►
From YouTube: 2021-11-04 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
Good
morning
I
only
have
15
minutes
today
I
have
another
meeting,
that's
conflicting
at
9,
15
or
in
15
minutes.
So
you
don't.
I
don't
know
what
time
I
always
forget.
B
Yeah,
I
think
we,
I
don't
have
many
topics
so
yeah,
let's
see
once
be
not
connect.
I
think
we
can
start.
B
Not
bad,
I
had
fun
holidays
on
monday,
so
honest,
it
was
a
shorter
one,
so
yeah
and
yeah
fixing
some
issues
in
the
project
this
week,
mainly
and
yeah.
I
found
a
big
one
without
your
instrumentation:
oh
no
yeah,
something
that
yeah
one
of
the
users
reported
in
the
forum
or
sorry
indeed
have
actions
and
and
yeah,
and
it
was
real.
A
B
We
were
filing
with
synchronous
network
requests,
so
we
could
only
handle
one
and,
if
any
other,
that
could
come
concurrently,
yes,
were
lost
and
we
didn't
instrument
them.
B
No,
no,
it
was
with
any
oh,
no
any
network
request
yeah
we
have
been
with
that
like
for,
since
the
library
landed,
and
it
was
just
because
one
line
was
out
of
the
place
it
should
be
so
it
was
getting
the
the
task
session.
It
was
getting
set
in
a
uid
for
the
session
each
time
it
connected,
but
instead
of
doing
it
for
each
connection
in
the
switzerland
thought
it
was
doing
outside
so
there's
the
first.
It
had
the
same
one.
So
it
didn't.
B
It
only
basically
reported
once
so
that
that
was
a
serious
one,
but
it's
fixed
and
already
merged.
So
I
think-
and
that
was
my
plan-
it
looks
like
we
we
can.
Let
me
share
the
comments,
so
we
can
follow
up,
as
you
only
have
15
minutes.
I
don't
know
sure.
B
Can
you
see
that
yeah?
I
think
I'm
saying
my
full
screen
today,
but
okay,
so
from
last
meeting
I
don't
know.
Maybe
key
point
is
the
compatibility
plan.
C
B
Hi
so
yeah
we
started
a
bit
earlier
because
bryce
has
a
meeting
in
10
minutes
or
so
sure
so
yeah.
Basically
we
have
talked
about
the
compatibility
plan
and
we
talked
about
the
minor
version
when
we
have
to
change
the
tools
and
nobody
said
nothing
against
it.
So
we
go
with
that
plan.
I
will
put
it
in
the
issue
right.
B
And
also
we
will
talk
about
the
environment
variables
names
that
mean
that
next
version
will
need
a
minor
number
update
that
we
talked
last
week.
So
my
that
will
change
some
things.
So
that
means
that
I
will
have.
I
will
have
a
new
number
to
change.
I
also
created
a
pr
to
yes
that
we
are
going
to
modify
it.
I
created
some
prs
that
can
come
with
this
minor
version.
Number
one
of
them
is
that
we
are
cleaning.
Definitely
the
products
that
started
with
the
leap
name
that
we
wish.
B
We
used
at
the
beginning
with
the
static
libraries
and
we
modified.
So
we
had
this
leap,
leave
open,
telemetry,
sdk
and
open
telemetry
sdk
in
the
package
yeah.
I
have
it
here.
Basically,
both
were
the
same,
and
now
that
we
are
changing
the
number
we
can
clean
up
clean
this
up.
This
is
mainly
the
change
removing
them.
C
B
B
To
be
dynamic,
yes
used
to
be
dynamic,
we
changed
that
like
three
or
four
versions
ago,
because
pipe
people
didn't
understand
the
difference,
it's
really
very
messy
with
spm,
but
so
now
that
both
were
the
same,
we
just
replace
the
names
and
use
a
static
for
everyone,
and
we
don't
have
duplicated
lines
here.
That
is
it's
really
confusing
when
you
are
having
the
project.
So
as
we
were
updating
the
miner
version,
I
have
also
created
that
pr
is,
is
not
approved
still,
but
yeah.
B
It's
just
that
change
another
pr
that
is
interesting
is
that
we
were
failing
with
the
network
instrumentation
with
url
session.
B
I
fixed
that
and
it's
merged
yeah.
This
is
the
issue
is
that
concurrent
network
requests
were
failing
to
be
reported
and
only
reported
the
first
one,
and
the
change
was
really
small
because
we
were
getting
the
task
id
from
the
session
outside
of
the
of
the
of
the
method
implementation.
So
we
were
getting
the
same
for
all
the
session,
so
only
one
could
happen,
and
now
we
are
just
moving
this
inside
the
method
works.
As
expected.
I
think
it
was
a
problem
with
my
rewriting
of
this
from
the
previous
version.
B
Parent,
only
the
parent
was
showing
up,
not
the
parent,
only
the
first
one
I
mean,
if
you
ask,
for
three
things
in
the
network
concurrently
before
waiting
the
first
to
answer
those
wouldn't
appear
until
the
first
one,
because
the
session
was
the
same.
So
it
was
thinking
that
the
session
was
already
used
and
didn't
update
anything
so
yeah
just
it
was
yeah.
It
was
a
really
stupid
pack,
but
it
was
working
bad.
B
B
B
One
is
implement
the
first
flash
spec
keeping
compatibility,
so
we
are
keeping
the
compatibility
with
the
timeout
and
implemented
on
the
batch
span.
Processor,
not
on
the
simple
spam
processor,
because
that's
the
simple
one
I
just
added
documentation.
So
people
know
that
should
use
the
other
more
or
less
yeah.
That
is
mainly
having
a
timeout
interval
optional.
B
We
have
the
default
previous
version
here,
without
anything,
so
should
work
with
for
current
users,
and
the
only
other
change
is
that
when
we
are
on
the
batch
spam
processor,
we
are
taking
it
in
account
and
on
the
simple
spam
processor
we
just
say
that
for
production,
environment
use
batch
spam,
processor
and
say
that
is
in
using
this
processor,
the
timeout
it
doesn't
make
sense.
I
mean
this.
B
This
shouldn't
be
the
one
used
on
production,
it's
more
for
testing
and
for
just
we
use
that
for
our
test,
mainly,
but
can
be
used
for
other
things,
also,
but
yeah,
so
that
one
and
the
other
interesting
is
yeah.
Also,
I
I
have
yes
that,
as
we
have
to
change
the
number
also
updated
the
semantic
attributes
and
resource
attributes
to
the
latest
version,
so
that
should
at
least
the
attributes
be
updated
to
the
latest
version
yeah.
B
Also
one
of
the
things
I
added
something
here
that
was
not
exactly
in
the
spec,
but
I
updated
the
spec
so
the
that
we
report
also
the
language
suite
as
a
new
as
a
telemetry
library,
so
that
that's
all
I
have
to
say
so
with
these
changes
when
these
peers
are
approved,
my
idea
is
releasing
one
one,
oh
as
soon
as
possible,
because
I
think
this
network
instrumentation
but
is
really
bad.
B
It
has
been
for
a
long
time,
but
it's
really
a
big
issue
so
yeah
we,
the
idea,
is
releasing
as
soon
as
possible
and
changing
the
number
and
just
do
some
cleanup.
As
I
said
so,
that's
my
plan.
A
B
B
Yeah
I
will
create,
then
I
release
tomorrow
in
my
morning,
so
we
can
just
have
a
version
with
that
and
and
people
can
use
yeah
the
network
instrumentation
issues
was
horrible.
I
thought
how
we
couldn't
realize.
Before
I
don't
know.
I
have
been
using
it
for
many
for
a
long
time
now
and
was
not
able
to
find
it
yeah.
A
B
I
already
improved
the
networking
stuff,
but
I
already
told
last
week
so
I
reduced
the
synchronous
areas
of
the
library,
but
this
was
not
related
to
that.
That
is
that
probably
it
was
most.
B
I
don't
know
why
we
couldn't
see
before
really
really
yeah
but
yeah.
I
think
as
soon
as
we
can
provide
it,
the
better
so
yeah
that
from
my
site,
all
right
anymore,
I
think
yeah.
B
C
I
don't
think
yeah.
I
didn't
test
multiple
urls
myself,
so,
okay
yeah,
that
is.
B
Was
I
can
show
you
I
had
a,
I
had
a
project
here
that
I
wrote
as
one
of
the
users
reported,
so
mainly
the
thing
was
that.
B
B
B
It's
just
okay:
this
is
not
gonna
work
now,
so
just
a
standard
network
call.
So
using
this.
You
know
this
web
page
that
just
returns
things
and
and
works
as
as
a
I
use
it
a
lot
it,
for
example,
is
usually
you
want
to
know
what
you
send
to
the
library
and
that
stuff.
So
I
use
it
in
my
test
and
also
in
the
library.
B
I
don't
know
if
you
know
what
I
so,
it
is
like
a
service
that
you
can
send
methods
and
it
will
just
provide
you
some
information
like
get
what
you
send,
for
example,
if
you
call
it
like
this,
so
you
can
so
it's
a
url
basic
url
that
just
do
things
so
I
was
creating
different
network
calls
and
printing
the
the
the
information
that
it
was
sending
so
created
the
batch
spam
processor
or
the
simple
it's
the
same
and
using
the
standards
out
exporter.
So
it
prints
the
expanse
information
in
the
standard
output.
B
Just
before
ending
one,
it's
already
running
okay-
and
we
can
also
make
here,
for
example,
make
the
request
slip
for
one
second,
for
example.
So
you
are
sure
that
all
the
networks
happen.
B
So
the
thing
is
that
when
you
run
it
yeah,
it
prints
here
just
to
know
that
all
of
them
just
worked.
Okay,.
C
B
A
B
That
always
works
and
just
a
very
useful.
You
know
my
peers
are
being
approved
now
by
sergey
so
and
just
sleeping
here,
so
it
takes
more.
B
Yeah
he
he
was,
he
was
in
the
the
committee
in
the
open,
telemetry
technical
committee,
so
the
effects
of
the
project
that
are
half
of
them
are
changed
each
year
in
the
last
elections.
He
he's
not
in
the
committee
anymore
right,
but
he's
still
there
I
mean
when
I
when
I
provided
a
project
to
open
telemetry,
because
I
created
it.
In
my
last
my
previous
company,
I
wrote
the
code
I
contacted
with
open
telemetry
project
and
I
said
I.
A
B
To
donate
the
original
code,
so
they
assigned
me
sergey
as
the
administrator
of
the
project
and
me
being
the
maintainer,
so
he
he
has
to
approve
my
my
peers,
for
example,
or
any
other
one.
So
at
the
beginning
he
was
the
only
who
could
approve
my
peers,
because
no
one
more
had
a
committed
access
to
the
project.
Now
bryce
also
is
a
committer,
so
he
has
rights
to
approve.
Also
my
peers
and
I
have
the
rights
to
approve
him.
B
B
He
has
never
been
in
any
match
meeting.
I
don't
know
why
he's
never.
I
have
never
seen
him
in
person,
I
think
either
we
have
been
he's
approving
my
you
can
see
here
that
he's
approving
my
yeah
right.
B
So
basically,
here
talking
about
this
is
five
network
requests.
Each
of
them
takes
more
than
than
a
second,
so
all
of
them
are
happening
at
the
same
time,
at
least
the
request
and
waiting
for
the
responses
synchronously.
B
B
C
So
so
now
so
you're
making
those
yeah
those
network
calls
from
that
asynchronous
block
of
code
yeah.
This
is
a
thunder.
B
B
And
so
here
we
have
mixed
information,
we
have
the
responses
and
we
have
the
expanse.
So
we
have
this.
This
is
the
response.
There
is
a
json
with
the
information.
This
is
the
first
span
printed.
B
It
was
failing
before
with
the
previous
code.
Only
this
one
will
have
pulled
up
here.
B
Yeah,
it
was
not
a
for
flash
thing
because
it
will,
with
10
seconds
it
should
send
all
the
spans
with
the
batch
spam
processor.
I
think
it,
but
just
to
be
sure
that
it
processes
all
of
them
and
export.
C
But
wouldn't
you
expect
it
to
have
five
different
spans
with
the
same
span
id,
because
that
is
what
you
had
set
right
to
there.
I
have.
I
have
different
spanish
right
right.
No,
but
before
you
had
like
you
know
the
spider
you
were
set
as
a
from
the
uuid
right.
B
Like
yeah,
it
was
getting.
No,
the
thing
was
failing
on
the
ins
inside
the
library.
It
was
yeah,
it
was
yeah,
it
is
a
local
pack.
It's
great.
It
was
failing
here.
C
B
Task
id,
instead
of
being
inside
the
block
it
was
outside.
B
B
B
B
B
A
B
Span
ended
up
the
first
one
that
log
the
response
just
removed
all
the
values
so
that
that
was
the
issue.
Now
it
works,
but
it
didn't
before.
C
C
So
if
anyone
was
making
a
synchronous
calls
multiple
url
calls,
they
would
have
hit
this.
C
Yeah,
this
is
a
I
mean
unless
you
use
it
with
multiple
yeah.
B
Some
it
was
reported
on
the
in
the
project.
I
will
just
reset
this.
B
Yeah,
it
was
reported
on
the
project
here
yeah
because
we
went.
We
had
one
user
of
the
kin
trace
exporter
that
he
was
using
quite
a
lot,
and
he
just
reported
here
that
this
wouldn't
work.
B
So
yeah,
I
guess
that's
if
he
was
waiting
for
the
batch
spam
processor
to
finish,
because
I
couldn't
believe
it
was
failing
and
we
hadn't
noticed
before
so
yeah,
so
that
was
nice.
B
B
A
B
C
With
yeah.
B
This
is
what
it's
done
in
the
sample:
yeah
yeah
yeah.
That
means
that
all
your
network
requests
will
be
instrumented,
but
if
you
are
sending
tipkin
expanse,
it
uses
the
same
url
session
thing
for
sending,
so
you
have
to
configure
it
using.
B
B
Then
yeah,
the
instrumentation
configuration
has
a
lot
of
callbacks
here
that
you
can
implement
for
extending
the
the
functionality,
for
example,
if
you
use
this
and
return
yes,
it
will
also
report
the
payload
in
inside
the
the
span.
It
captures
the
that
in
a
buffer
and
reports
that
this
one,
for
example,
says
if
you
want
your
you
with
these
requests,
if
you
want
to
instrument
that
request
or
avoid
that
and
is
the
one
using
here.
B
If
the
url
is
this
one
just
on
the
instrument,
because
it's
your
span,
your
your
15
response,
your
sending
is
here
the
other
things
here,
for
example,
changing
the
name.
B
B
C
But
before
this,
your
instrumentation
of
the
url,
you
already
configured
the
zipkin
exporter
right,
yeah,
you
you.
B
B
C
C
Set
up
the
zipkin
exporter
right,
you
don't
have
to.
I
think
I
should
look
at
that
example
that
you
have
do
you
have
to.
B
You
have
options
for
configuring,
but
the
option
is
mainly
the
endpoint
by
default.
Is
your
local
endpoint?
B
C
B
C
So
you
can
so
what
is
the
yeah
I'm
a
little
confused,
so
the
span
processor,
so
you
can
create
spans
outside
of
the
url
instrumentation,
but
the
url
instrumentation
can.
So
you
can
go
to
two
different
places.
That
is
essentially
what
this
means.
This
you
know,
the
spans
outside
of
the
url
session
can
go
to
one
endpoint
and
the
url
instrumentation
can
go
to
another
endpoint.
B
Yeah,
it's
no,
you
mean
sending
different.
C
B
C
Do
you
mean
yeah
yeah,
but
I
I
don't
understand.
Why
do
you
need
two
different
endpoints?
Because
you
know
you
would
no.
I
have
only
one
yeah
so
so
so
my
my
question
is
you
specify
the
span
exporter
with
a
zipkin
and
endpoint
for
the
spam
yeah
and
and
then
you
start
a
url
instrumentation
right,
yeah,
the
network
instrumentation.
You
know
where
you
create
the
trapper
so
whatever
span
that
is
generated
before
you
know
so,
let's
say
you
created
some
other
spans
right
before
you
get
to
the
url
instrumentation.
B
The
parent
yeah
yeah,
if
you
yeah,
I
mean
it-
will
take
the
active
span
as
the
parent,
so
yeah
exactly.
C
B
C
B
C
Then
you
know
the
specifying
the
url
here.
What
does
it
do?
Because
you
already
specified
the
zip
code
span,
endpoint
yeah.
C
B
Okay,
let
me
if
you,
if
you
use
the
default
instrumentation
this
one
okay,
this
one
will
be
default,
so
it
captures
everything.
B
You
will
see
that
what
we
have
now.
The
thing
is
that
the
tipkin
exporter
will
send.
A
B
C
C
So
what
was
the
closed
exception
available.
C
C
C
Now,
actually,
in
your
example
right,
you
have,
you
can
just
cut
and
paste
that
zipkin
example
and
that
should
do
it.
C
So,
just
like
you
know,
do
like
a
docker
ps,
okay,
yeah
go
to
command
line.
C
Because
I
I'm
using
oh
yeah,
I'm
using
zipkin2
I'm
using
zipkin,
but
I'm
using
a
docker
compose,
so
that
makes
it
easier
to,
but
I
think
you
also
have
don't
you
have
one
of
the
or
you
have
the
prometheus
example
that
is
metrics.
It's
not
that.
C
C
One
yeah:
how
can
I
run
this
I'll,
just
say:
docker
compose
minus
docker
compose
up
up
and
down
okay,
so
I
must
go
to.
B
B
C
B
I
mean
I
don't
know
if
he's
timing
out
the
batches
pump.
Professor
has
a
longer
time
so.
B
B
C
B
Yeah,
I
know
why
we
are
not.
We
are
not
seeing
it,
but
it's
happening
what
I
am
stupid.
Sorry
it
was
working
as
respect,
but
what
I
did
was
that
I
we
don't
have
the
standard
output
here.
Oh
you
only
have
spam
yeah
yeah.
C
B
C
B
B
B
I
don't
know:
okay,
I
don't
know
how
it
it
doesn't
get.
C
Moving
yeah
yeah,
but
the
previous
one
you
had
it's
typically
get
and
post
okay.
So
if
you
click
on
the
show,
I
think
sure
will
give
you
the
go
back
to
the
zipkin
yeah
click
on
this
show
click
on
the
show,
and
that
will
give
you
oh
it's
not,
and
if
you
click
on
the
that
greater
than
oh.
C
A
little
bit,
but
the
parent
span
should
actually
close.
After
all,
the
other
requests
have
come
back
right.
B
Yeah,
no,
the
the
child
expands
can
be
longer
than
the
the
yeah
they
could
be
longer
because
they
can
be
asynchronous,
and
here
you
see.
A
B
Yeah,
we
are
our
sending
of
the
thief.
Information
is
being
captured
also,
so
that
means
that
that
that
that
was
the
reason
that.
C
You
have
to
so
the
one
that
we
are.
You
know
capturing
the
call
to
zipkin,
that
is
by
itself
right,
yeah.
B
B
B
C
B
Because
it's
been
avoided
here
method,
so
the
instrument.
C
So
that's
that's
basically,
so
it
would
return
true
if
it
is
no
yeah.
B
For
example,
if
you
are
using
another
url
that
you
don't
want
to
report
like
sending
your
metrics
or
things
like
that,
you
just
avoid
the
url
using
this
callback.
So
it's
like.
C
B
Or
any
condition
I
mean
you
don't
want
to
instrument
if
it's
from
10
a.m
to
12
of
night.
So,
for
example,
it's
any
condition
here,
but
just
comparing
the
url,
because
you
pass
the
request.
You
can
also
compare
any
other
if
a
scheme,
whatever.
C
Yeah
yeah,
I
think
we
we
should.
There
will
be
another
meeting
right
in
this
yeah
yeah
we'll
talk
next
week,
then,
okay,
all
right
yeah,
thanks
inside
all,
right
bye,
take
care,
bye.