►
From YouTube: 2020-11-25 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
A
Hello,
hello,
eater.
I
cannot
hear
you,
oh
what
person.
B
Is
on
my
side,
can
you
hear
me
now
yeah.
A
No,
not
really,
I
think
it
wasn't
a
not
very
nice
time
zone
for
me.
B
C
C
C
C
C
Okay
is
anyone
else
here,
maybe
from
aws
who
knows
about
what
the
first
item
is
about,
so
that
we
can
discuss.
D
I
know
alita
wants
to
talk
about
just
the
prometheus
receiver
and
the
issue
that
we
have
in
general
about
that,
but
I
think
she's
the
right
person
to
actually
go
into
detail
about
what
we
want
to
discuss
there.
C
C
B
Sure,
thank
you,
sir.
Hey
everybody
good
to
see
you
again.
There
was
an
issue
opened
on
the
contrib
folder
contrib
repo
around
there's,
a
user
using
a
signalfx
client
to
export
via
zipkin
v2
to
a
zipkin
receiver
and
then
export
to
new,
relic
and
datadog,
and
these
errors
are
being
are
not
marked
as
errors
in
vendor
backends
and
that's
because
from
my
reproduction
there's
a
in
the
format
they're
being
exported
as
their
the
way.
An
error
is
marked
as
an
error.
It's
just
a
tag
called
error.
B
B
Did
some
great
work
around
unifying
some
of
the
v1
and
v2
zipkin
translation
code
and
anyway,
it's
unclear
to
me
basically
whether
one
there's
an
issue
with
the
zipkin
translation
code
itself
and
there's
work
that
needs
to
be
done
in
the
collector
to
fix
that
or
resolve
it
or
whether
the
issue
is
on
the
client
end.
B
This
signalfx
client,
which
is
like
a
fork
of
a
datadog
client,
seems
kind
of
older,
and
that
is
is
now
forming
the
zipkin
format,
and
so
basically,
I'm
not
totally
sure
where
to
basically
who
to
bug
to
get
this
fixed
or
if
the
answer
is
you
know
it's
not
going
to
be
fixed.
That's
totally
fine
too,
when
I
can
just
add
code
to
the
datadog
exporter
or,
and
I
think
the
new
relic.
B
Fellow
can
you
know
work
around
that
depending
on
if
the
the
user
is
important
to
them
so
eyes
here
would
be
appreciated
or
any
input,
because
I'm
not
super
familiar
with
zipkin
in
general,.
E
So
it's
been
a
while,
since
I
did
that
work.
But
if
I
recall
right,
I
thought
it
was.
You
know
any
tags
that
didn't
or
attributes
that
didn't
match.
You
know
anything
were
just
passed
on
through,
as
is
so
if
they
were
coming
in
in
the
zipkin
a
certain
way
they
should
get
sent
as
attributes
when
it
gets
converted.
So
that's
all
going
to
get
converted
to
otlp
and
then,
depending
upon
whether
you
know.
B
E
C
B
B
Dropped
the
issue
is
that
the
span
status
isn't
being
marked
as
an
error.
It's
it's
not.
You
know
spam
status
and
zip
code.
V2
relies
on,
I
think
status
code
or
something
it
had
some
kind
of
gnarly
heuristics
to
look
at
specific
tags.
So
I
don't
know
if
that
was
not
semantically
correct
and
that's
been
dropped
for
good
reason
or
whether
it
was
just
you
know,
missed
in
the
translation.
So
I
definitely
don't
need
an
answer
off
top.
I
mean
you
know.
B
C
E
B
I
wasn't
not
blaming
anyone
just
yeah
just
trying
to
appreciate
it
mission
accomplished.
I
guess
thank
you
all
for
the
time.
A
E
A
Otlp
yeah
yeah.
We
should
probably
create
a
github
action
to
to
fix
those
issues
on
prs,
because
I've
seen
so
many
but-
but
I
think
eric's
point
here
is
very
interesting
because
I
think
it
there
is
an
opportunity
for
us
to
create
some
sort
of
more
integration
test
or
a
an
acceptance
test,
or
you
know
more
broad
tests,
so
that
vendors
can
then
assert
whether
spans
that
are
being
sent
on
a
given
format
are
then
viewed
with
the
correct
properties.
A
C
C
Probably
I
guess
we
have
the
performance
test
and
we
have
the
correctness
tests,
but
the
correctness
tests
they
verify
that
things
pass
through
when
when
you
go
from
otlp
to
otl
key,
but
they
don't
verify
the
the
the
the
password
for
for
when
the
origin
of
the
data
is,
is
in
a
different
format.
Right,
so
kevin
wrote
those
correctness
tests
which
are
great.
I
think
we
need
a
bit
more
on
this,
so
I
agree
with
you:
yeah
yeah,
yeah.
B
E
F
For
us,
like
at
least
for
the
exporters,
because
our
customers
want
like
hey,
give
me
like
the
exporters
that
this
is
like
proven
to
be
working
well,
so
you
know
we
created
this,
like
conformance
test
for
the
exporters
outside
of
the
project.
So
maybe
we
can
collaborate
to
you
know
bring
it.
F
Yeah,
that's
that's
the
that's
one
of
the
reasons
that
distribution
exists
like
because
you
know
we
just
want
to
be
able
to
verify
and
such
so
I
think
that
we
can
collaborate
instead
of
like
doing
more
work
there
I'll
send.
I
put
a
few
links,
maybe
to
the
test
framework.
We
can
evaluate
and
discuss
later,
like
in
the
later
meetings,
what
we
can
do.
C
Yeah
yeah,
I
think
that's,
that's
a
definitely
good
idea
to
maybe
move
at
least
a
portion
of
what
you
do
on
your
own
distribution.
Maybe
you
still
need
to
do
something
that
is
different
from
the
open,
telemetries
distribution,
but
some
things
likely
things
that
you
do
in
your
repository
can
probably
be
done
here
as
well
right.
They
are.
A
Yeah,
okay,
so
yeah
a
couple
of
ci
tests
for
for
the
builder
itself.
So
if
you're
using
the
open
three
collector
builder,
then
you
could
you
could
reuse
some
of
those
tests
there.
It
basically
just
starts
up
the
the
resulting
binary
and
verifies
the
health
check,
and
I
think
in
one
of
the
users
of
the
builder
that
we
have
at
red
hat,
we
ensure
that
data
is
going
through
jager.
A
So
jaeger
is
our
end
point
our
you
know
final
back
end
and
then
we
we
generate
a
trace
with
a
trace
jam,
and
then
we
check
on
jaeger
by
calling
a
jager
api
to
see
if
the
number
of
traces
that
we
expect
is
there.
C
Okay,
let's
see,
did
anyone
join
anyone
who
had
issues
here
in
the
documentary?
I
don't
think
so.
I
don't
see
anybody.
Okay.
Is
there
anything
else
that
we
need
to
discuss?
B
B
Are
we
not
going
to
discuss
the
prometheus
stuff?
Are
the
right
people
not
here.
C
Yeah
we
can,
or
or
we
can
discuss
it
next
week-
there
is
a
c
meeting
next
week
as
well.
Okay,
after
you,
if
you
want
to
schedule,
works
for
me,.
F
Yeah,
like
we
can
do
it
next
week
and
let
people
know
that,
like
we
will
actually
be
discussing
it,
so
they
show
up.
F
C
Okay,
anything
else
I
see
stuff
is
being
added
for
the
document,
or
is
it
what
we
just
discussed?
That's
about
the
tests
right,
yeah
yeah!
It
is
just
what
we
just
discussed
here:
okay,
okay,.