
the
agenda
so.
B
B
B
B
So
probably
need
to
figure
out,
because
I
I
know
that
that
max
is
the
admin
for
this
repo.
In
case
he
has
given
specific
permissions
for
right
access
to
this
repo,
then
I'm
seeing
this,
but
I
mean
earlier,
I
was
feeling
probably
it's,
because
I'm
part
of
this
high
contract
proverbs
group.
That's
the
reason
why
I'm
seeing
this,
but
I
think
that's
not
the
case.
In
that
case,
let
me
figure
it
out
how
to
do
it,
because
max
max
is
only
as
I
understand
he
won't
be
there
this
week.
D
Well,
I
think
I
asked
you
about
getting
getting
resource
from
frostburn.
I
think
this
is
required
for
the
shuttle
exporter
to
populate
the
process.
B
B
Pull
request
was
there
for
some
time,
so
he
he
did
create
a
request.
So
so
it's
a
long
history.
I
created
a
police
request
where
I
was
adding
that
resource
as
part
of
span
to
through
span
every
span.
I
mean
the
resource
would
be
propagated
to
the
exporters,
but
josh
has
a
concern
that,
because
this
resource
is
going
to
be
safe
for
every
span,
we
should
not
be
propagating
it
as
part
of
the
span,
probably
through
through
processor.
B
B
E
If
somebody
else
wants
to
continue
so
so
here's
the
thing
with
resource
and
span,
I
think
we
should
solve
the
instrumentation
library
and
span
problem
and
resource
will
follow
along
because,
if
it
we
need
to
solve
both
for
otlp
and
one,
the
resource
problem
is
a
little
easier
than
instrumentation
library,
okay,
okay
and,
I
think,
there's
two
ways
we
can
go
about
doing
this
and
I
threw
together
a
prototype
which
is
really
junky
in
one
of
those
cls,
and
I
don't
know
if
it's
the
way
we
should
go
forward,
because
I
think
the
options
are
a
little
ambiguous
here.
E
There
were
concerns
on
at
both
options,
but
the
two
ways
I'd
propose
are
one
is
we
keep
a
reference
to
the
tracer
shared
pointer
inside
of
a
span
and
instead
of
using
this
recordable
interface,
we
pass
this.
This
exportable
span
thing
I
tried
to
create
that,
has
the
recordable
and
has
the
original
tracer
and
then
exporters
which
need
access
to
instrumentation
library
and
resource,
get
it
off
of
the
tracer.
E
The
second
option
is,
we
literally
write
instrumentation
library
end
resource
into
recordable,
which
is
what
you
would
propose
lily
in
your
original
cl.
That
has
a
lot
of
duplication
of
of
information
because
you're
writing
into
every
single
span,
and
I
think
those
are
the
two
options
on
the
table.
Yeah
the
I
tried
to
detangle
span
and
recordable,
and
it's
it's
it's
a
very
aggressive
pr
that
does
this.
E
I
have
it
and
I
sent
like
that
first
cut
of
it
right,
but
it's
ugly
as
hell
and
it
needs
to
be
cleaned
up,
and
so
I've
been
trying
to
do
this
like
send
things
in
fragmented.
If
somebody
else
wants
to
take
a
crack
at
it,
I'm
more
than
happy
to
like
step
back
and
let
somebody
else
do
that.
That's
fine,
because
I
I
know
my
bandwidth
has
been
limited.
E
I've
been
my
p0
has
been
the
metrics
data
model,
getting
that
marked
as
stable
and
that's
been
taking
way
more
time
than
I
expected.
So
I
apologize
for
availability
but
that
I
think
those
are
our
two
options,
and
so
you
know
question
number
one
is
which
which
direction
do
we
want
to
go
there
and
because
I
don't
know
if
my
counter
proposal
to
your
original
pr
is
actually
the
best
either.
I
think
I
think,
there's
trade-offs.
We
just
have
to
decide
on
and
then
the
second
question
is
who
does
the
work?
B
E
B
Okay,
so
so,
okay,
so
probably
I
think
I'm
just
adding
that
the
current
set
of
tasks
which
people
guys
are
doing
so
trace
context.
I
mean
I
just
want
to
have
more
clarity
on
the
water
leadership
we
are
having
and
who
is
doing
what
so
trace
context
right
now,
george,
you
already
raised
the
pr.
E
There's
a
windows
build
issue,
that's
and
then
a
clang
format
issue
and
then
I
think
that's
done.
B
Yeah,
so
probably
I
think
we
can
purposefully.
I
just
wanted
to
wait
for
a
couple
of
days
to
see
if
johann
can
review
it,
I
think,
and
from
once
these
issues
are,
I
think,
fix.
I
think
we
should
be
good
to
merge
it
and
then
the
multiprocessor
support
would
you
like
to
I
mean
following
this:
would
you
like
to
have
appear
for
this,
because
I
think
you.
E
E
Library,
I
don't
think
is
called
out
here
at
all,
but
that
one
that
one,
I
think,
also
needs
to
be
handled
resource
and
instrumentation
library,
I
think,
should
be
the
same
person.
E
Multi-Processor
can
be
someone
else
what
I
wanted.
The
only
reason
I
looked
into
it
was
to
make
sure
that
when
I
did
the
resource
thing
I
kind
of
made
it
so
you
can
never
have
multiprocessor.
So
that's
the
only
reason
I
looked
at
it.
If
somebody
else
wants
to
take
that
known
it,
I'm
fine.
If
somebody
else
wants
to
take
resource,
instrumentation
library,
I'm
fine
with
that
too,
like
I
don't,
I
don't
need
to
own
all
of
this.
E
I
just
wanted
to
make
sure
that
I'm
not
designing
something
that
break
like
prevents
us
from
making
progress
later.
That
was
all
okay.
B
I
I
gotta
have
a
look
into
multi-processor,
that's
something
I
think
you
already
raised
a
pr
for
that
and
the
changes
are
already
there
in
your
previous
vr.
Probably
I
mean
correct
me
if
my
understanding
is
wrong
and
we
just
need
to
take
it
from
there
and
probably
create
a
new
pr
on
top
of
that,
and
do
you
see
any
further
issues
I
mean
in
that
I
mean,
apart
from
just
taking.
E
I
I
think
that
should
work,
there's
there's
the
issues
that
johann
brought
up,
that
I
think
we'll
still
have
to
resolve.
It's
like,
I
said,
there's
a
design
trade-off
here
of.
Basically,
who
pays
the
memory
cook
price.
B
B
B
E
Yes,
I
will,
I
will
summarize
all
the
discussions
and
things
we
had
into
an
issue
and
sorry
for
the
scatterbrained
plaster.
B
Did
see
your
comment?
Basically
you
don't
really.
I
mean
your
concern
was
that
the
carrier
should
not
be
tightly
coupled
with
the
propagator.
E
Right
and
so
it's
no
longer
a
global
propagator
it'd
be
like
I'm
going
to
install
the
propagator
for
curl
or
I'm
going
to
install
the
propagator
for
our
http
library
right.
It's
not
that's.
My
main
concern
is,
is
the
usability
of
it
with
the
template
like,
but
obviously
the
suggestion
I
have
is
not
as
performant
yeah.
B
Definitely
definitely
this:
this
is
more
of
a
compile
time
generation
so
that
definitely
yeah.
Let.
B
Probably
I'll
get
some
more
idea
from
there
because
I
think
if
I
know
dotnet
also
using
a
version
of
propagators,
so
probably
they
will
also
have
similar
issues.
Let
me
see
how
they
do
it
and
probably,
if
you
want
to
fix
it,
we'll
do
a
fix
for
this.
Otherwise
we
may
have
to
remove
global
propagated
together,
because
that
does
not
really
solve
the
purpose
of
having
a
global
propagator
yeah
and
I
think
civility
conversion
would
be
clear
nobody's
really
working
on
that.
B
I
did
have
a
look
into
whether
we
can
really
generate
a
code
from
the
configuration.
I
don't
see
any
tool
which
can
generate
a
c
plus
plus
code
from
that
java
configuration
which
we
have
for
semantics.
B
So
probably
we
may
have
to
write
it
manually
if
there
is
nothing
like
that
available
so,
but
I
think
I'm
not
touching
it
as
of
now
baggage.
I
know
somebody's
already
taken
nickel
has
already
taken
it
over.
You
have
done
some
refactor
so
probably
take
care
still.
I
think
we
don't
have
anybody.
B
B
So
this
is
more
for
the
tracking
I
mean.
The
only
only
take
here
is
that
we
definitely
need
at
least
initial
trace
context,
multi-processor
resource
integration
for
a
beta
release.
I
mean
with
that
that
and
probably
for
release
candidates.
B
You
should
have
all
these
so
right
now
beta
release,
I
think
you're
targeting
for
end
of
april
and
the
release
candidate.
Would
this
and
if
you
see
the
the
the
milestones
you'll
see,
the
beta
release
is
targeted
for
end
of
april
and
the
release
candidate
for
end
of
may
that's
more
stringent
timelines,
I'm
sure
at
least
the
release
candidate
will
stretch
out
a
bit
but
yeah
so
probably
at
least
for
release
for
beta
release
and
at
least
resource
integration.
Multi-Processor.
B
To
drop
yeah
thanks,
everyone,
and
probably
I
think
we
get
into
45
minutes
we
get
20
minutes
back
now.