►
From YouTube: 2021-09-07 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).
B
B
Now
this
one
is
first,
so
this
one
applied.
This
one
applies
to
our
version
branches
and
mateish.
We
had
looked
while
we
were
on
that
call.
We
had
looked
at
because
last
week
the
governance
committee
was
supposed
to
discuss
whether
we
were
allowed
to
do
this,
since
it
is
essentially
subverting
the
or
the
the
cla
check
and
we
pulled
out.
B
We
read
the
notes
and
they
said
that
it
was
okay
as
long
as
there
was
no
either
copyrightable
changes
which,
as
applies
to
our
gh
pages
since
we're
just
putting
those
results
in
there
for.
C
B
And
the
or
if
it
was
something
like
it
was
just
like
cherry
picked,
you
know
like
no
no
changes,
so
that
should
cover
us,
for
here,
at
least
until
they
figure
out
how
to
make
it
work
for
real,
with
the
the
github
action
spot
and
the
ecclhx.
B
Yeah
good,
I'm
glad
they
I'm
glad
that
wrote
it
down
in
the
meeting
notes
and
were
legit.
B
A
Well,
I
have
one
kafka
pr
that
I
would
like
you
both
to
review
if
it's
possible,
because
it's
well,
this
one
is
not
that
controversial,
but
it
might
be
the
the
next
one
will
be
so
I
tried
to
implement
the
you
know
two
spans
one
for
receive
one
for
process
and
have
the
process
span
be
a
child
of
the
receives
one.
I
didn't
manage
to
do
that
second
part,
yet
because
of
various
reasons,
but
what's
what's
a
bit
new
and
what's
controversial?
Is
it's
not
in
this
pr?
A
Yet,
but
by
the
time
the
pro
yeah
by
the
time
the
process
consumer
span
starts,
they
receive
span
is
already
finished,
so
you
have
to
manually
construct
the
context
with
the
correct,
receive
spam
spam
context
and
pass
it
around.
So
it's
something
that
hasn't
been
done.
Anything
l
anywhere
else
in
this
repo.
So
it's
a
kind
of
a
new
thing.
A
B
A
B
A
A
So
it's
it's
kind
of
complex
and
strange,
but
I
think
that's
because
of
this
whole
batch
processing
and
but
receiving
scenario
that
it
makes
it
a
little
bit
overcomplicated.
A
B
A
Didn't
at
this
parent
child
relation
yet-
and
you
have
like
separate
traces
for
each
receive
span,
but
I
was
planning
to
do
that
in
an
xpr.
So
I
had
a
separate
suppression
strategy
for
receiving
process
spans
because
they
are,
they
have
different
semantics
and
make
it
possible
to
process
spam.
Be
a
child
of
the
receive
span.
B
I
see,
but
in
this
one
it
this
one
introduces
the
receive
span.
B
Okay
and
in
this
one,
the
consume,
the
the
processband
still
parents,
the
producer
span.
A
Well,
let's
leave
it
aside
for
a
while
in
case
of
spring
batch,
I
changed
the
structure
a
bit
and
because,
because
there's
a
batch
processing
involved,
so
we
have
only
one
process
span
for
all
records
so
that
batch
processing
consumer
has
links
to
all
producers-
okay
yeah,
which
is,
I
think,
correct
according
to
the
spec
right
now
I
think
yeah,
but
going
back
to
kafka
clients,
because
we
instruments,
you
know,
beat
the
list
they
travel.
I.
A
I
know
I
yeah
I'm
not
so
sure
that
those
spans
should
parent
the
receive
spanner
so
because,
when
you
compare
kafka
clients
to
string
but
sorry
spring
kafka,
batch
receiver-
I
mean
in
the
the
second
one
everything
both
receiving
and
the
process
fund
pretty
much
happen
in
one
class.
So
it's
not
really
possible
for
any
context.
A
Modification
I
mean
and
starting
any
other
span
between
those
two,
it's
not
possible
to
do
that,
but
in
when
you're
using
the
raw
kafka
polling
api,
you
can
pull
in
one
class
and
just
pass
the
records
all
over
your
application
and
iterate
over
them
yeah
somewhere
else
yeah,
and
you
would
if
we
like
forcefully
override
the
parent
of
those
process
spans.
You
might
lose
context
between
those
two.
So.
B
C
B
A
Oh
well,
I
mean
I,
I
think
I
would
like
at
some
point
not
to
instrument
the
kafka
clients,
the
consumer
records
thing,
but
yeah
we
probably
have
to
fully
instrument
like
the
whole
spring
kafka
library,
the
vertex
kafka.
If
then,
there's
a
reactor
kafka
lady,
so
there
are
at
least
three
like
higher
level
kafka
libraries
that
we
have
to
cover
before
that.
C
A
B
B
B
C
C
A
You
because
the
span
is
closed
by
has
next,
I
think
so,
if
you.
A
Next
or
if
you
like,
iterate
in
any
exchange,
oh
the
list
iterator,
for
example,
is
not
supported
correctly
completely.
So
if
you
iterate
with
an
iterator,
it
works.
If
you
work
with,
if
you
iterate
with
a
list
it
read
reader,
it
doesn't.
I
think
the
list
dot
get
methods
are
not
implemented
too.
So
it's
like
it's
leaking.
B
C
C
B
Did
you
make
the
last
release
out
of
here.
C
B
C
I
guess
the
problem
was
just
because
I
was
calling
in
and
I'll
think
yeah.
I
guess
yeah,
it's
compiling
me.
So
there
shouldn't
be
a
problem.
Never
mind
got
a
bit
confused.
There's
this
one
method
to
get
resource
from
the
auto
configure,
which
I
was
using
because
of
the
whole
resource,
is
not
passed
a
sampler
issue
that
is
going
on
forever,
but
yeah
that
method
got
changed.
If
it's,
that
was
the
cause
of
the
breakage,
but
I
was
actually
calling
that
api
internal
details
would
be
fine.
Yeah.
B
Thanks
for
doing
this,
mateish,
yes,
you
saw
I
I
was
there.
I
was
looking
at
it.
There's
some
mother.
B
A
From
on
the
pr
from
the
original
outer
or
the
issue,
and
apparently
the
yet
another
problem
with
remove
last
methods,
but
yeah,
it's
still
not
not.
That
related
I
mean
it
was
it
wasn't
working
correctly
before
and
it
doesn't.
C
A
C
C
C
B
The
the
placement
of
the
channel
handler-
or
maybe
I
missed-
maybe
I
missed
what
you
were
saying.
C
A
B
B
B
C
B
A
I
think
he
is
not
on
vacation
he's
working
well,
wasn't
he
present.
On
the
last
cig
I
mean
I
was
out
because
I
was
on
pt
on
thursday
on
friday,
so
I
wasn't
there,
but
I'm
not
sure
about
laurie.
B
Okay,
yeah!
No!
No!
I
don't
think
he
was
on
the
in
thursday
sig
on,
because
I
I
was
gonna
chat
with
him
about
some
of
the
context.
B
That
doesn't
matter.
I
just
noticed
that
the
instrumentation
repo
had
been
kind
of
quiet,
because
I
was
out
and
you're
out
and
nikita's
out.