►
From YouTube: 2021-09-01 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
A
B
Yeah,
I
think
there
were
a
little
bit
more
on
the
quiet
side
like
there
was
not
much
discussed
there.
One
of
the
recent
topics-
and
this
is
what
one
brought
by
me-
was
about
possibility
of
adding
a
lock
log
record
receipt
timestamp.
B
I
actually
brought
this
idea
to
the
recent
open,
telemetry,
specs,
sick
and
bogdan,
didn't
like
this
idea.
He
he
said
that
if
we
need
it
for
locks,
why
we
don't
need
it
for,
let's
say
metrics
or
yeah
or
tracing,
and
I
I'm
wondering
myself
like,
because
maybe
this
another
field
is
another,
let's
say
complexity,
but
I've
put
some
rationale
behind
that
in
the
issue.
So.
A
I
think
the
difficulty
here
is
once
you
recognize
the
fact
that
there
is
no
single
time
stop.
Then
you
can
come
up
with
several
timestamps
right
at
the
time
when
it
was
generated
when
it
was
written
to
file
when
it
was
read
from
file
when
it
was
forwarded.
Let's
say
in
the
collector
received
by
the
server
like
you
can
come
up
literally
with
half
a
dozen
interesting
time
stamps
and
then
what
do
you
do
you?
Do
you
record
all
of
those
unclear?
I
I
don't
know.
A
What's
the
answer
to
this,
I
think
it
was
in
the
past
discussed
a
bit,
but
we
never
came
to
kind
of
a
nice
proper
solution
for
this.
I
don't
think
the
solution
is
to
like
literally
have
six
seven
different
timestamp
fields,
because
they
are
also
not
probably
limited
to
whatever.
We
can
imagine
right
now
right.
Something
can
be
imagined
that
I
mean
there
is
something
that
you
can
probably
record
that
we
don't
even
know
that
should
be
recorded.
B
Yeah
yeah,
exactly
also
like
another
thing
that
could
help
here
is
maybe
having
some
field
like
as
a
semantic
convention.
Even
that
describes
how
given
timestamp
was
associated
with
the
record,
because
now
you
can
let's
say,
handle
this.
This
time
stand
differently.
It
was
assigned
through
some
parsing
or
it
was
assigned
through
received
timestamp.
If
it.
B
C
Yeah
we
we
talked
about
that
a
little
bit
last
time.
I
think
christian
brought
that
up
that
I
I
really
thought
that
was
a
great
idea
and
it
you're
just
in
that
it
allows
you
to
if
you've
got
a
timestamp
that
looks
bad
and
it
was
parsed.
Then
okay
treat
it
as
probably
bad,
like
there
were
a
lot
of
things
that
come
out
of
just
having
a
flag
there
that
I
thought
was
kind
of
cool.
A
A
D
So
yeah
I
wanted
to
give
this
update.
We
are,
we
are
like
there
is
no
like
piper
release,
but
they
can
still
right
now
install
the
through
the
github,
so
I'll
get
my
antennas
released.
The
experimental
packages
and
couple
of
people
have
also
showed
interest
in
you
know
using
this
and
then
yeah,
so
we
wanted
to
get
their
feedback
as
well.
That's
update
there.
D
One
one
question
I
have
is
so
the
we
I
was
verifying
like
the
whole
functionality
with
the
color.
I
was
unable
to
look
at
the
like
if
the
trace
context
is
reaching
the
collector
or
not
with
the
logging
exporter
was
wondering
you
know
if
you
can
add
that
information
on
there
as
well.
A
D
So
like
there
are
like
four
or
five
people
who
showed
interest
in
this.
Using
this,
we
will
have
something
like
feedback
from
them
as
well.
Hopefully,
yeah
yeah.
A
No,
this
is
great
great.
Thank
you.
Let's,
maybe
let's,
let's
kind
of
post
an
announcement
in
whatever
channels
are
available,
trust
this
luck,
maybe
to
let
people
know
that
this
exists
and
we
are
seeking
feedback,
but
this
is
great.
Thank
you.
Thank
you
for.
D
A
D
E
I
have
an
update
on
java.
If
you
want,
I
didn't
put
it
in
the
agenda.
Yes,
please
so
yeah
we're
making
some
good
progress
on
the
java
side
as
well.
There's
a
couple
of
pr's
that
have
gone
in,
but
we
have
a
couple
of
more
that
should
happen
in
the
next
two
weeks
and
then
there
should
be
a
working
prototype
in
the
java
instrumentation
and
yeah.
It's
it's.
E
A
And
sorry,
the
the
prs
that
you
said
went
in
are
they
in
the
official
open
telemetry
java
sdk
is
the
key
repository?
Oh
nice.
E
So
yeah
it's
progressing
and
yeah,
so
that'll
be
usable
soon
from
the
sdk
side
and
I
believe,
there's
already
auto
injection
stuff.
That's
in
the
java,
auto
instrumentation
that
injects
trace
id
and
various
other
things
into
logging
libraries.
So
I
think
that
already
exists
there.
E
So
yeah
I'll
keep
you
posted.
Hopefully
we'll
have
some
good
news,
I'm
going
well.
Hopefully
we'll
have
some
good
news.
Before
I
leave
town.
We
have
this
every
other
week
right.
This
call
yes,
okay,
then
I
won't
update
for
another
month.
Actually,
I'm
gonna
miss
the
next
one.