►
From YouTube: 2022-08-03 meeting
Description
cncf-opentelemetry@cncf.io's Personal Meeting Room
A
B
So
we
have
the
pr
from
david.
I
think
this
one
was
already
revealed
by
some
folks
here
right.
B
New
issues,
okay,
yesterday
I
tried
to
run
the
workflow
on
my
box
and
the
stack
exchange
regis
failed.
I
always
get
four
spends.
I
never
get
eight.
I
didn't
have
time
to
kind
of
debug
and
see
what
may
be
going
there,
but
at
least
in
my
box
is
failing
consistent.
I
don't
know.
Perhaps
you
guys
give
a
try.
Just.Net
nuke
workflow,
perhaps
is
something
on
my
setup.
B
I
do
have
a
kind
of
underpower
machine,
not
lots
of
memory
and
stuff
like
that,
but
I'm
consistently
get
just
for
expense.
B
Yeah,
just
just
if
you
folks
have
a
chance
when,
when
you
wait
on
something
just
do
the
workflow,
if
it
fails
in
somebody
else's
machine,
then
I
think
I
will
try
to
dig
on
that.
But
perhaps
you
can
feel
my
on
the
power
machine.
I
don't
know
but
yeah
when
I
tried
it
failed.
B
B
A
Because
it
took
two
additional
retries
to
make
it
work.
B
I
was
gonna
say
I
had.
I
had
one
run
yesterday
that
was
stuck
on
hours
on
ubuntu,
but
then
on
the
rerun
is
passive.
You
know
I.
C
And
if
you
look
at
the
the
first
failure,
it's
even
more
interesting,
I
think
we
are
looking
at
the
second
one,
the
first
one,
all
of
it
failed
in
all
the
environment.
If
you
go
up
and
switch
the
first
run,
this
is
the
second
attempt.
The
first
attempt.
I
think
it.
Oh,
if
I
remember
correctly
all
this
is
when
robert
created
an
issue
like
I
see.
B
Yeah,
I
I
think
perhaps
we
we
could
keep
it
open
for
a
bit
more,
but
we
without
kind
of
kind
of
consistent
rapper
is
hard
to
do
anything
about
right.
So.
B
B
B
This,
I
think,
came
up
about
from
the
dpr
when
ad
instrumentations
right,
so
we
need
some
way
to
quickly
refresh
the
integrations,
json
or
eventually,
perhaps
move
out
of
it
right.
I
don't
know
which.
B
A
So
I
don't
know
that
it
needs
to
be
something
that
we
include
in
in
the
current
milestone,
but
might
be
a
nice
to
have
with
one
of
the
upcoming
ones.
B
Yeah
I
I
tend
to
think
that
we
don't
need
at
this
stage,
because
we
are
still
trying
to
get
more
instrumentations
and
kind
of
get
more
people
to
try,
and
this
is
purely
internal
right.
So
there
is
no
visibility
to
them,
so
I'm
I'm
tempted
to
keep
as
it
is.
You
know
not
add
to
to
the
milestone
or
anything
like
that
for
the
time
being,.
B
Let's,
let's
add
two
to
one,
but
then
we
we
we
have
to
do
in
some
time.
We
have
to
go
over
then,
and
there
we
decide
if
we
really
need
to
do
or
not.
So
for
now,
just
adding
to
this
milestone
we
re-evaluate
later
sounds
good.
B
B
I
had
some
failures
regarding
I'm
not
sure
if
was
this,
but
there
is
one
thing
that
perhaps
is
also
robert
machine,
because
it's
a
splunk
thing
that
from
time
to
time
they
kind
of
block
some
port
range
and
actually,
I
think,
actually
from
microsoft,
I'm
not
sure
microsoft.
I
mean
kind
of
windows
that
blocks
some
port
range
and
it's
kind
of
a
bit
of
random.
I
don't
know
if
you
guys
experience
that
too.
B
A
Yeah,
I've
definitely
seen
something
like
that
happen
before
so
I
know
in
other
integration
tests
I've
written
I've
had
to
test
the
availability
of
the
port
before
trying
to
actually
use
it
in
a
test.
B
B
I'm
tempted
to
kind
of
give
this
some
priority,
but
I'm
already
asking
you
guys
to
run
dotnet
nokia
workflow.
If
this
is
a
issue,
it
should
show
up,
with
the
caveat
that,
depending
on
on
the
failure,
we
don't
do
part
of
the
workflow.
B
So
we
may,
if
there
is,
there
are
other
errors
we
this
one
may
be
hidden.
You
know,
because
I
think,
for
instance,
if
I
have
multiple
failures,
let's
say
on
dotnet
core
3.1:
I
get
the
multiple
failures,
but
then
there
is
no
run
for
net
6.
For
instance,
you
know
so
it's
possible
that
if
we
had
somewhere
earlier
failure,
maybe
it
hides
this
one.
B
So
perhaps.
B
I'm
I'm
tempted
to
kind
of
add
a
milestone
for
this
one
just
for
us
to
be
sure
that
we
are
clean
on
our
long
run
of
the
tests
and
especially
for
this
stuff,
that
has
framework.
B
We
need
to
be
running
on
our
machines
because
it's
our
gap
in
coverage
in
ci
because
of
that
thing
of
it's
a
windows,
but
we
need
to
run
the
docker
in
linux.
So
it's
our
gap
in
coverage
right
now.
So
I'm
kind
of
I
see
here
asp.net,
I'm
not
sure.
If
it's
this
score
or
not,
but
if
it's
framework
we
should
be
running
this
locally.
It's
the
only
way
right
now,
so
I'm
kind
of
tempted
to
kind
of
put
it
here
for
at
least
for
now
for
the
beta.
B
So
we
are
sure
that
we
are
good
here
and
then,
if
it's
something
that
perhaps
require
fix
or
not
we
decide,
but
we
should
at
least
investigate
it.
B
So
this,
I
think,
will
be
one
of
the
issues
that
is
unblocked
right
now.
Robert
is
going
to
be
out
for
some
time
until
the
end
of
the
month.
His
vacation.
C
But
the
question
is:
if
you
would
like
to
switch
to
the
alpha
or
better
release
for
for
the
one
for
oh,
I
see
I
see
so
if
yes,
we
can
unblock
this
and
few
other
last,
probably
otherwise.
It
should
wait
as
long
as
we
decide
that
we
want
the
upgrades.
In
fact
we
have.
In
fact
we
have
tons
of
pull
requests
from
dependable
related
to
new
release,
but
still
same
question,
what
to
do
with
them.
C
Yeah
one
more
thing
to
that,
like
we
already
have
this
implementation
as
a
part
of
auto
instrumentation,
all
we
need
to
do
is
remove
that
from
the
auto
instrumentation.
So
it's
not
a
real
task.
I
think
we
should
wait
for
the
proper,
like
it's
not
a
real
blockage.
We
just
introduced
that
label,
which
is
confusing
us
over
here.
B
Yeah
and
and
to
be
fair,
I
actually
glanced
at
over
the
board
and
I
didn't
see
that
you
guys
already
discussed
targeting
ds2
2rc.
You
know
so
I
was
kind
of
assuming
that
was
target
for
for
beta,
so
which
is
not
the
case.
C
B
A
A
C
A
So
1.4
will
not
depend
on.net,
7
or
the
new
diagnostic.
C
A
B
Yeah
they
they
that
that
that's
what
I
heard
yesterday
yesterday,
I
went
to
the
sdkc
meeting,
because
I
I
had
a
pr
about
open
tracing
and
I
actually
asked
not
only
these
packages,
but
the
other
packages
should
be
published
still
with
the
current
version
before
they
publish
the
ones
with
dot
net
seven,
but
they
are
going
to
switch
the
their
main
branch
to
net
seven
relatively
soon.
B
Okay,
so
just
recapping
here
they
should
publish
a
package
with
dotnets
that
use
the.net
6
version.
You
guys
already
discussed
it.
This
should
be
for
the
release
candidate,
but
it
should
be
unblocked
relatively
soon.
I
think
just
then
it's
a
matter
of
how
this
balance
against
the
other
stuff
that
we
have
to
do
in
the
priorities.
A
Yeah,
so
about
that,
so
I
know
that
we've
tried
to
depend
on
the
the
stable
releases
of
of
the
sdk,
but
to
do
what
you're
talking
about
it
sounds
like
we
need
to
depend
on
an
alpha
release
of
the
sdk.
B
Yeah
that
that
that's
true
because
but
yeah
no
that's
that's
true.
I
I
was
gonna,
I'm
thinking
because
we
depend
on
rc
and
beta
for
packages
and
instrumentation,
but
not
for
the
sdk.
You
are
right.
B
Yeah
we
we
have,
we
have
a
a
reasonable
backlog
here
for
the
beta.
B
I
think
we
may
have
to
cut
some
of
this
stuff
here.
I
don't
know
if
you
guys
want
to
go
over
that
right
now,
but
perhaps
we
should
plan
for
that
because,
just
by
the
number
that
we
have
here,
I'm
not
sure
if
you
can
go
over
all
of
them.
C
C
B
Yeah,
I
I
I
think,
that's
true.
I
basically
on
our
side
as
robert
make
clear.
I
think
wcf
is
a
big
one
for
us,
so
yeah.
I
I
think,
and
my
assumption
is,
that
actually
reflects
for
all
vendors
in
this
space
so
yeah.
I
think
this
one
and
the
bug
really
should
be.
We
should
be
really
commit
for
this.
This
tool.
A
A
B
Yeah,
I
think
I
think
for
us
that's
the
case.
We
have
some
some
some
people
that
are
eager
to
try,
but
they
are
they.
They
use
the
wcf
heavily.
You
know.
A
B
Yeah,
so
I
think
for
the
time
being,
I'm
gonna
move
this
to
for
committed,
and
we
really
try
to
get
to
the
bottom
of
this
shoe
complete
wcf
and
investigate
the
bug,
but
probably
I
think
next
week
we
track
the
progress
and
also
review
this
stuff.
Here
we
need
to
go,
as
I
said
before,
over
stuff
like
this
one
here
you
know
especially
the
stuff
that
is
framework.
A
Yeah
I
mean
if
we
still
have
time
at
the
end
of
this
meeting,
I'm
fine
with
going
through
through
this
list
with
whoever
wants
to
stick
around.
B
A
Yeah-
and
I
think
there's
another
one-
that's
already
in
progress
in
that:
okay
yeah,
it
was
already
moved
the
one
that's
in
the
in
progress
column,.
B
B
All
right
I
think
this
week
completed.
Then
we
have
your
topic
quiz
about
the
version
of
seven.
A
Yeah
and
we've
kind
of
talked
about
that,
and
how,
for
this
milestone,
we
should
hold
off
on
any
of
those
updates
and
wait
for
the
next
milestone,
just
to
reduce
risk
with
the
current
release
and
prevent
us
from
taking
on
the
pre-release
dependency.
B
Yeah,
so
I
I
would
say,
as
piotr
red
said
and
by
the
meeting
that
I
have
yesterday,
so
we
should
be
safe
for
the
core
packages.
We
need
to
pay
attention
to
any
instrumentation
or
anything
just
kind
of
double
check.
If
there
is
no
dependence
with
the
newer
seven
version,
you
know
we
should
have
one
reduced
but
yeah
whenever.
B
Packages,
let's
remind
ourselves
to
take
a
quick
peek
and
see
the
dependencies
to
to
not
put
us
in
a
bad
situation.
B
All
right-
and
I
think
that
I'm
in
benefit
of
our
friends
in
eastern
europe
we've
been
talking
about
moving
the
time
of
the
meeting.
I
think
you
guys
had
some
agreement
last
week
when
I
was
not
around
here,
but
should
be
9
a.m.
Pacific
time
is
everybody
on
board
with
that.
B
Cool
cool,
so
I
predict
calendars
and
everything
so
by
next
already,
if
you
have
some
because
a
lot
of
people
usually
copy
the
the
things
from
the
calendar
and
the
updates
there,
don't
manifest
directly
to
yours
be
aware
that
the
next
one
is
9
a.m.
Pacific
time.
A
Yeah,
so
a
couple
of
things
to
check
when
that
happens,
do
you
know
if
we'll
run
into
any
meeting
conflicts.
B
I
still
don't
know
that,
but
we
have
to
figure
out.
You
know
we
wanted
the
the
time
of
the
meeting
to
work
for
us,
independent
of
that.
So.
B
All
right,
so,
if
you
guys
have
time,
perhaps
let's
go
over
the
the
backlog.
B
So
this
one
was
something
that
I
opened
quite
some
time
ago.
B
B
I
think
it
is
a
really
good
to
have
so
people
can
reference
quickly
if
they
see
some
strange
experimentation
that
should
be
kind
of
easier
to
get,
but
on
the
other
side,
what
I
was
thinking
afterwards
is
that,
because,
typically
this
matter
when
people
see
something
on
the
experimentation,
the
instrumentation
itself
by
open
telemetry
specs
should
be
including
the
version
of
its
implementation.
You
know
so
in
a
sense,
it's
very
good
to
people
read
before,
but
in
the
practical
case
that
information
should
be
readily
available.
B
B
I
think
we
had
a
bunch
of
things
that
had.
B
And
actually
there
was
some
other
discussion
about
the
these
names.
I
I
think
I'm
I'm
totally
fine
to
move
this
to
one
rc.
B
A
Or
environment
variable
names.
B
B
Yeah,
it's
a
good
point.
Typically,
that's
what's
gonna
happen.
Keep
you
on
the
board.
We
are
targeting
two
four.
B
Erasmus,
you
did
that
work
for
the
constant
did
you
because
you
know
the
constants
are
different.
I
was
gonna
say
if
somebody
had
a
second
look
at
that.
B
We
we
have,
I
is
because
I'm
not
sure
of
the
current
states
of
this.
That's
why
I'm
asking,
because
I
do
remember
that
we
had
some
difference,
but
I
think
a
few
of
them
were
fixed.
I
just
don't
know
the
current
state
of
this
if
we
had
a
bunch
of
difference
yet
between
the
traces
and
the
metrics,
these
were
configuration
settings.
B
B
A
B
A
Yeah
I
was
going
to
say
I
thought
I
remembered
seeing
us
having
at
least
one
test
in
the
past.
That
was
just
a
startup
hook
without
the
profiler
enabled.
But
I
I
don't
know
if
that's
changed
over
time,
because
we've
we've
refactored
a
lot
in
in
that
space
and
some
of
that
behavior
might
have
changed.
B
I
I
think
this
is
the
important
enough
scenario,
especially
for
for
for
something
that
microsoft
plans
to
to
use.
I
think
it's
worth
to
to
keep
on
on
three.
A
B
Yeah
this
is
a
nice
idea.
I
think
I
was
having
this
conversation
with
in
some
pr.
We
could
make
the
things
kind
of
go
faster,
but
let's
move
to
to
the
next
milestone.
B
Oh
no
is
is,
regarding
the
I'm
also
tempted
just
to
move
to.
I
I
just
a
quick
pause,
because
I
I'm
thinking
kind
of
the
more
instrumentations
that
we
add
is
important
to
keep
this
space
of
releasing
you
know.
So
we
eventually
have
to
get
to
to
things
like
this,
but
let's
kind
of
sprinkle
them
in
between
the
the
milestone
so.
C
A
Yeah
so
graphql,
I
suspect,
is
lower
usage
and-
and
so
I
I
think
we
can
include
that
with
the
next
round
of
instrumentation
updates.
I
think
the
semantic
conventions
finally
merged
for
for
graphql,
but.
B
Robert
did
a
bunch
of
fixes
there.
What
spending
yeah.
A
So
this
just
to
catch
you
up,
so
the
existing
page
has
already
been
updated
and
renamed
to
focus
on
library,
instrumentation,
libraries
and
what
we
want
to
do
is
add
a
page
to
the
documentation
that
focuses
on
our
project.
B
I
I'm
I'm
tempted
to
kind
of
put
to
the
next
milestone.
Show.
B
Unless
you
guys
have
some
kind
of,
as
I
said
it
is
not,
that
is
more
a
thing
like
kind
of
okay,
let's
clean
up
the
the
issues
and
select
just
a
few
for
the
next
release
and
we
keep
doing
those
as
we
go.
B
B
B
Yeah,
the
the
bottom
one
kind
of
we
already
decided
the
bottom
one
I
mean
the
other
one
bug
about
test,
I'm
just
looking
to
see
if
this
is
similar.
B
Yeah,
I'm
I'm
tempted
to
say
that
we
keep
for
three,
but
probably
probably
there
is
a
bit
of
assumption
here.
This
is
infrastructure
related,
it's
not
our
code
itself.
B
We
have
to
look.
This
is
just
my
guess
of
a
quick
look
so
because
of
the
test,
as
I
said,
about
the
asp.net
tests
also,
I
think
we
should
keep
for
beta,
at
least
to
have
a
second
look
at
those.
If
there
is
no
wrapper,
then
we
we
are
good,
but
because
of
the
coverage
gap
in
integration
tests
that
are
for
net
framework.
I
think
we
really
should
pay
attention
to
this
kind
of
issue,
just
to
be
sure
that
oh
yeah
is
really
infrastructure
related.
B
So
I
we
covered
all
of
them.
I
don't
have
to
go
back
to
636
and
I'll
change.
The
milestones
this
first
one
here.
B
One
question:
did
you
guys
talk
last
week
about
a
date
for
the
zero
three
release.
B
Okay,
I'm
just
just
thinking
if
there
was
something
in
mind
or
not,
I
don't
think
we
need
to
put
down
a
date
right
now,
but
I
don't
think
he
also
is
that
far.
You
know.
B
B
A
So
I
just
want
to
give
everyone
a
heads
up
that
I'll
be
out
next
week,
and
so
I
won't
be
responding
to
to
get
up
or
slack.
A
Yeah,
though
the
it's
a
whole
company
break
so.