►
From YouTube: 2022-04-26 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
So
did
you
because
I
only
heard
on
thursday
evening,
did
you
hear
onorak's
news?
No.
B
Yeah,
that's
something
I
did
not
expect,
but
yeah.
That
means
we
are
losing
two
maintainers
in
open,
telemetry
java.
A
A
Yeah
john
brought
it
up
in
this
morning's
maintainer
meeting.
Oh
just
that
the
java
java
is
going
to
be
down
to
one.
A
B
So
what
about
like
other
meetings
like
today's
meeting,
it's
is
santa,
I
mean
what,
when
does
he
live?
Do
you
know?
Is
it
like
right
now
or.
A
So
yeah
we'll
have
to
regroup.
I'm
glad
that
at
least
the
the
metric
stuff,
I'm
really
glad
that
the
metric
stuff
is
gonna
be
stable
yeah.
Now
it's
only
logging
left
yeah.
A
lot
I
can
do
logs
logs
are
easy.
A
Copy
paste
batch
span
processor
over
the
batch,
lock
processor
yeah.
I
mean
it's.
B
B
Have
you
been
to
iraq?
Is
he
going
to
join
today
or.
B
B
Yeah,
I
I
kind
of
hope
that
anurag
will
be
here
today.
They
wanted
to
ask
about
the
start
and
end
times.
B
A
A
A
Oh,
are
you
taking
some
time
between
jobs.
D
B
Well,
have
a
good
rest
and-
and
I
think
we'll
be
missing-
you
yeah.
D
D
D
D
A
Yeah,
I
was
just
thinking
about.
What's
gonna
happen,
nobody's
gonna!
Well,
I
don't
know
matanosh
might
keep
picking
away
at
groovy.
Oh.
B
D
A
Sometimes
I
like
those
brain
brainless.
A
Tyler
benson
once
gave
me
a
compliment.
Oh
that
that's
what
I
remembered
gabe
was
giving
a
compliment,
a
compliment
saying.
I
was
really
good
at
menial
tasks.
D
D
B
Yeah,
I
actually
had
a
similar
line
of
thinking.
I
always
will
whenever
I
was
like
the
the
senior
person
or
leading
it
in
my
team,
I
kind
of
just
hope
all
the
best
stuff
for
myself,
but
I
would
just
use
the
usually
take
care
of
renames
and
like
meaningless
refactorings,
and
all
the
new
features
would
be
done
by
other
people.
A
Matthias
wanted
to
catch
up
on
the
start.
End
time,
yeah.
B
There
was
an
issue
about
that
and
I
didn't
completely
didn't
remember
the
context
for
that.
Why
we
didn't
do
the
fight.
I
remember
there
was
some
discussion.
A
Yeah,
I
can.
The
action
item
is
on
me
to
open
a
spec
issue,
which
is
that
so
we
do
this
clock
per
route
spent
load
per
local
route
span.
A
A
So
then
we
use
the
nano
time
within
that.
A
B
So
if
we
would
just
replace
that
we
would
get
like
shifts,
possibly
yeah,
okay,
I
understand.
D
B
Okay,
cool
so
we're
waiting
for
spec
clarification;
okay,
good.
A
But
still
good
to
yeah
I'll
try
to
get
you
can
harass
me
on
next
monday.
If
I
don't
have
a
spec
issue
open
by
then
okay.
B
B
Is
it's?
It
doesn't
affect
api,
but
it
affects
like
the
behavior
slightly,
so
you
know
it.
We
don't
really
have
to
fix
that,
but
we'd
like
it
to
be
good
to
know
where
we
are
and
what
are
we
going
to
do
with
that
in
the
future?
B
So
yeah
I
mean
I
probably
have
finished
everything
that
I
wanted
to
do
in
the
instrumentation
api.
So
if
you
have
a
bit
of
spare
time,
please
take
a
look
at
it
and
I
think
we're
probably
good
to
go
to
receive
in
the
next
race.
A
A
Yeah,
let's
try
to
get,
would
we
we
would
just
call
it.
We
wouldn't
actually
change
the
tag
or
anything.
B
I
also
like
started
working
on
cleaning
up
the
java
agent
instrumentation
api
and
the
extension
api
and
I've
removed
all
the
all
the
classes
that
we
talked
about
last
time
from
the
in
java
agent,
instrumentation
api
and
now
it's
time
to
merge
them
both
into
one
module,
and
I
have
a
problem
that
I
don't
really
know
a
good
solution
too.
So
I
wanted
to
consult
with
you
so
java
agent
instrumentation
api
is
supposed
to
be
in
the
bootstrap
class
holder.
B
How
are
how
can
we
like
separate
it
into
two
parts
in
great
because
we
need
to
know
in
the
gradle
java
agent
module
like
which
artifacts
are
supposed
to
place
on
the
bootstrap
dependencies
list
and
which
are
supposed
to
be
placed
on
the
java
agent
dependencies
list?
Is
there
any
way
to?
I
don't
know,
maybe
configure
like
separate
source
sets
or
two
artifacts
like
to
produce
two
jars,
just
for
the
generation
build
purposes.
D
Two
source
sets
is
probably
the
most
structured
way
to
do
it,
but,
of
course
any
yet
because
we
can
package
the
jar
even
that
jar
can
be
packaged
into
the
agent
scheme
with
the
instant
all
that
stuff
and
that
can
be
had
done
sort
of
manually
within
a
real
like
we
have
that
relocate
task.
I
think
in
the
javascript
and
some
other
projects,
of
course,
and
that's
more
manual
than
you
can
probably
done.
Also.
D
B
I
think
I
probably
wanted
to
call
it
the
extension
api,
but,
like
the
name
is
it's
a
secondary
concern,
let's
say
that's.
B
B
Would
we
produce
two
jars
like
shaded
separately
or
produce
one
jar
with
oh,
like
we
could
do
the
relocation,
the
instant
class
data
stuff
inside
the
extension
api?
Did
I
get
that
or.
D
D
B
A
Can
we
put
that
combined
artifact
in
both
places
and
have
a
filter
to
filter
out
the
classes,
files,
the
opposite
set
of
class
files.
A
A
Like
you
know,
what
what
am
I
looking
for.
B
Yeah,
so
right
now
it's
just
project
without
any
any
exclusions,
and
I
I
honestly
not
I'm
not
sure
how
how
to
exclude
like
a
package
or
a
class
from
from
the
whole
configuration.
B
B
A
A
Take
a
look
I
sent
on
this,
so
the
I
was
trying
to
build
on
sort
of
your
the
instrumentation
api
instrumentation,
so
yeah
I
like
I,
I
like
the
idea
of
instrumenting
span
key
as
opposed
to
continuing
to
build
more
and
more
into
that
context,
key
bridge,
and
I
think
we
could
then
move
out
these
also
and
then
the
context
key.
B
A
Would
only
be
the
true
like
open,
telemetry,
api
context,
keys,
I
think
the
baggage
and
the
span.
B
A
Yeah
and
I'm
wanna
then
look
at
I'll
look
at
if
this
once
I
finish
that,
then
I'll
look
to
see
if
this
can
be
done
without
the
reflection
stuff,
but
kind
of
needed,
the
the
application
dot
version
of
stuff.
In
order
to
do
the
wrapping
and
stuff
that's
required.
A
Oh
yeah,
this
was,
I
found
a
funny
error
warning
message
in
my
logs
because
it
didn't
update
it
actually
printed
out
this.
B
A
A
A
A
Ended
up
unrelated
no,
what
I
sort
of
re-
oh
it
is,
and
now
I
noticed
you
were
doing
shut
down
now,
which
is
good
at
least
not
like
blocking,
because
I
think
that's.
The
problem
that
I
had
seen
before
with
like
shutting
down
thread
pools.
Is
it's
easy
for
those
to
kind
of
get
stuck
or
take
a
long
time
to
shut
down.
D
D
D
A
It
does
not
just
like
if
you
manually,
re-manually,
like
re-do
the
application
context
or
something.
A
D
D
You
might
have
some
in
progress
or
not
even
database,
like
you,
have
your
graceful
shutdown
of
your
request,
hamburger
as
well
so
like
spring,
when
the
jvm
gets
a
shutdown,
it's
going
to
wait
for
the
request
to
stop
and
then
stop
listening
for
those
close
down.
The
database
connection
we'll
make
sure
there's
no
quarry
still
in
the
box
or
something
so
like.
There's
a
lot
of
things
that
get
closed,
that
probably
shouldn't
be
closed
properly
and
these
things
don't
generally
have
a
flush
method.
A
D
A
I
saw
our
good
friend,
dns
lookup
looks
like
it
attacked
again
on
the
client
side.
Now.
C
D
A
D
D
D
B
Up
golden
week,
not,
but
we
have
the
first
and
the
third
of
my
national
home
place.
A
D
A
Well,
enjoy
the
time
off
next
week
and
on
rock
yeah
enjoy
lots
of
time
off.