►
From YouTube: 2021-12-08 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
A
C
Oh
he's
here:
oh
hey,
hey,
yeah
hi,
everyone
did
they
change.
Was
the
password
changed
like
I,
I
thought
it
was
always
like
five
sevens,
even
with
the
new
link
for
the
meeting,
so
new
link.
B
C
C
C
A
Yeah,
I
just
have
the
one
exporter
configuration
thing
that
I
want
to
talk
about,
but
we
we
can
do
that
asynchronously,
because
I
know
c
joe,
you
haven't
had
a
chance
yet
to
look
at
it.
I
don't
know
if
michael
is
gonna
make
it
today.
B
Yeah,
maybe,
like
I
say,
like
update
for
anyone
who
might
be
looking
at
their
notes,
we
can
say
that
1.2
is
the
release
is
blocked
until
the
spec
is
stable.
B
It
looks
like
the
spec
is
trying
to
make
it
happen
in
the
next
few
days,
but
I
don't
know
for
sure
it
will
happen
before
end
of
month.
I
do
see,
like
george
did
mention
this
morning,
that
we're
trying
to
make
it
happen
by
end
of
this
month.
So
until
then
we
can
only
release
like
rc's,
so
we
may
do
like
more
one
more
rc
in
the
next
week.
A
B
B
Yeah,
if
there
are
no
other
topics
to
be
discussed
today,
we
can
end
early.
If
michael
is
not
joining,
I
think
we
can
discuss
that
pr
offline.
So
it
looks
like
he
has
a
builder
pattern
approach
and
you
have
a
slightly
different
approach,
maybe
like
now
that
both
years
are
out
there
we
can
compare
and
command
in
the
pr
itself
and
maybe
like
next
week
we
can
make
a
decision
on
that.
A
Yeah,
I
think
that's
fair,
I
think.
Well,
we
can
talk
about
it
more
once
you
have
a
chance
to
take
a
look
at
it,
but.
C
A
My
opinion
is
that
I,
like
I,
like
the
the
direction
the
builder
pattern
is
going,
but
there
are
some
elements
of
that.
My
pr
gets
at
that
I'd
like
to
I'd
like
to
discuss,
maybe
having
both
or
elements
of
both.
So
we
can
get
into
that
once
everybody's
had
a
chance
to
look
at
it.
A
I
had
one
other
quick
question,
maybe
just
because
I
was
curious.
I
I
just
saw
today
there
was
this
issue
that
I
had
responded
to.
You
had
responded.
First,
see
joe.
A
Or
so
about
well,
the
guy
now
has
a
question
about
hey.
The
self
diagnostics
worked
for
him,
but
he
has
running
containers
and
he'd
like
those
to
go
out
to
a
console,
and
I
was
kind
of
researching
this
a
little
bit
to
see
if
there
was
any
tooling
like.
I
assume,
he's
running
linux
containers
so
like
lttng
or
something
like
that
ability
to
pipe
event
source
out
to
the
console.
I
don't
know
if
that's
possible.
B
A
B
I
mean,
of
course,
we
can
write
our
own,
even
source
listeners
and
write
it
to
console
or
wherever,
instead
of
file,
because
we
already
do
that
do
the
writing
to
file,
so
we
should
be
able
to
modify
it
to
write
to
like
console
so
definitely
doable.
B
I
don't
think
that,
right
now
we
expose
any
knobs
to
change
the
output
from
file
to
something
else.
A
B
Alternately
alternately
like
anyone
can
write
their
own,
even
listeners,
even
source
listeners,
sure
that
also
should
work.
I
mean,
given
this
be
a
common
ask
going
forward.
I
think
we
should
be
able
to
handle
that
in
the
self
diagnostic.
There
was
also
ask
about
like
typing
it
through
I
logger
as
well,
so
I
mean
it's
a
little
bit
more
trickier
to
do.
B
I
logger
because
you
need
to
have
a
doc,
get
a
hold
of
an
eye,
logger
instant,
to
do
the
login,
so
we
cannot
really
provide
that
from
like
a
file
like
it
has
to
be
like
programmatically.
B
So
I
have
to
like
do
some
descending,
basically
something
like
when
you
initialize
the
sdk.
You
would
provide
a
logo
instance
to
be
used
if
there
is
a
logging
feature
turned
on
and
loading
feature
can
be
turned
on
as
before.
B
So
there
was,
I
think,
yeah
even
that
pr
is
attempting
to
do
that,
but
it's
only
doing
it
from
a
the
extension
start
from
stream.package
right,
but
we
should.
B
Because
there
is
no
nothing
which
ties
it
to
the
hosting
thing.
All
we
need
is
just
a
instance
of
a
logger.
The
sdk
should
have
an
instance
of
file
occur,
that's
all,
then
we
can
write
it
to
the
dialogue
if
the
login
itself
is
turned
on.
B
That
the
ilo,
the
dotnet
runtime
team,
has
an
issue
open
for
in
general
about
how
does
libraries
get
a
hold
of
high
logger?
So
it's
not
very
straightforward.
If
you
even
look
at
our
tracing
and
metrics,
it's
very
easy.
You
just
create
a
activity
source
or
meter
yourself
and
then
start
using
it,
but
for
I
logger
you
cannot
just
go
and
like
new
up
a
idler
or
local
provider.
You
have
to
get
it.
You
have
to
get
the
right
one
which
has
all
the
configuration.
So
it's
not.
A
B
I
don't
know
whether
the
daily
work-
you
know
there
is
an
issue
open.
I
don't
know
who
opened
that
issue.
It
is
being
tracked
in
the
runtime
repo.
Maybe
if
you
can,
I
can
find
that
that
also
would
contain
the
list
of
other
items
being
tracked
for
net
seven
consideration.
B
So
actually,
let's
do
it
this
way.
So
can
you
open
like
issues
in
our
repo,
the
same
report?
You
are
already
in
oh
yeah,
sure
just
just
yeah,
just.
B
And
just
for
the
label
require
runtime,
something
there
is
and
maybe
open
the
yeah
first
one.
It
should
have
the
link
crawled
all
the
way
down
yeah
observability
candidate
items
for
right.
That's
the
link,
I
believe,
yeah,
the
very
first
one.
I
love
libraries
that
don't
use
di
to
use
very
first
one
from
the
top
oh
yeah,
and
also
there
is
a
related
issue
explorer
using
event
source
before
I
logger.
B
So
I
don't
think
we
opened
it,
but
whoever
opened
it
like
they
might
be
facing
the
similar
concerns,
as
we
were
just
discussing
like
we
want
to
like,
we
are
using
even
even
source
for
our
logging,
but
users
might
want
it
in
eye
logo.
B
There
are
some
issues
yeah,
I
don't
think
like.
We
ever
like
spend
enough
time
like
prioritizing
this,
but
if
there
is
enough
demand,
then
we
should
be
like
prioritizing
this,
but
it
doesn't
look
like
a
lot
of
people
asked
mostly.
We
are
fine
with
the
self
diagnostic
module
and
the
best
thing
is.
It
can
be
turned
on
and
off,
ad
hoc
without
restarting
the
process.
B
So
mostly
it
should
be
satisfying
the
needs.
But
if
there
is
a
considerable
demand,
we
can
enhance
the
self-diagnostic
module
to
pipe
it
to
someplace
else
like
eye
dogger.
Most
likely
I
don't
get
along
would
do
because
then
I
logger
itself
as
a
console
provider,
so
it
can
be
redirected
to
console
or
elsewhere
right
yeah.
It's
just
that,
like
someone
has
to
spend
some
time
like
designing
and
coming
up
coming
up
with
a
proposal.
So
let's
just
wait:
if
there
is
enough
feedback,
we
can
we
prioritize
that
yep.
B
I
like
that,
makes
sense
for
the.
A
Yes,
we
were
doing
that
though
we
actually,
we
deprecated
that
exporter,
because.
B
B
We
expect,
like
dotnet
team,
to
clarify
that,
because
the
best
thing
with
even
sources,
you
don't
need
anything
to
be,
provided
you
just
create
it
and
log
and
like
I
logger
where
you
really
need
someone
to
give
it
to
you,
so
it's
more
like
da
friendly
than
the
event
source
right.
That's
why
these
issues
are
open
just
one
day
like
maybe
you
have
a
noise
back
from
vacation
like
we
can
ask
like
whether
this
is
something
which
they
are
actively
considering
or
it's
just
like
put
it
as
a
placeholder
for
7.2.
B
That's
how
I
heard
you
had
like
one
more
issue
open
right:
the
one
like
retry,
something.
A
This
was
the
one
we
were
just
looking
at
it
was.
It
was
originally
about
retry
functionality
for
jrpc,
which
I
understand
where
still
waiting
on
some
functionality
from
the
grpc.net
library.
B
I
see
okay,
we
don't
retray
that
it's
the
only
thing
which
I
know
for
sure
I
think
like
when
you
say
like
we
are
waiting
for
like.
Are
we,
depending
on
the
grpc
client
library
itself,
to
provide
that.
B
A
A
I
mean
maybe
add,
support
for
grpc
client
libraries.
This
seems
done.
Oh
okay
yeah.
Maybe
maybe
we
can
implement
retry,
because
that
would
be
super
valuable.
We
have
experienced
new
relics,
experienced
customers
who
have
data
dropped
from
their
clients.
C
B
A
Yeah,
I
might
actually
look
into
this
this.
This
looks
maybe
promising
that
maybe
the
okay
features
there.
There
is
another.
B
Thing
which
was
like
kind
of
hidden
in
the
ask
if
you
go
back
to
the
issue,
so
the
other
issue
is
like
this
is
only
going
to
help
with
like
a
very
simple,
like
network
error,
where
you
can
retry,
but
it
won't
really
help
with
extended
things.
So
I
think
you
need
to
go
back
like
one
more
nah.
That's
it
yeah,
okay,
yeah,
the
very
first
one
yeah
right.
B
So
there
is
a
oh
like
we
need
to
put
things
in
that
temporary
queue
and
keep
retrying,
something
like
that,
yeah
that
one
right
yeah.
So
that's
a
little
bit
more
involved
one,
because
we
might
want
to
keep
things
in
disk
or
something
like
that.
There
is
an
active
pr
which
someone
in
my
team
started
but
never
finished,
which
is
basically
to
provide
a
like
apa
to
store
things
into
a
disk
and
retrieve
it.
B
Pr
should
be
still
there.
I
don't
think
it
ever
got
much,
so
I
think
it.
B
B
A
B
B
Yes,
someone
from
our
team
would
be
working
on
it.
Quite
I
mean
I
won't
commit
to
it,
because
the
first
priority
is
to
fix
that
in
assure
so,
but
we
like
to
have
it
neutral
so
that
we
can
put
it
in
this
report
itself
and
have
all
the
other
exporters
liberated.
A
Cool
cool,
but
in
the
meantime
you
know
like
the
otlp
exporter.
A
B
Like
a
temporary
500,
there
is
a
like
set
of
spec
written
for
that
like.
When,
should
you
try,
so
we
should
be
able
to
do
that.
This
is
more
like
a
even
deeper
into
that
thing
like
if
it's
down
for
an
extended
period
of
time,
we
don't
want
to
like
start
losing
data,
because
our
circular
buffer
has
a
capacity
right.
B
So
if
the
exporter
is
like
step
four
like
couple
of
cycles-
and
I
think
we
would
hit
the
limit
of
the
circular
buffer
and
start
losing
data
so
to
prevent
that,
then
only
we
need
this
persistent
file
based
so
right.
So
it's
a
like
even
deeper
approach
to
prevent
data
loss,
but
I
totally
agree
that
otp
one
should
be
like
relatively
straightforward.
We
don't
need
to
like
wait
for
this
to
happen.
We
should
be
able
to
leverage
the
underlying
libraries
feature
relatively
easy.
A
Yeah:
okay,
okay,
okay,
yeah
great!
This
is
good
context.
B
B
A
If
you're
up
for
it,
I
know
that
you
broached
the
issue
earlier
this
year
about
working
to
get
our
pipeline
a
little
bit
more
healthy
yeah.
B
A
B
A
B
Maybe
I'll
speak
to
them
offline
and
see
if
one
of
them
can
pick
it
up
cool,
okay,
okay,
I
have
to
go
now,
so
I
hope
there
are
no
other
agendas
so
see
you
again
next
week.