►
From YouTube: 2021-02-05 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
B
C
C
C
A
B
C
C
C
C
A
B
C
B
I'm
glad
you
reached
out
and
asked
I'm
gonna
make
everything
really
easy
for
you,
because
you
did
the
right
thing.
I
was
like,
oh
all
right,
although
he
hasn't
figured
out
what
the
what
what
how
he's
gonna
do
today,
but
he's
like
I'll
make
sure
it's
good,
because
you
you
asked
the
question
before
you
sign
something.
C
D
C
C
C
C
Which
I
mean
we
have
the
pump
file,
so
the
pump
file
serves
that
purpose.
I
think
that's
like
the
chat
shadow
plugin
puts
the
com
files
in
so
there
are
attributions
just
not
licenses
basically,
and
if
that's
enough
for
it
like
it,
wouldn't
be
enough
for
a
closed
source,
but
it
might
be
enough
for
an
open
source.
C
E
Yeah,
we'll
just
do
whatever
I'm
I'm
I'm
good
with
doing
our
best
based.
C
E
C
E
C
E
B
C
E
E
I
yeah
yeah.
Let's
do
that
and
you
know
if
we
happen
to
hear
something
different,
we'll
adapt,
but
otherwise
we'll
go
forward
with
that.
C
C
E
So
fab
21
is
a
sunday.
Is
that
what
you.
C
C
E
I
was
just
gonna
go
check
that
issue
not
closed
yet.
Is
it
the
pr,
though,
or
release
one
zero,
zero.
C
C
B
That
won't
impact
releasing
the
sdk,
though,
because
that's
all
part
of
yeah
auto
configuration
right,
yep.
E
C
E
The
private
key
is
is
really
for
reverse,
like
the
collector
has
to
authenticate
back.
E
B
Mdns,
it's
like
a,
I
don't
know
it
stands
for
mini
or
micro
or
it's
oh
multicast.
I
see
yeah
see
I
didn't
even
know
what
it
was
for,
but
this
is
like
what
the
when
you
have
your
mac
and
you
can
do
like
whatever.local
because
it
has
like
auto
network
discovery
built
in
like
it
does
some
sort
of
broadcast
multicast
dns,
it's
very
handy,
but
it
has
nothing
to
do
with
mtls.
B
Oh,
I
don't
know,
I
do
think
I
did
want
to
get
in
that
hotel
resource
attributes
auto
configuration
pr
if
we
can.
So
if
you
can
take
a
look
at
that,
I
mean
if
that
needs
to
go
and
later
it's
part
of
the
it's.
You
know
it's
not
part
of
the
stable
stuff
like
that.
Still
would
be
nice.
It
would
be
helpful
for
splunk
at
least,
but
I
don't
think
it's
a
big
deal.
B
Useful
just
to
be
able
to
have
these
in
the
default,
like
the
rest
of
the
things
end
up
in
the
default
like
if
it
feels
weird
to
have
all
of
that
the
stuff
in
our
resource
module
show
up
in
the
default.
But
these
ones
not
like
this,
which
is
like
the
most
basic,
makes
sense.
E
B
Properties,
yeah
they're,
coming
with
documents
or
properties
so
that
you
can
use
them
in
instrumentation
to
do
things
that
are
not
specified
by
the
spec.
C
C
Yeah,
that's
the
whole.
Let's
log
correlation,
though
right
we
just
haven't
done
an
open
telemetry,
yet
so
you're
doing
that
in
your
instrumentation.
But
theoretically
I
have
once
we
have
logs
in
open
telemetry,
yes
yeah!
This
is
definitely
a
short
term.
Hopefully,
okay,
I
don't
know
it's
only
for
logs.
I
thought
you
were
using
it
in
something
more
general,
okay,
yeah.
E
Now,
once
once,
you
start
putting
it
in
the
system,
properties.
E
B
B
Yeah,
so
I've
been
working
on,
I
put
in
a
pr
actually
today
to
upgrade
spring
sleuth
open
telemetry.
I
don't
know
what
it's
called
whatever
that
integration
is
o
15.
They
had
upgraded,
oh
fourteen
one,
but
had
never
hadn't
gone
and
removed
all
the
deprecations.
So
that
was
like
I
needed
to
go.
There's
a
whole
bunch
of
decoration,
scrubbing
that
had
to
happen
so
yeah.
I
think
jason
just
put
in
the
link.
B
But
then
that
also
uncovered
this
whole
weirdness
with
global
propagators
that
since
they're,
using
our
unpublished,
instrumentation
apis
that
all
use
by
default,
you
had
to
set
the
global
as
a
part
of
your
setup
and
then
their
testing
all
broke
everything
when
I
removed
that
and
then
15
broke
it
because
you
can't
set
it
more
than
once.
So
all
the
tests
were
broken,
so
I
had
to
put
reset
for
test
in
the
middle
of
their
production
config,
which
is
you
know,
super
not
awesome.
B
E
Hey,
I
was
really
happy
for
a
reset
for
tess.
I
was
happy.
I
was
helping
the
azure
sdk
folks
update
to
zero
15
also,
and
the
tests
were
breaking
also
right
with
the
camp
set
more
than
once,
and
I
was
like
oh,
I
know
that
one.
B
F
F
B
Yeah
yeah
at
some
point,
so
it's
really
on
there
and
you
know
if
they
don't
get
to
it
before
release
16
I'll,
just
update
it
the
pr
to
update
it
to
16..
So
you
know
it's
all
good.
I
also
really
confused
about
their
like
of
that
part.
You
were
just
looking
at.
I
don't
know.
Maybe
someone
who
knows
maven
better
understand
why
they
import
every
dependency
individually
and
then
exclude
any
sub
dependencies.
B
B
C
B
B
C
B
E
So,
john,
I
have
a
question
about
so
that
you've
looked
at
some
of
this
code
because
I'm
thinking
about
for
us-
and
I
mean
so
now
that
the
once
the
sdk
is
one
zero.
It
would
be
awesome
if
the
flute
folks
could
officially
release,
but
they
have
this.
They
have
a
dependency
on
our.
You
know
instrumentation
api,
which
we're
not
planning
to
have
stable
for
a
few
months,
probably
a
couple
few
months
yeah.
E
Do
you
think
they
can
vendor
this?
In?
Did
you
happen
to
notice
close
and
look
closely
enough
like?
Are
they
exposing
that
as
public
api
surface
or
what
they're
using
this?
For
so.
C
C
F
B
Originally,
it
was
all
marked,
it's
all
been
marching,
although
it
looks
like
the
the
other
person
tommy.
F
B
C
B
C
E
I
I
tired,
under
our
metrics
interop
instrumentation,.
B
Actually,
speaking
of
metrics,
so
we
haven't
really
talked
about
my
pr
to
to
spam
the
logs
on
suppressed
spans,
but
I
was
wondering
whether
this
actually
is
a
use
case
we
could
use.
We
could
do
this
with
metrics
like
we
could.
We
could
actually
count
suppressed
spans
and
tag
them
with
the
like
label
them
with
the
instrumentation
and
just
have
some
have
a
counter
reading,
putting
it
in
the
logs.
B
G
B
It's
all
good
yeah,
so
I
may.
I
may
update
that
to
use
some
metrics
to
count
that
stuff
and
hopefully
we
can
make
it
easy
to
set
up
the
metrics
logging
metrics
exporter.
So
we
can
tell
people
for
customer
support
to
turn
that
on,
and
so
we
can
see
what's
being
suppressed.
B
That's
that's
a
call
for
trask.
I
think.
E
Well,
the
the
oh,
the
library,
instrumentation
yeah,
the
java
agent
has
no
api
surface
and,
from
the
from
our
understanding
in
bogdan,
you
can
tell
us
if
you've
heard
any
differently.
The
latest
we've
seen
in
the
spec
about
telemetry
stability
is
that
that's
not
going
to
be
required
for
one
zero.
G
In
general,
as
a
one
or
one
two
zero,
we
said
that
configuration
is
stable
if
they
are
not
in
the
spec.
Probably
a
good
discussion
would
be
to
add
them
in
the
spec
as
as
incrementals
to
the
spec,
because
spec
spec
is
not
freezing
at
one
zero.
I
mean
we
can
it's
freezing
from
the
perspective
of
not
breaking
it,
but
we
can
always
add
new
capabilities
and
stuff.
E
Okay,
yeah,
that's
something
we
should
discuss,
then
maybe
we
can
take
a
couple
minutes
now
because
from
previous
discussions
on
the
java
agent,
specifically,
we
had
agreed
that
we
would
have
stable
config
properties
for
anything
that's
in
the
spec,
but
we
wouldn't
guarantee
stability
for
things
that
are
not
in
the
spec.
E
G
I
I
don't
understand
you,
and
especially
if
specs
comes
and
define
a
slightly
different
viable
for
the
same
purpose,
and
you
already
define
one
not
sure,
but
I
can
tell
you
from
the
user
perspective
that
will
annoy
hell
of
them.
If
we
change,
if
we
change
the
variable
so
no
matter
what
is
the
reason
they
will
be
annoyed
on
us,
so
I
don't
know
what
to
how
to
to
deal
with
this
problem.
But
this
is
the
reality.
E
E
G
I
mean
if
you,
if
you
hide
them
behind
the
experimental
flag.
I
think
it's
okay,
I
mean
if,
if
somebody
shoots
themselves
in
the
foot,
it's
one
thing
versus
we
declare
something
stable
and
then
they
don't
get.
E
E
Cool,
I
will
I'll
open
a
issue,
a
v1
issue
and
we
can
go
from
there
just
so
we
don't
lose
track
of
that.
E
For
some
reason
I
dumped
this
in
here:
let's
see
what
it
was.
Oh,
oh,
yes,
the
agent,
whether
we
want
the
agent
to
only
instrument
global,
open,
telemetry
or
not.
E
I'm
not
I'm
not
totally
convinced
that
we
wouldn't
want
the
agent
to.
I
don't
know.
I
can
see
both
sides
of
like
if
somebody
has
a
an
application,
say:
they've
written
it,
you
know
a
spring
app
and
they've.
You
know
added
some
instrumentation
injected
open,
telemetry
object
and
then
they
want
to
add
the
java
agent
later
on.
C
C
B
Yeah,
I
think
what
I
was
calling
out
here
is
it's
really
weird.
It
would
be
really
weird
and
surprising
if
they'd
done
a
whole
bunch
of
configuration
and
then
the
agent
came
along
and
made
all
of
their
the
instrumentation
work,
but
their
configuration
was
not
used
at
all.
So
I
think
that's
a
that's
kind
of
the
weird
case
that
I
mean
that's
also
going
to
be
the
case
if
they
assign
something
to
a
global,
open,
telemetry
like
that'll,
be
ignored.
Also,
I
don't
so.
I
don't
know
that
there's
a
good
way
around
it.
E
But
do
you
think
it
will
be
less
surprising
that
the
agent
does.
B
G
I
wanna,
I
wanna,
ask
you
something
trust.
What
does
it
happen
if
if
the
user
brings
an
sdk
and
the
agent
comes
with
another
sdk,
because
that's
the
only
possibility
where
they
said
something
to
the.
E
C
G
E
B
My
guess
is:
there's
no
right
answer
to
this.
Some
people
will
be
surprised
by
one
and
so
only
the
surprise
value.
How
come
how
hard
would
it
be
to
make
it
an
option
like
hijack?
I
don't
know
dash
dash,
hijack
all
open,
telemetry
instances
equals
true,
I
don't
know
just
making
stuff
up,
but
you
know
having
an
option
in
the
agent
that
would
it
would
force
all
of
them
to
be
there
if
a
user
really
wanted
that
but
defaulted,
just
the.
G
G
The
other
question
would
be
they're,
probably
gonna,
be
users
that
want
only
all
the
plugins
to
be
installed
by
the
auto
instrumentation
and
everything
except
the
sdk,
and
they
want
to
have
their
own
sdk,
configure
and
stuff.
Is
that
a
possibility.
F
B
Was
that
you
like
somebody
wants
to
be
able
to
run
in
both
modes,
switching
back
and
forth,
and
so
they
might
configure
everything
and
then
run
it
and
test
things
out
and
they're
like.
Oh,
let's
go
to
the
agent
and
not
change
their
code
at
all,
just
plug
in
the
agent
and
then
things
will
maybe
behave
differently.
F
E
Yeah
we're
going
to
get
a
long
tale
of
of
people
doing
things
that
are
not
yeah
doing
all
kinds
of
things.
I.
C
E
Yeah,
I
I'm
definitely
good
with
this.
I
mean
it
because
the
the
typical
agent
user
is
like
hey
just
give
me
auto
instrumentation
and
then
they
want
to
come
in
and
add
you
know
their
custom
stuff,
and
so
it's
easy
to
say:
hey
use.
The
global,
open,
telemetry
object
for
your
custom
stuff
and
I
feel
like
that's,
going
to
be
the
the
big
majority,
but
yeah
we're
going
to
have
a
lot
of
other
stuff
too
yeah.
G
I
think
I
think,
starting
with
that
goal
and
clearly
documenting
that
that's
the
usps
and
that's
how
we
suppose,
people
to
use
it's
it's
a
good
start.
I
mean
we
can
always
improve
things,
but
documenting
that
this
is
the
expected
behavior.
This
is
what
we
want.
We
expect
users
to
do.
It's
the
right
thing
to
start.
G
B
C
B
The
top
I
forgot,
what
did
I
put
I
put
in?
No?
It
was
a
in
a
discussion
I
think
max
from
from
splunk
put
in
the
pr
to
document
to
add
documentation
for
that.
C
B
C
G
G
E
E
This
whole
section
definitely
needs
to
be
like
here.
If
you
want
to
know
the
the
details,
yeah
go
somewhere
else.
A
G
The
other
thing,
by
the
way,
john
to
blame
on
us.
We
should
also
have
the
same
thing
on
the
main
repo.
B
Oh
yeah,
I
agree,
I
agree.
Actually
the
thing
that
I
think
is
most
interesting
is
that
we
have
a
whole
bunch
of
examples
that
are
very
they
like
cover
a
bunch
of
cases
and
we've
no
one
ever
finds
them.
Like
I
answered
the
same
question
over
and
over
again
like.
I
need
an
example
of
how
to
do
tracing
manual
tracing
with
with
propagation
and
hdp
I'm
like.
Well,
we
have
this
nice
example.
B
B
G
I
always
read
that
getting
started.
It
is
half
a
page
because
I
want
to
have
something
running
and
then
I
stopped
reading
documentation
and
try
by
myself.
So
I
think
I
think
the
getting
started
is
the
most
important
and
critical
part
and
to
be
short
like
just
having
something
working
and
then
expect
people
to
to
just
play
around.
E
So,
in
that
sense
that
we
do
have
the
agent
has
a
nice
getting
started,
it's
you
know,
half
a
page
here
and
you
do
get
up
and
going,
but
it
completely
like
skips
over
any
of
the
custom
instrumentation
stuff.
G
F
B
E
This
still
refer
to
hotel,
trace.
B
C
C
E
C
B
B
B
E
C
C
I'm
also
I'm
planning
on
publishing
the
aws
distro
as
a
base
image.
Also-
and
I
don't
know
if
we
want
to
do
that-
but
the
base
image
that
just
packages
the
agent
and
sets
java
tool
options.
It's
not
going
to
handle
every
case
like
if
they
want
to
set
their
own
java
to
options
just
like
that.
But
many
apps
can
then
just.
E
C
E
Of
like
which
java
distro.
E
C
C
F
E
C
F
Do
we
open
telemetry
java,
instrumentation,
already
published
to
dockerhub.
C
E
C
I
think
the
main
thing
that
we
talked
about
today
is
like
we
don't
mention
global
telemetry
and
the
last
paragraph
I
just
link
straight
to
the
tracer
api,
but
there
we
need
to
add
a
bit
about.
You
can
get
a
tracer
using
global,
open,
telemetry
and
then
use
the
tracer
api
to
do
a
lot
of
stuff.
That
would
be
a
huge
improvement
in
that.
C
C
C
F
E
C
E
Only
down
the
downside
I
see
here
is
it's
not
as
obvious
to
people
how
to
contribute
to
these
pages,
where.
G
A
E
Is
hiding
behind
your
your.