►
From YouTube: 2021-08-18 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
C
E
It's
fine,
it's
fine!
So
let's
look
at
the
agenda
and
start.
F
Yep,
hello,
everyone,
so
I
have
two
questions
for
the
open,
telemetry.
F
Sorry,
like
I
just
had
some
noise
coming
back,
so
I
have
two
questions.
The
first
question
is
that
when
I'm
trying
to
access
it
into
the
open,
telemetry
corrector,
it
doesn't,
let
me
do
it.
So
I
think
it's
because
of
the
base
image
doesn't
support
shr
bash.
So
is
it
any
conscious
decision
or
why
is
it
not
leading
us
to
ssh
into
the.
G
F
Okay,
so
if
it's
any
conscious-
which
I'm
totally
fine
because
like
when
I'm
trying
to
configure
open
telemetry
collector
to
write
the
traces
into
a
log
file
into
a
pvc,
so
when
I
want
to
test
that
out,
look
into
the
data,
so
I'm
unable
to
access
it
into
the
container,
which
means
I
need
to
scale
down
the
collector
and
then
start
a
test
container
to
mount
the
same
pvc
to
look
into
the
data.
So
it
was
a
bit
hassle.
F
So
I
just
wanted
to
confirm
on
that
so
yep
and
the
other
thing
is
that
yep.
So
when
I
was
like
when
I
and
like
I
was
talking
to
a
few
other
folks
who
are
trying
to
evaluate
open
telemetry
collector
to
use
so
like
they
do
not,
they
always
ask
me:
what
is
the
demo
setup?
I
can
use
to
evaluate
and
show
it
to
the
team
to
push
the
usage
of
open,
telemetry
collector
forward.
F
So
why
my
proposal
here
is
that
why
don't
we
maintain
our
own
clone
of
hipster
shock,
because
I
see
light
step
honeycomb
all
like
everyone?
Even
the
google,
like
the
hipster
shop,
has
started
from
the
google
project,
so
I'm
a
demo.
F
E
B
Yeah
I
mean
we
need
to
just
open
up
an
issue
on
the
collector
and
just
say
hey.
You
know
this
is
the
demo
that
I'd
like
to
add
and
just
add
a
one-page.
You
know
proposal
for
it.
B
B
Yeah
I
mean
again,
I
just
was
doing
a
general
call
out
for
any
pr
reviews
that
folks
are
waiting
on.
I
know
there
are
pr
reviews
that
would
be
welcome
for
the
one
I
listed
that
the
emanuel
is
working
on,
but
there
are
others
also,
I'm
sure
if
folks
are
blocked
in
any
way.
You
know
just
please
list
it
we'll
follow
up
on
it.
J
K
K
B
H
Yeah
for
sure
I
welcome
them.
Yeah
no.
E
E
While
we
were
in
in
a
bad
state
with
the
with
the
matrix
transition
to
attributes.
So
half
of
the
components
will
try
transfers.
Half
of
them
will
not
and.
C
E
I
have
a
pr,
I
think,
I'm
done
with
chord.
Then
I
need
to
to
make
a
release
day
for
record.
K
E
Yeah
it's
used
by
by
others
as
well
yeah.
I
will
probably
add
that
anyway,
for
tracing
doesn't
have
any
impact,
but
for
metrics
it's
randomly
working.
So
that's
what
it
is
right
now.
E
I
will
I'll
respond
to
you.
I
just
woke
up.
K
B
I
think
bogdan
we
can
give
an
update
on
the
move.
Anthony
did
you
want
to
give
a
quick
update,
because
just
so
that
folks
are
aware
of
what
we
are
doing
on
the
collector
core
to
contrib
move.
K
Sure
so
I
guess
I
can.
I
can
update
this
group
from
bargain
certification.
I
don't
think
anything's
really
changed
since
we
talked
yesterday
because
we
were
waiting
for
this
release
to
finish
up,
but
where
we're
at
is
that
we
have
identified
a
list
of
components
that
will
move
from
the
core
repo
to
the
contrib.
K
Repo
punya
has
given
us
a
proof
of
concept
that
shows
how
we
can
maintain
the
commit
history,
so
we're
working
with
that
process
to
move
over
first,
some
of
the
dependencies
when
we
when
we
started
out,
we
identified
that
internal
test
data
and
a
couple
of
other
internal
packages
in
collector
core
were
dependent
on
by
some
of
these
components
that
were
moving
so
we're
going
to
move
first,
those
internal
dependencies
into
contrib,
so
that
they're
available
there
as
well.
Then
we
will
move
the
the
packages
out
of.
K
We
will
create
prs
that
will
copy
the
commit
history
for
individual
components
into
contrib.
We
will
get
those
merged
in
so
that
the
components
will
appear
there
along
with
all
of
their
prior,
commit
history
and
will
update
some
of
the
necessary
references.
K
Like
the
references
to
the
new
internal
packages
that
exist
in
contrib,
once
we
have
moved
all
of
those
or
copied
all
of
those
over
to
contrib,
we
will
then
start
removing
them
from
core
deleting
deleting
the
the
copies
from
the
from
the
head
branch,
at
which
point
we'll
then
have
to
go
back
and
update
contrib
to
ensure
that
any
references
to
dependencies
that
were
moved
have
also
been
updated.
K
So
it'll
be
kind
of
a
three-step
process
of
make
copies
of
them,
remove
them
from
core
and
then
fix
what
has
been
broken
in
contrib
punya
was
trying
to
help
us
build
a
dependency
graph
so
that
we
could
maybe
ease
some
of
the
pain
in
that
third
step,
but
I
think
we've
run
into
some
snags
there.
So
we
expect
this
approach
will
work
and
that's
what
we're
proceeding
with.
B
Yeah
and
and
we're
going
step
by
step
so
component
by
component
so
again
just
super
careful
about
not
breaking
tests
and
also
retaining
commit
history
and
any
other
dependencies.
As
anthony
mentioned.
E
Yeah,
so
in
that
regard
anthony,
I
would
ask
you,
because
you
are
also
so
first
of
all
for
everyone.
Anthony
became
the
responsible
for
this.
Thank
you
anthony
for
for
doing
it.
Every
time
when
you
apply
when
you
move
one
component
to
a
contributor
make
a
copy
there,
I
think
you
should
be
very
careful
and
go
through
the
prs
in
core
and
at
least
close
them
until
you
tell
the
the
person
to
to
open
in
in
the
other
level,
yep
sure
I
can
do
that.
B
I'll
work
bogdan
with
anthony
to
make
sure
that
that's
what's
flagged.
C
E
E
A
E
D
Just
quickly
regarding
this
last
topic
is
that
issue
3474
the
one
to
track
for
when
this
will
be
sort
of
like
released.
B
Or
public
yeah
eric
yeah
we're
just
putting
everything
in
there.
You
know
that's
the
tracking
issue.
B
E
I
have
one
problem
in
country
if
somebody's
one
of
these
people
are
here.
Let
me
check
who
was
this
so.
E
Is
pablo
here.
A
E
Yeah,
I
don't
know
where
that
comes
from
that
vmi
also,
it
seems
to
be
another
problem,
it
seems
we
have
and
maybe
jurassic
year
we
have
a
problem
with
jager.
We
a
circular
dependency
that
we
probably
have
to
at
one
point
address.
E
E
Yeah
so
so
the
problem
is,
and
we
will
remove
that
from
core
with
your
work.
Yeah.
B
E
But
but
the
problem
that
we
have
is
that
we
need
from
miaga
we
need,
we
need
the
models,
because
we
want
to
transfer
translate
from
into
jager
format,
but
that
comes
with
the
with
the
dependency
on
us
as
well,
and
it's
it's
it's
a
mess.
They
depend
on
0
28,
which
is
super
old
and
stuff.
So.
C
B
Okay,
so
again,
I
think
I
think
bogdan
this
will
get
fixed,
because
when
we
move
jager
you
know
jaeger
will
we'll
have
the
dependencies
also
looked
at
right,
so
we
can
at
least
flag
it
that
it
needs
to
be
updated
and
then
work
with
jurassic
too
yeah
get
it
fixed.
E
E
E
H
E
A
K
K
B
There
was
one
more
thing
I
just
wanted
to
call
out,
and
this
is
just
useful
for
folks
who
are
looking
at
deploying
the
collector
using
an
operator
using
our
hotel
operator.
We
made
some
modifications
and
you
know
enhanced
the
operator
which
jurassic
and
our
team
has
been
working
on,
and
we
also
added
a
helm
chart
to
the
to
hotel
where
you
can
actually
deploy
the
operator
and
the
collector
instances.
So
please
take
a
look
and
you
know
everything
has
been
merged
at
this
point.
B
We
also
published
some
extensive.
You
know
configuration
and
implementation
details,
so
I
just
shared
the
post
but
should
be
super
useful.
You
can
actually
use
that
for
pulling
in
metrics
from
any
infrastructure.
You
know
on-prem,
self-managed
or
services
to
be
able
to
ingest
that
into
any
service
endpoint
for
prometheus
remote
right,
especially.
E
So,
okay
thanks
everyone
anthony.
If
you
don't
mind
if
you
have
an
example
on
on
that
retreat
thing.
Otherwise
I
would
read
the
documentation,
but
that
was
just
lazy.
K
Yeah,
I
think
the
the
documentation
probably
discovers
that
you
put
a
retract
statement
with
the
version
that
you
don't
want
anybody
to
use
anymore.
Okay,
but
do
you
have
to
put
it
on
that
release
or-
or
you
put
you
you
put
it
in
the
next
release,
so
you
would,
you
would
be
in
the
go
mod
for
point.
33
would
say
retract
point
32.
Oh
I
see
I
see.
I
see.
Okay,.