►
From YouTube: 2021-06-02 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
A
Right
so
I
had
one
thing
I
wanted
to
talk
about,
but
jurassic
is
not
here.
I
wanted
to
talk
about
the
oath,
so
I
guess
that
we
need
to
wait.
Maybe
he
joins
a
bit
later.
C
I
have
a
not,
I
don't
need
to
talk
about
it.
I
was
curious
if
there
was
a
release
scheduled
this
week
or
today.
A
C
No
worries:
it's
not
critical.
I
rather
just
trying
to
communicate
something
to
an
end
user.
A
A
B
A
couple
of
things
I
I,
I
missed
a
couple
of
meetings
because
of
some
family
issues,
but
I
remember
that
alolita
had
asked
earlier.
If
I
could,
if
I
could
be,
if
I
could
receive
triage
privileges,
is
that
something
that's
still
possible
or
what's
the
next
step
there?
I
think
that
was
part
of
a
larger
set
of
questions
that
you
had
asked,
I
believe
was.
B
B
Discussion
it's!
I
can
come
back
to
that,
the
other
the
other
question.
So
this
is
a
more
technical
one.
There's
a
discussion
last
time
about
the
z
pages
extension.
I
was
wondering
if
there
are
any
stats
on
like
do.
We
know
if
z
pages
is
widely
used
today
or
we
what.
A
B
A
B
You
know
sure
makes
sense.
I
I
guess
my
the
reason
behind
my
question
was
I
wanted
to
ask.
I
was
thinking
of
using.
I
was
thinking
of
perhaps
enhancing
the
z
pages
to
help
people
understand
what
resources
are
being
detected.
B
So
as
an
as
a
user
of
the
collector,
I
would
know
okay,
I
if
I
want
to
override
such
and
such
thing
here
is
where
I
should
override
it
or
I
tried
to
override
and
my
selection
lost,
here's
how
I
can
fix
it.
B
A
Sounds
like
well.
If
we
try
to
generalize
this,
it
sounds
like
some
way
of
understanding,
observing
the
individual
components
right,
the
state
of
the
individual
components
I
think
well,
yeah
I
mean
it
sounds
reasonable
to
make
it
part
of
the
z
pages.
Okay,
maybe
have
some
sort
of
way
for
components
to
declare.
A
B
A
E
A
So
you're
asking
whether
in
your
product,
you
should
be
in
your
back
end
product.
You
should
be
implementing
otlp
receiving
natively
versus
implementing
an
exporter
in
the
collector.
I
guess
both
are
valid
approaches.
It's
it's.
I
don't
think
we
can
make
a
recommendation
or
the
blanket
recommendation
right.
It's
probably
a
product
choices
that
you
will
need
to
be
making.
A
So
if
you
have
an
application
which
is
instrumented
using
open,
telemetry
sdk
the
that
application
can
emit
telemetry
in
otp
directly
and
that
can
go
directly
to
your
backend
without
the
need
to
use
the
collector
if
you,
but
if
you
implement
an
exporter
in
the
collector
the
collector
becomes
among
that
which
is
in
that
case.
A
So
there
is
that
that
slight
advantage,
I
would
say,
but
I
guess
you
see,
there
is
a
lot
of
expertise
in
the
collector,
so
it
tells
that
a
lot
of
vendors
actually
find
it
maybe
easier
or
prefer
to,
instead
of
modifying
their
back
end,
to
actually
have
a
support
for
their
proprietary
protocol
in
the
in
the
character.
So
both
both
approaches
are
valid.
It's
not
like.
We
believe
one
is
right
and
the
other
is
wrong.
It's
more
of
an
individual
decision
for
each
member.
E
Yeah
makes
sense,
so
another
question
is
that
if
we
have
like,
if
more
and
more
receivers
are
coming
into
the
open,
telemetry
collector
space,
so
I
see
the
like,
the
exporter
should
explicitly
handle
all
these
receivers
right.
E
So
if,
if
there
is
a
new
like
let's
say,
some
new
format,
receiver
is
coming
and
then,
if
I
write
an
exporter
I
need
to,
if
I
need
to
consume
that
data,
then
I
need
to
write
my
exporter
in
a
way
that
it
formats
the
data
which
is
coming
from
that
specific
receiver
plug-in
to
how
I
want
to
forward
it
to
my
backing.
A
No,
that's
not
that's
not
correct.
Actually,
all
the
receivers
translate
the
data
into
common
data
model,
internal
data,
which
is
called
p
data,
and
the
exporters
are
responsible
for
doing
the
opposite.
Translation
from
that
common
data
model.
To
whatever
is
your
destination
format,
so,
regardless
of
what
new
receivers
are
added,
assuming
that
they
correctly
translated
into
the
common
data
model,
your
exporter
should
continue
working
correctly.
A
F
A
A
If
you
take,
for
example,
so
I
can
tell
the
the
specific
combination
that
definitely
works
is
prometheus
receiver
with
other
metric
exporters.
That
combination
is
very
widely
used.
You
use
prometheus
receiver
to
scrape
some
data,
for
example
collector's
own
metrics,
and
then
this
data
gets
converted
into
whatever
vendor
format
you
want
to
using
a
specific
vendor-specific
instructor
that
is
very,
very
widely
used
configuration
the
opposite
may
not
be
so
widely
used,
but
conceptually
that's
the
expectation.
A
You
are
right
in
the
sense
that
open
telemetry's
metric
data
model
is
more
extensive.
It
has
data
types
that
are
not
directly,
maybe
perhaps
or
presentable
in
prometheus,
but
that's
expected
right
there.
That
may
be
specific
conversions
that
are
not
possible
or
maybe
awkward
or
sub-optimal
in
a
sense.
A
They
may
maybe
losing
data
that
that
is
possible,
but
the
generic
concept
still
applies,
that
there
is
a
common
data
model
of
the
data
that
flows
through
the
pipeline
and
the
processors
operate
on
this
data
model
and
the
receivers
and
experts
are
expected
to
do
their
job
to
to
normalize
on
the
receiving
of
the
data
to
this
data
model
and
to
do
the
opposite.
One
in
the
exporters,
when
sending
good
estimations.
A
We
have
what
is
called
correctness
tests
in
the
testbed
and
it's
it's.
It
verifies
the
combinations
of
receivers
and
exporters
on
a
variety
of
data.
Yes,
it
exists
not
for
all
receivers
and
exporters
for
a
few
only,
but
it
can
be
added
it's
a
generic
testbed
which
allows
you
to
test
an
automatically
generated
variety
of
data
with
any
combination
of
receivers
and
extenders.
F
A
A
Right
so
I
see
a
couple
things
in
the
agenda:
yeah
do
you
want
to
so
I
guess
we
talked
about
the
z
pages.
The
other
thing
was
about
triaging.
D
Yeah
I
mean
tigran,
we
had
already
chatted
with
about
pune,
also
helping
on
the
triaging,
so
he
and
I
have
been
working
on
this
together
anyway,
so
I'll
we'll
just
work
on
getting
him
and
getting
an
issue
for
him
and
then
go
forward
from
there.
Okay.
D
The
only
other
point
I
wanted
to
bring
up
sorry
was
on
the
pending
reviews
for
some
of
the
prs
again.
These
are
prs
based
on
the
phase
one
and
phase
two
backlogs.
So
could
I
just
tag
you
again
tigran,
because
I
think
that
some
of
the
general
approvers
are
probably.