►
From YouTube: 2022-03-09 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
C
B
A
B
C
D
Is
that
resolved,
I
think
rover
phone
solution.
It
was
resolved
just
okay.
C
Great
so
and
the
test
failure
is
unrelated
right.
D
C
D
On
the
test,
yeah
and
fixing
the
bug
that
didn't
account,
plus
cancellation.
C
Okay,
so
I
would
take
a
quick
look
but
that
there
is
approvals
already
so
after
the
other
one.
I
will
give
priority
to
the
other
one
that
I
just
started
tests.
C
C
D
C
B
C
Oh
no.
This
is
a
follow-up
to
the
the
code
that
chris
did
basically
is
about
kind
of
just
recording
the
exception
on
the
top
current
activity
and
then
just
dispose
the
order.
C
The
current
code
doesn't
have
anything
like
this,
so
just
adding
this
as
a
separate
stream,
so
it
seems
reasonable
just
that
noah
pointed
out
that
it's
the
choice
of
having
this
handler,
not
this
code
specifically
but
just
having
the
handler
is
kind
of
you-
are
making
a
choice
on
trying
to
submit
the
data
and
be
happy
risking
the
time
for
shutdown
and
perhaps
even
the
error,
if
you
error
before
they
think
the
application
may
show
some
all
the
information
as
there
that
cause
determination.
C
But,
as
you
said,
the
risk
of
that
happening
is
much
smaller.
So
it's
a
trade-off.
There
is
a
risk
there,
but
I
think
we
are
aware
of
that
and.
E
For
now
one
question
in
that
area,
I
asked
noah
to
review
that
pr,
because
I
also
felt
the
same.
That's
why
I
wanted
to
get
his
like
opinion
on
that.
So
should
we
have
a
flag
or
something
like
another
environment
variable
where
we
could
give
the
control
to
the
user
there,
because
there
are
chances,
because
I
have
seen
in
places
where
it
instead
of
crashing
it
hangs
and
people
don't
know
why
it
is
happening
actually.
So.
E
B
E
E
C
Yeah,
I
I
think,
because
the
risk
of
this
bringing
the
determination
problem
first
guess
right,
but
my
my
first
guess
is
that
by
default
we
should
enable,
but
then,
if
people
start
to
see
the
application
hanging
on
termination
or
something
then
is
the
time
that
they
have
the
environment
variable
to
change
the
default.
You
know
because
I
think
the
risk
is
relatively
low,
but
is
a
real
risk.
C
C
That
on
the
issue,
just
to
be
sure
that
people
agree-
or
perhaps
people
have
a
better
argument
to
a
different
default.
You
know
this
is
just
what
comes
to
my
mind
at
first
consideration.
C
Piatri,
this
is,
is
very
good.
I
I
just
thinking
about
because
it's
a
long
time
we
may
have
to
be
very
careful
if
the
cooled
up
stream
changes
a
lot.
I
don't
think
it's
a
simple
matter
of
just
pulling,
so
this
needs
to
be
done
very
carefully.
I
think.
A
E
A
C
We
should
be
doing
that
you
know,
but
I
I'm
just
calling
attention
that
we
should
be
very
careful
with
the
code,
because
the
change
that
and
the
long
time
that
we
don't
pull
from
upstream,
that
there
may
be
some
other
kind
of
not
readily
visible
stuff,
that
we
need
to
take
care.
B
Yeah,
I
think
in
general,
the
apis
are
additive,
so
I
don't
think
there
anything
really
was
removed,
but
at
least
there
were
two
recent
fixes.
I
think
you
pointed
out.
Those
should
get
ported,
no
matter
what.
C
Yeah
yeah
yeah
cool,
so
just
doing
carefully.
One
thing
to
think
here:
do
we
need
those
for
beta
or
not.
C
C
Yeah,
so
for
this
in
progress,
chris
told
me
that
it's
not
done
but
he's
gonna
try
to
follow
up
and
and
finish
this
and
by
the
way,
if
you
guys
need
something
for
chris.
He
just
said
that
he
can
join
the
meeting,
but
you'll
be
able
to
do
some
work
and
such
stuff
so
he's
around
if
needed.
E
I
have
a
question
based
on
this
dashboard.
What
is
it
like?
Tentative
time?
Are
we
planning
for
our
beta
release.
A
E
Okay,
and
also,
if
you
look
at
it
the
currently,
we
don't
have
anything
for
log
or
metric
and
open
telemetry.
Sdk
has
advanced
in
those
areas
and,
if
I
remember
correctly,
open
telemetry
is
just
in
rc,
because
the
metric
spec
is
not
like
a
stability.
So
what
are
our
plans
like?
When?
Are
we
going
to
start
the
the
work
on
those?
Should
we
plan
for
that
in
the
coming
weeks,.
C
I
think
I
think
we
we
should
focus
on
the
release
of
the
beta
with
tracing
but,
as
you
said,
because
if
we
are
going
to
match
the
versions,
this
get
me
to
other
point
that
I
wanted
to
talk
about.
That
is
versioning
numbers
that
we're
going
to
use
and
we
have
to
follow
something
from
open
telemetry.
C
So
in
that
sense
we
should
kind
of
be
matching
the
sdk
it's
not
necessarily,
but
I
think
you
make
things
much
simpler
for
the
users,
but
then
we
get
to
that
strange
point
where
okay,
we
are
in
1.2,
but
let's
say
what
we
support
for
metrics.
What
we
support
for
lag
for
logs
from
the
point
of
view
of
the
auto
instrumentation,
you
know,
I
think,
in
the
same
way
that
we
set
up
the
sdk
for
tracing.
C
C
The
perhaps
some
library
that
you
are
using
is
already
instrumented
with
some
metrics
and
we
want
to
be
able
to
set
up
and
collect
those
I'm
trying
to
be
kind
of
not
ambitious
here
in
the
sense
that
perhaps
in
the
future,
we
could
consider
leveraging
the
the
instrumentation
code
itself,
the
bytecode
instrumentation
itself
to
do
things
like.
Oh
I
wanna
time
such
functions.
You
know,
I
I
don't
think
all
because
some
small
functions,
probably
a
lot
of
people,
don't
want
to
time
those.
C
C
Enable
with
tracing
and
then
we
do
a
follow-up
to
enable
metrics
logs.
I
I
kind
of
I
think
it
of
course
is
important,
but
I
kind
of
give
less
priority
than
metric
is
in
my
mind.
You
know,
because
logs
typically
people
have
already
some
solution
in
place:
the
collection
of
runtime
methods,
that's
something
that
will
be
new
for
a
lot
of
users.
You
know.
E
So,
are
we
going
to
have
a
if
the
plan
is
to
have
a
different
branch
to
track
the
metric
effort,
or
are
we
going
to
do
it
on
the
main
branch,
along
with
the
traces.
B
C
But
I
think
it
keeps
the
work
simple
and
we
can
have
kind
of
use
the
configuration
options
to
kind
of
okay,
it's
disabled
by
default.
You
know,
I
think,
makes
our
work
much
simpler.
C
Yeah,
I
think
I
think
we
need
to
have
a
a
discussion
about
what
we
want
to
do
there.
You
know.
C
C
Yeah,
but,
but
I
I
think
like
this,
on
the
other
hand,
that's
why
I
think
we
need
to
have
a
a
a
real
conversation
about
this
and
involving
the
sdk
too.
But
if
you
look
to
the
future
of
this
thing,
a
lot
of
the
things
are
moving
to
that
interface,
the
I
logger
you
know.
C
So,
do
you
want
to
kind
of
really
get
out
of
the
way
kind
of
of
the
let's
say,
more
official
path
to
kind
of
not
official,
but
the
traditional
and
the
kind
of
recommended
path
to
go
to
support
some
kind
of
log
library
that
doesn't
support
that
you
know
I'm
trying
to
reduce
the
work
for
as
you
open
telemetry.
C
C
But
but
I
I
think
it's
a
a
discussion
that
we
need
to
have
you
know
kind
of
to
to
make
that
explicit
and
kind
of
okay.
We
can
start
with
a
logger,
but
then,
if
there
is
really
the
demand
and
people
really
want,
we
can
look
at
supporting
something
out
of
file
order,
but
once
more
I
logger
is
is
the
present
and
the
future.
I
think
anything
else
is
kind
of
legacy.
C
Yeah
and
and
also
just
to
to
be
sure,
I'm
just
kind
of
giving
my
rough
thoughts
here.
We,
of
course
we
have
discussed
this
and,
of
course
I
can
change
my
position
and
let's
see
what
we
agree
as
a
as
a
group.
You
know
work
on
this
this
stuff,
so
just
to
recap
and
be
sure
that
what
I'm
saying
here
is
is
just
kind
of
the
rough
line
yeah.
Let's
when
we
get
to
log,
let's
start
with
phi
logger.
C
If
we
feel
the
need
of
the
demand,
then
we
consider
perhaps
some
extension
points
to
work
on
that,
or
perhaps
we
do
adapters
to
other
loggers,
so
they
have
the
I
logger
interface.
You
know
so
not
sure.
C
So,
just
briefly,
to
recap
the
thing
about
version.
C
We
can't
because
we
need
to
have
the
position
that
I'm
I'm
kind
of
linear
I'm
trying.
I
think
we
should
try
to
match
major
and
minor
versions
with
the
sdk,
but
realistically
speaking,
we
are
not
going
to
be
able
to
match
patches.
You
know,
because
we
will
have
bugs
to
fix
that.
Perhaps
they
the
sdk,
doesn't
have
anything
to
fix
and
we
need
to
kind
of
have
separate
release
which
is
kind
of
weird,
but
that's
at
least.
What's
coming
first,
to
my
mind
you
know
we
match
major
minor
and
patch.
A
A
C
I
I
I'm
thinking
more
just
about
the
sdk
because
of
this
problem
that
I
mentioned
because
the
other
packages
they
don't
need
to
follow
that.
So
I
export
I
can
do
a
major
upgrade
while
the
sdks
is
still
the
same
version,
but
because
we
do
inject
a
version
of
the
sdk
you
know.
So
I'm
thinking
that
perhaps
we
should
try
to
match
the
sdk
only
the
sdk.
We
can't
control.
As
you
said,
the
example
of
export,
I
think,
is
the
best
one.
Somebody
can
create
a
2.0
of
exported.
C
The
open,
telemetry
spec
supports
that,
and
then
we
are
going
to
have
perhaps
a
minor
patch
to
incorporate
that
export,
or
something
like
that.
You
know
actually
that
that
is
not
a
good
example
of
something
for
minor
patch.
But
let's
say
something
like
this:
that
we
move
some
feature
or
some
fix
that
we
want
to
add.
We
are
gonna
have
to
to
move
the
minor
patch,
so
major
miner,
you
match
the
sdk
only
that
we
are
using,
but
not
everything
that
we
bring
along.
C
You
know
I
I
I
think
that
is
kind
of
it's
not
ideal,
but
I
think
it's
kind
of
the
realist
thing
that
we
can
hope
to
do.
C
So,
for
instance,
right
now
the
sdk
you'll
be
something
like
1.2,
some
patch
number
rc.
C
If
we
release
we
probably
using
1.2,
you
should
release
1.2
something
you
know
I
I
think
doing
a
version
completely
different.
We
can
do
that
also,
but
it
doesn't
match
the.
I
think.
Actually
we
can't
do
that
because
it
doesn't
match
the
open,
telemetry
recommendation
spec
for
vision.
You
know
we
need
to
match
the
spec
that
we
are
using.
A
C
A
C
That's
true:
we
are
not
going
to
be
a
hundred
percent
following
the
semantic
version,
as
people
expect
that,
on
the
other
hand,
the
sdk
follows
and
people
you'll
know
the
version
of
the
sdk
and
the
spec
that
we
are
using.
But
you
are
right.
You
are
not
going
to
be
following
the
semantic
version
too,
but
my
once
more
same,
as
I
said
before,
regarding
logs,
this
is
kind
of
my
first
impression.
C
My
first
take
on
the
problem
would
like
to
hear
what
people
think
about,
and
perhaps
they
have
different
suggestions.
You
know,
but
I
think
putting
together
the
sdk,
the
open,
telemetry
spec
for
versioning.
That's
the
place
that
we
end
up.
That's
my
my
current
thought.
You
know
kind
of
I
can't
I
can.
C
C
B
C
C
Okay,
this
is
what
I
had
in
mind.
Do
you
guys
want
to
bring
something
else
up.
C
I
saw
there
thank
for
opening
that
this
reminds
me
something
raj.
I
remember
that
you,
I
don't
remember
what
was
the
exact
feature,
but
I
remember
that
you
wanted
to
ask
for
some
feature
for
net
seven
right.
I
I
I
don't
remember
if
it
was
upgrade
or
downgrade
of
package
or
something
like
that.
C
Is
that
emotional
or
do
we
need
to
kind
of
look
into
that.
E
I
could
not
get
the
right
guys
after
zara
bleh
to
reach
them
out,
because
there
are
different
teams
here.
I
don't
know
who
is
the
right
one
to
reach
for
that?
It
is
about
the
same
as
of
now.
We
have
an
additional
dips
which
will
work
perfectly
for
framework
dependent
applications,
but
there
are
cases
where
this
additional
depth
may
not
work.
So
those
are
the
places
we
want
to
have
something
different
and
I
was
reaching
out
the
dot
net
runtime
team
for
that
earlier.
E
So
all
of
a
sudden,
that's
not
left,
and
I
also
missed
following
up
with
them
in
a
proactive
basis
after
that.
So
I
need
to
get
back
to
them
to
get
an
update
on
that.
If
we
have
that,
we
are
like
our
instrumentation
will
be
very
strong.
We
might
consider
like
99
percent
of
the
cases
we
will
be
able
to
attach
with
additional
debt
support.
C
I
see
yeah
because
thinking
here,
as
far
as
I
remember
dot
net
seven
is
this
schedule
temptatively
for
november.
So
probably
they
are
going
cold
freeze
around
june
yeah
yeah
yeah.
Don't
do
others.
C
E
Three
months
yeah,
even
in
the
last
planning,
they
said
it
is
doubtful
because
if
it's
going
to
change
the
entire
thing,
I
think
it's
not
an
lts
feature
dot
net
seven.
So
they
said
it
is
doubtful
to
take
it.
If
it's
going
to
go
through
a
major
change
but
like
at
least
they
are,
they
wanted
to
take
a
look
and
help
us
out
other
ways
in
other
ways,
so
I
I'll
follow
up
with
them
that
there
is
like
another
channel
to
engage
them.
So
I
will
go
through
that.
E
F
Yeah
so
paulo,
I
was
able
to
join
late,
but
we
both
talked
about
if
there
were
any
other
things
that
we'd
like
to
see
added
to
net
seven,
that
that
would
make
things
easier.
And
I
know
that
I
haven't
looked
into
things
and
I
don't
know
if
anybody
else
in
this
group
has
taken
a
look
at
things
as
well.
But
that'll
be
something
to
consider.
C
Yeah,
I
I
think
we
should
look
into
that.
You
know
so,
for
instance,
after
we
we
chat.
I
I
had
the
I
remember
something
that
maybe,
in
the
future,
for
instance,
open
telemetry
wants
to
do
profiling.
C
I
know
that
the
spec
team
didn't
want
to
handle
that,
but,
for
instance,
the
profiling
we
use
async
local
with
the
spam
contacts
to
update
to
to
know
the
the
span
with
the
stack
for
activity.
Current.
That's
the
same
conceptual
thing:
the
current
is
not
going
to
expose
the
the
handle
because
their
sync
local
is
hidden.
We
just
get
activity
current,
so
I
think
I'm
gonna
ping
dave
about
that
to
see
kind
of
what
he
thinks
is
the
best
path
for
us
to
do.
That's
the
equivalent
in
open
telemetry.
C
This
is
just
an
example,
but
I
think
there
are
other
things
like
that
that
perhaps
needs
or
api
changes
or
runtime
support
you
know.
So
we
should
try
to
think
about
those
as
soon
as
possible.
The
the
freeze
is
in
june.
So
if
you
have
any
ideas.
E
Yeah,
it
is
also
come.
That
is
a
reason
for
me
to
ask
the
other
question.
Also
apollo.
Like
initially
I
brought
up
right
like
the
we
haven't,
started
the
work
for
the
log
exporter
and
metric
exporter.
So
we
don't
know
the
challenges
in
that
area.
Yet,
as
we
did
not
do
anything
on
that,
so
it's
better
to
do
some
prototype
at
least
we
will
get
to
know
like.
Are
we
going
to
hit
some
implementation
and
we
can
reach
out
faster
for
help
to
get
something
changed
there?
E
I
also
I'll
try
to
see
if
we
like,
if
we
definitely
try
out
in
that
way,
I'll
try
to
see
if
we
can
get
someone
from
the
dot
net
arc
to
continue
joining
with
us
for
the
sigma
meeting.
I
know
earlier
david
used
to
join
us
like
so
I'll
check
with
he
is
more
into
profiling,
so
this
needs
more
than
that.
So
I
will
check
if
some
other
pm
also
can
join
us
where
they
can
directly
take
her
request
and
like
and
get
the
responses
for
us
in
the
coming
weeks.
C
Yeah
that
that
will
be
very,
very
useful.
You
know,
I
think,
the
the
the
thing
that
crossed
my
mind
is
kind
of.
We
have
this
opportunity
for
net
seven
to
fix
these
things,
so
we
should
try
to
do
a
kind
of
brainstorm
in
their
effort
and
use
your
help
and
access
to
the
people.
So
we
can
get
stuff
if
needed
on
their
runtime
or
apis.
C
We
get
that
4.9,
so
the
future
kind
of
looks
brighter.
You
know.
D
Yeah,
it's
a
very
good
point
about
logging,
because
it
is
the
case
taking
hard
reference,
probably
to
microsoft.
Exchanges.
Logging,
then
probably
the
same
issue
as
with
the
activity
service.
E
C
Okay,
so
I
think
we
have
this
path
to
the
beta,
but
I
I
think
as
soon
that
is
done,
kind
of
or
in
parallel,
as
people
can,
but
at
least
when
that's
done,
we
really
should
kind
of
prototype
and
do
this
kind
of
stuff
to
bring
everything
that
we
can
in
time.
You
know,
at
least
you
have
some
discussion
conversation.
C
F
To
bring
up
that,
I
don't
know
if
got
discussed,
but
there
was
a
issue
submitted
to
fill
out
some
of
our
integration
test
gaps
and
I
just
wanted
to
call
out
if
somebody's
working
working
on
that
or
planning
to
work
on
that
reach
out
to
me,
I
have
some
ideas
on
how
we
can
simplify
some
of
the
boilerplate
that
that
we
do
so.