►
From YouTube: 2022-04-19 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
A
Yeah,
it
was
easter
over
here
and
we
have
a
bank
holiday
monday.
A
B
Yeah,
I
think
in
the
us
at
least,
like
goes
back
to
like
the
official
days
that
banks
were
closed,
were
like
one
set
of
like,
like
the
most
official
set
of
holidays
because
like
if
you
ask
like
what,
when
schools
are
closed,
like
they
have
all
these
other
holidays,
like
columbus,.
A
B
Did
you
already
know
about
john,
or
did
you
see
on
slack
this
morning?
Okay,
I
I
just
saw
it
this
morning.
A
It
was
was
a
surprise
for
me.
Well,
he
he
changed
the
team
right,
so
we
haven't
been
talking
that
much
recently.
C
B
It's
certainly
been
in
observability
from
new
relic
and
splunk.
Now
for
a
good
while.
B
B
The
release
did
go
out
on
a
good
handful
of
issues
that
worked
through
some
that
held
up
the
release
and
some
that
were
just
found
by
the
release
that
I
fixed
after
the
release.
B
So
third
time's
the
charm
can
trip
repo
in
tomorrow,
or
I
guess
I
should
ask-
is
there
anything
there's,
probably
nothing
that
we
need
to
hold
up,
but
they
can
trip
release,
for.
I
can
probably
try
to
run
that
tomorrow.
B
You
should
let
me
do
this
one
yeah:
what's
what's
the
we've
gone
through
all
three
smoothly
with
this
new
stuff
then
have
at
it.
B
Want
to
I
want
to
use
it
actually
there's
some
wild
bit
wise.
Why
the
best
one,
though,
let
me
find
this
one,
oh,
my
goodness,
double
tara,
long
bits.
B
B
B
B
Made
no
this,
I
could
not
understand
for
the
life
of
me.
It's
wacky,
the
the
exponent
encoding
for
I,
that
yeah.
C
C
A
Saw
this
vr
earlier
and
seeing
how
complicated
it
is
and
that
it
has
like
2
000
lines
of
code
of
changes.
Okay,
I'll
leave
it
for
later.
A
Yeah
yeah,
I'm
really
glad
and
we'll
probably
start
migrating.
Our
database
connection
pool
and
the
instrumentations
over
to
the
ultimate
represent
nice
yeah.
C
A
We're
waiting
for
the
jvm
matrix,
it
was.
B
A
A
On
it,
it
has,
I,
I
haven't,
been
attending
the
jvm
metrics
meetings
recently.
A
And
my
calendar
and
the
number
of
things
that
I
have
to
do
got
too
big,
so
I
think
jason
was
attending
this,
so
I
just
dropped.
B
Yeah
we
didn't
nobody
joined
this
morning
except
jack,
and
I
so
hopefully
it's
just
cyclical.
I
will
get
jonathan
in
particular
back.
B
Yeah
I'll
get
bugged.
In
a
couple
days
I
made
I
got
ping
riley,
I
have
a
now.
I
have
a
friend
on
the
tc
side
channel
to
the
tc,
see
if
he's
got
if
he
feels
like
chiming
in
and
asking
bogdan
nice
also.
B
Well,
yeah:
I
was
waiting
on
this
one
to
give
you
a
chance
to
look
and
weigh
in
matish,
I'm
sure
today.
I.
A
Okay
yeah,
I
briefly
saw
your
reviews
on
the
spanky
vr
and
I
completely
agree
with
the
local
root
span.
Comments
that
you
made.
I
mean
the
whole
br
shows
that
the
concept
of
the
local
span
is
just
pretty
much.
It
should
be
a
first
class
concept
in
the
instrumentation
api.
A
So
while
I
was
writing
it,
I
had
like
two
ideas
about
it:
either
we
like
remove
server
spam
and
migrate
to
local
respawn
completely,
or
we
have
a
separate
local
spam
and
a
separate
thing
called
http
server
spam,
which
is
kind
of
like
a
more
specific
thing.
It
tries
to
extract
the
http
server
span
using
the
spanky,
and
if
it's
not
there,
it
uses
the
lockout.
B
A
Oh,
it
should
be
even
used
in
in
less
places
than
before,
because
all
this
http
server
instrumentations
got
migrated
to
the
http
root
holder
thing,
so
it's
probably
mostly
for
customizations
like
by,
for
example,
I
know
we
have
one
customer
that
uses
it
to
set
just
some
responding
and
probably
some
you
know,
jxws,
instrumentation
and
stuff.
Something
like
that.
B
A
A
I
think
so
I
mean
unless
they
assuming
that
we
merge
this
vr
unless
they
switch
to
the
non
suppression
strategy,
it
will
always
mean
like
http
server
spam.
If
you,
if
you
call
local
root
span
in
the
controller
method,
it
will
always
be
hdb
server
spam.
A
You
do
something
weird
like
changing
the
separation
strategy
to
don't
suppress
anything.
A
A
B
B
So
this
is
a
package
protected.
So
first
I
was
like
well,
we
don't
need
the
disclaimer,
but
then
I'm
like
wait.
B
A
I
don't
know
I
yeah,
I'm
not
sure
what
magic
it
uses
to
make
it
work,
but
the
I
just
pretty
much
look
at
the
dragon
box
of
the
two
method
handles
look
up:
factory
methods
and
the
lookup
one
says
that
it
can
see
non-public
methods
basically
or
the
public.
One
says
that
it
can
only
see
public
ones
yeah.
Something
like
that.
So
I
think
that
by
just
by
using
the.
A
B
This
method
is
color
sensitive,
see.
I
think
that
it
it.
It
depends
on
the
color,
where
you
call
look
up
that
it
inherits
the
permissions.
B
C
A
Fairly,
it
works
now,
but
we
don't
test
it
test
it
on
jpms
on
any
kind
of
module
scenario.
So
I
wonder
how
it
will
work
in
that
case.
B
No,
no,
I
just
I
so
with
reflection.
B
C
All
right,
but
I
mean
if
we
wanted
some
sort
of
mid-ground
that
were
easy
to
reason
about,
like
a
package
private
class
with
public
methods
that
I
can
reason
about,
because
I
always
know
that
factors
private
classes,
it
shouldn't
actually
matter
in
the
wrong
time.
That's
my
understanding
of
them
as
long
as
the
methods
are
public
that
I
would
expect
to
work
no
matter
what
so,
but
it's
still
kept
out
of
the
compiled
compilability
from
the
user.
C
B
Wants
to
that's
another
biggie.
A
Yeah,
you
can
leave
that
one
to
me.
I
was
hoping
that
anna
will
manage
to
finish
their
her
refactoring
of
the
previous
pr,
so
that
you
don't
have
to
build
the
agent
twice
because
it
was
going
pretty
well
so,
but
I
probably
can
take
a
look
at
it.
First.
B
All
right
yeah,
if
you
don't
get
to
it
in
the
next
few
couple
days,
I'll
probably
be
I'm.
I
I'm
just
burnt
out
on
reviews
right
now,
but
give
me
two
days
and
I'll
be
up
for
looking
at
another
one.
Another
big.
A
B
B
C
C
B
Yeah,
so
I'm
not
too
real,
and
also
I
just
I
mean
at
least
we
don't
have
very
many
j9
customers
running
j9
and
then
the
the
smoke
test-
one,
I
think,
is
mostly
just
the
container
running
out
of
memory,
which
is
what
laura
is
looking
at
yeah.
I
think.
A
B
B
Then
yeah
this
one.
This
is
a
great
another
great
contribution.
I'm
really
happy
for
the
jboss
log
manager
support.
I.
B
Also
so
I
think
we're
on
a
good
track
there
I
agree
with
you
about.
I
don't
think
we
need
this
instrumentation.
A
Yeah
and
the
funny
thing
I
I
looked
at
the
look
for
j1
instrumentation.
That
also
has
these
methods
and
probably
that
either
the
javascript
look
for
j1
libraries.
You
know
copied
from
each
other,
so
I
looked
at
our
instrumentation
and
if
you
comment
that
one
it
still
works.
So
I
have
no
idea
why
I
wrote
it.
A
If
you
look
and
look
for
j1,
it
has
an
instrumentation
instruments,
two
methods.
There
was
one
that
of
this
one,
the
md
yeah,
there's
one
that
it's
very
similar
j
plus
one
one,
that's
just
a
vlogging
event.
I
think
yeah,
one
that
just
retrieves
one
time
of
nbc
and
the
other
one.
That's
called
get
mdc
copy,
but
doesn't
actually
return
anything.
It's
just
a
void
method.
A
B
Yeah
but
since
we
are
storing
the
span
or
should
be
context
into
the
logging
event,
we
are,
we
don't
need
to
copy
the
thread
local
in
yeah
when
they
do
that
copy.
B
All
right
anything
to
look
at
in
the
core
repo.
C
C
C
C
B
A
I
think
we
talked
about
it
on
thursday.
A
B
Yeah
someone
asked
about
hikari
metrics,
so
right.
B
My
I
I
got
well
as
you
can
see
with
this,
but
the
lovely
lighting
that
I
have.
I
I
ordered
some
lighting,
so
maybe
you'll
see
my
face
on
thursday.
Maybe
I'll
have.