►
From YouTube: 2021-10-28 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).
C
Just
before
this,
but
nice,
but
I
have
not
prepared
the
the
document
for
today.
Let
me
prepare
it
answer.
D
C
B
C
C
Nice,
I
don't
know
much
about
lambda,
I
must
say.
C
Okay,
so
do
you
have
any
topic
for
today.
A
I'm
not
okay.
I
was
looking
at
github.
I
was
just
curious
if
that
that
versioning
issue
that
we
were
talking
about
last
week
had
any
traction
on
it.
A
I
brought
that
up
to
the
hotels,
maintainers
meetings
and
people
seem
to
have
a
lot
of
thoughts
on
that,
but
I
was,
I
asked
them
all
to
to
put
it
into
the
the
document
that
we,
okay,
that
we
that
we
or
the
issue
that
we
created,
but
I
think
that
there
also
might
be
a
larger
kind
of
drive
to
to
provide
some
guidelines,
but
at
the
end
of
the
day
I
think
it's
just
going
to
be
sdk
specific.
C
Yeah,
I
think
the
only
the
only
traction
that
it
got
was.
Sir
look.
A
C
Yeah
so
yeah,
we
can
give
it
another
week
to
see
if
anyone
has
any
issue
with
it,
but
I
think
we
will
change.
Also
one
of
the
changes
that
be
not
made
also
implied
a
small
change
that
change
a
bit
the
behavior.
So
maybe
we
should
change
the
minor
version
with
that,
and
also
I
merged
yesterday
appear
that
improved
somehow
the
network
instrumentation
oh
cool,
just
to
simplify
a
bit
the
synchronous
areas
that
should
be
handled
and
also
modified.
C
C
So
he
was
creating
a
network
request
and
when
it
ended,
it
tried
to
create
that
when
the
span
ended,
it
tried
to
create
the
shipkin
trace
immediately
and
the
zipkin
exporter
also
takes
the
network
and
uploads
immediately.
So
that
means
that
the
network
got
stuck
yeah,
because
this
one
was
finishing.
While
this
one
was
trying
to
write,
so
I
fixed
that,
but
also
improved
simple
processor.
C
So
it
now
does
that
work
of
exporting
on
the
background
and
synchronizes
on
the
flash
on
and
on
the
end,
so
the
pr
was
approved
yesterday
by
sergey
that
he's
not
in
the
committee
anymore.
I
don't
know
if
you
have
seen
the
boating
of
the
open
telemetry,
so
sergio
is
not
anymore
in
in
the
committee,
so
I
don't
know
if
they
will
put
any
other
committee
member
in
the
project
or
will
continue
being
the
administrator.
I
don't
know
anything
about
that.
C
He
was
our
our
administrator
of
the
predator
or
the.
I
don't
know
how
what's
the
exact
name
for
that,
but
yeah
so
yeah.
I
did
that
change
so
that
probably
also
changed
somehow
this
simple
processor
that
probably
many
people
is
using,
because
it's
the
only
one
using
the
example
so
really
will
improve,
live
for
a
lot
of
people
so
yeah
so
yeah,
probably
if
they
don't
change
that
we
can
keep
that
for
a
week
and
even
we
don't
have
any
comment
against
it
or
any
other
idea
that
we
can
debate.
C
Probably
we
can
close
next
week
and
continue
with
that.
Okay.
So
for
the
last
meeting
actions,
do
you
have
any
topics?
Sorry
before
going
with?
No
no.
C
Okay,
great,
we
also
have
received
a
new,
a
new
person
committing
changes
to
the
project
that
he
he
has
been
quite
active
recently
with
zipkin
fixes
and
stuff
that
I
have
been
reviewing.
So
I
can
probably
tell
him
to
connect
to
this
meeting
next
week
or
something
like
that.
So.
C
C
Wait
another
week
or
feed
back
and
go
with
it
if
no
more
ideas,
yeah
and
the
environment
variables
name
error,
that's
the
one!
You
change
right,
venus.
C
So
fixed.
C
In
the
code-
and
so
we
are
not
keeping
compatibility
with
existing
names,
I
don't
think
that's
a
big
issue
because
probably
no
one
was
using
this
functionality
for
and
not
at
least
not.
I
was
using
it,
but
I
am
using
the
same
version
for
both
the
target
and
the
creator.
So
so
what
right.
A
I
was
I
was
just.
I
was
just
curious
about
just
how
I
guess
how
useful
this
this
feature
is
cross
process,
the
transfer
with
the
environmental
variable
like
that's
not
going
to
be
usable
in
production.
Is
it
that's
only
going
to
be
something
that
you
could
do
like
locally
through
testing.
A
C
Yeah
that
that
one
yeah
that
definitely
won't
work
on
a
mobile.
You
have
no
way
to
share
information
between
apps,
except
if
the
same
provider,
and
they
have
some
file
format.
A
It
would
have
to
be
like
on
a
shared
bundle,
that's
right
or
something
yeah
yeah.
I
always
I
always
forget.
I
always
forget
that
this
could
be
used
on
a
you
know.
Yeah.
C
C
And
that
you
want
to
run
that
tool
under
your
trace
or
something
like
that,
probably
in
a
mobile
you,
you
cannot
share
that
anyway,
because
you
cannot,
even
with
a
network
request.
You
cannot.
C
I
know
you
can
create
a
local
server
and
you
can
pass
the
context
using
network
request
to
your
local
server.
I
did
something
like
that
in
the
past,
with
a
local
server
and
transmitting
video,
and
things
like
that,
it's
usually
done
in
another
process,
but
that's
quite
ugly,
but
you
could
do
that.
Okay,
but
not
with
environment
variables.
It
will.
You
would
need
to
use
network
request.
A
A
C
So
I
think
there
are
no
more
things
from
last
meeting
yeah
this
one
that
you
fix
the
environment
variable
and
you
change
also
the
transparent
thing.
C
C
B
So
what
about
the
trace
state
and
trace
parent?
I
actually
you
know
I
looked
it
up,
and
so,
where
would
you
actually
use
it
is
that
within
the
I
think,
I
I
looked
for
wherever
the
references
were
in
the
code.
Where
would
you
use
that
that
variable
that
it's
not
an
environment
variable
right?
It's
a
that's,
a
spec,
the
state
and
the
trade
state
in
the
transparent.
C
Yeah,
it
was
in
the
in
the
I,
it
was.
B
In
the
pr
right
wasn't
in
there
yeah
I
I
looked
up
that
link
and
you
know
you
guys
were
discussing
that.
That's
why
I
referred
refer
to
that,
but
so
this
is
something
that
that
is
used
to
when
you're
passing
the
context
right
to
another.
C
B
Yeah
yeah,
so
this
would
be
when
you're,
when
you're
at
the
back,
when
you're
stitching
together
the
two
spans.
C
A
C
D
C
Yeah,
I
am,
I
am
running
posix
spawn
but
running
on
a
simulator.
So,
as
you
are
on
a
you,
are
running
the
simulator
on
the
mac,
you
can
run
any
process
of
your
mac
with
posix
spam
when.
C
D
C
A
C
In
the
in
the
simulator
wow
and
things
like
that,
so
yeah,
some
some
fancy
stuff
and
it's
open
source.
So
oh
okay,.
A
That's
that,
actually,
is
something
that
I
was
curious
about,
like
crash
reporting
in
open
like
like
how
useful
would
that
be
for
the
yes
to
the
the
open
telemetry
product,
there's
a
lot
of
there's
a
lot
of
like
unknowns
in
there.
I
guess,
like
the
main
pro.
C
The
main
problem
with
crash
in
ios
is
that
you
don't
have
the
symbols
file.
C
That's
the
main
problem
you
have
to
make
the
back-end
serve
your
get
your
symbol
files
depending
how
you
destroy
with
your
app.
You
won't
get
the
symbol
files
if
you
are
using
bitcode
and
the
app
store
generates
that
so
you
have
to
actually
go
to
the
store
and
say
give
me
my
symbols,
so
you
can
fit
your
back
end
with
those
symbols
and
symbolicate
the
grass.
A
A
Yes,
yeah
correct
the
so.
C
So
you
can
partially
symbolicate
grasses
just
with
swift,
but
will
only
provide
you
the
name
of
the
of
the
function
yeah.
A
Not
the
yeah,
not
the
line
yeah
so
so
table
that
for
a
second
okay
I
have
not.
Actually,
I
have
not
had
a
deployed
ios
app
in
quite
a
long
time
and
I'm
kind
of
curious,
like
nacho.
Have
you
seen
apple's
ui
for
crashing
at
all,
like
that?
Do
they
have
a
crash
tool
I
feel
like
they
do
they.
A
C
A
C
You
can
you
can
with
xcode,
they
have
the
symbols
because
they
have
access.
Yes,.
C
A
Yeah,
you
have
to
feed
it
to
them
yeah
I
was
just.
I
was
curious
because
I'm
I'm
wondering
how
valuable
it
would
be
to
try
to
do
a
another.
You
know
another
crash
reporting
tool
right
like
if,
if
apple's
already
providing
something,
maybe
they
don't
have
the
the
whole
like
sure
you
can
get
crashes
from
from
production,
but
you
don't
necessarily
get
the
full
context
of
how
many
crashes
you're
getting
or
what
the
distribution
of
crashes
are
or
that's
what,
like
some
of
the
other.
C
A
C
Was
the
one
who
said
that
this
plan
has
did
in
this
in
their
library
program?
So
it
looks
like
everyone
has
a
cross
reporter
now,
yeah.
A
A
Yeah
for
sure,
for
sure
yeah,
that's
that's
yeah,
that's
kind
of
like
how
I
feel
about
it
too,
and
also
like
how
how
would
that
data
fit
into
the
the
the
open
telemetry?
A
Oh
gosh,
what's
the
term
that
they
use
for
the
data
points
like
the
like,
like
how
would
that
fit
into
like
a
metric
or
a
log
or
a
or
a
or
a
trace
like?
I
don't
think
it's
not
like
a
very
it's,
not
a
very
compatible
data
format.
C
Yeah,
that's
right.
I
am
reporting
crashes
because
I
also
have
implemented
a
crash,
not
not
a
crash
reporter,
exactly
because
I
am
symbolicating
myself.
For
example,
yeah.
C
Pl
price
reporter
also
for
recovering
the
crash
logs
and
symbolicating
that
after
but
the
the
the
thing
is
that
yeah
as
it's
for
I
am
working
on
tests
and
that's
not
on
production.
So
you
don't
have
to
have
that
difficulties,
but
yeah.
D
C
C
You
can
probably
also
report
them
each
threat
in
a
in
an
attribute
child
of
I
don't
know
yeah.
I
wonder.
C
And
yes,
and
some
things
but
yeah,
it's
not
very
down
to
that.
The
only
thing
is
that
you,
you
must
save
the
context
of
the
thread
that
crashed
that
it's
something
you
must
set
to
the
crash,
reporter
library
being
ks
crash
or
pl
plus
reporter
you
can
in
both
set
the
set
the
data
of
the
thing.
That's
working,
so
you
can
recover
that
if
it
crashes,
so
you
can
set
them
the
span
id
or
the
trace
id
in
order
to.
C
C
A
C
But
didn't
yeah,
then
I
I
think
that
ks
crash
a
developer
was
working
for
another
observability
company
and
implemented
the
same
ks
crash
library
inside
their
observability
library.
I
don't
remember
exactly
which
one.
A
C
In
the
past
so
yeah
I
don't
know.
A
Yeah,
I
mean
that's
what
that's
that's.
Why
I
ask
is
we're
looking
at
adding
a
crash
reporter
as
well
and
yeah?
I
was
just
planning
on
using
the
pl
crash
report
now
that
dot
or
now
that
microsoft
has
been
maintaining
it
because
yeah,
because
I
was
using
that
at
new
relic
and
but
I
had
like
you
know
it
hadn't
been
updated
since,
like
2008.
B
C
C
B
A
It
was
kind
of
abandonware
for
a
while,
but
but
yeah.
I
was
just
wondering
how
useful
that
would
be
to
try
to
make
it
like
it.
For.
C
C
They
have
something
with
xct
issues
or
something
like
that.
Some
strange
features
there,
but
not
documented
and
yeah,
but
they
I
don't
they.
They
have
their
own
crash
because
they
capture
the
crashes
and
report
them,
but
yeah
everyone
is
using
bo
or
ks
crash
or
pl
crash
reporter
now,
yeah.
C
Yeah
ks
crash
is
not
maintained.
Officially,
okay,
so
he
only
accepts
pr
from
people,
he
validates
them,
and
so
it's
more
or
less
a
community
driven
project.
C
C
Be
partially
symbolicated,
but
you
cannot
get
some
information
and
ks
crash
is
much
more
structured
that
because
it
provides
you
the
dictionaries
with
the
data
and
it's
and
it's
quite
nice
library,
but
yeah.
I
use
it
in
the
past
before
being
acquired
by
datadog
in
my
company.
We
used
that,
but
we
changed
to
build
crash
reporter
because
we're
the
one
used
yeah
yeah,
for
example,
I'm
excited
yeah
yeah
ks
crash-
is
a
really
nice
library
also
and
it
works.
C
Maybe
it's
not
not
having
an
official
support.
Yeah
yeah
can
be
not
a
good
option,
maybe
because
you
don't
know
when
things
will
be
fixed
but
yeah
both
both
works
quite
well.
C
In
fact,
the
idea
of
setting
the
information
in
the
grass
that
I
implemented
in
pl
crash
reporter
was
based
on
the
functionality
that
that
I
didn't
so
right
now
that
was
the.
That
was
the
reason
I
implemented
that
because
I
really
needed
to
identify
why
this
panda
crust.
So
you
can
follow
the
trace,
or
at
least
reconstruct
the
span
after
the
grass,
with
the
grass
information,
so
yeah,
it's
a
it's
a
it
could
be
created
inside
a
an
instrumentation
itself.
A
A
A
Yeah
for
sure
yeah,
that
was
the
other
thing
is,
I
wonder,
how
they're
handling
it
in
the
other
sdks
like,
because
you
know
you
might
have
a
an
exception
in
like
java
or
something,
and
you
want
to
pass
that
along.
I
guess
you
just
throw
that
on
as
a
part
of
a
trace
or
something.
C
B
C
B
You
know
if
they
have.
You
know
if
the
web,
that
side
of
things,
if
it
crashes
right,
they
have,
I
think,
har
right
some
archives
that
you
can
upload.
C
A
C
A
Yeah,
I
think
I
think
that
is
in
there
in
their
roadmap.
Oh
okay
or.
C
I
also
think
that
yeah
it
makes
sense
to
have
that
information
inside
server
code.
You
have
exceptions
also,
there
we
go.
A
C
It
supports
no
it.
It
supports
the
concurrency
library,
oh
cool
yeah,
from
ios
13,
so
they
are
giving
support
to
ios,
13,
14
and
15..
Oh
wow.
I
think
a
weight
and
everything
wow.
C
A
C
But
that
means
that
we
can
have-
probably
some
users
of
the
library
will
will
want
to
use
a
single
weight
method
so
yeah
we
are
supporting
that
already
awesome.
A
C
Yeah
so
yeah,
that's
a
really
nice
one.
We
will
have
to
check
if
everything
works
as
expected.
Yeah.
A
We
should
we
should
do
like
a
press
release
on
that
or
something
I
don't
know.