►
From YouTube: 2020-08-12 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
B
B
Good,
I
saw
the
open
telemetry
page
and
I
thought
I
might
just
hop
on
and
see
what
these
meetings
are
sort
of
about.
With
regards
to
the
logging
support
that
you're
trying
to
add.
B
I'm
sort
of
interested
I'm
trying
to
learn
a
little
bit
more
about
the
general
framework.
But
I
was
just
curious
to
see
if
there
was
any
sort
of
I'm
mostly
interested
in
sort
of
like
data
contracts
in
general,
and
so
this
is
sort
of.
As
far
as
I
can
tell
you
guys,
are
discussing
sort
of
the
contracts
around
how
open
telemetry
might
interact
with
certain
backend
data
sources
for
spans
and
logs
and
whatever.
But
in
this
particular
case
logs-
and
I
guess
it's
just
sort
of
of
interest
to
me.
B
B
A
Yeah,
this
is
right
meeting
seems
like
it's
going
to
be
a
quiet
meeting.
I
don't
know
where
is.
A
Else
on
vacations,
perhaps
the
open
telemetry
defines
a
data
model
for
logs.
It's
part
of
the
specification,
if
you
haven't
seen
it,
it's.
B
Yeah
I've
been
reading
through
some
of
the
meeting
notes,
but
I
wasn't
able
to,
I
guess,
get
everything.
A
B
Yeah
well
much
like
many
of
the
other
people
that
I've
noticed
on
the
call.
My
company
does
a
lot
of
stuff
with
logs.
So
I
guess
I'm
just
curious,
because
I've
been
in
a
number
of
different
discussions
around.
B
How
how
data
is
formatted
such
that
it
can
be
used
for
sort
of
searching
as
well
as
for
actually
writing
sort
of
like
a
matching
engine
where
you
can,
which
is
sort
of
like
searching
where
you
create
some
sort
of
secondary
artifact
when
you
match
against
the
log-
and
I
guess
that
has
some
tangential
relationship
to
this,
because
this
is
more
about
when
the
logs
are
being
created.
Inside
of
your
cloud
environment
or
whatever
else
and
then
being
piped
out
to
the
back
end
itself.
A
Yeah,
that's
that's!
That's
typically,
where,
where
open
telemetry
focuses
open,
telemetry
does
not
usually
is
not
concerned
with
how
things
work
on
the
back
end,
how
you
search
for
the
telemetry,
the
the
primary
concern
of
inflammatory,
is
how
the
telemetry
is
generated
and
how
it's
transmitted,
how
it's
delivered
collected,
but
not
how
it's
used
afterwards.
How
are
you
right.
B
A
With
it
how
it's
it's
stored
or
searched,
or
but
it
it
certainly
deals
with
things
that
are
important
for
the
back
end.
B
B
Yep
yeah,
that
makes
sense,
yeah
and
plus,
I'm
sure
you
know,
we
use
a
variety
of
different
things
for
gathering
our
data.
We
don't
use
open
telemetry,
so
I
had
some
interest
in
sort
of
better
understanding
it
to
see.
If
there's
it's,
because
it
seems
like
a
fairly
aspirational
project
in
terms
of
the
different
things
it
can
collect,
we
focus
mostly
like
our
own
production
systems.
B
We
have
somewhere
where
we
collect
metrics
and
we
have
somewhere
where
we
have
a
cabana
log
stash
for
logs,
but
and
that
could
I
don't
know
if
there's
interrupt
operability
with
open
telemetry,
but
it's
just.
B
I
was
thinking
if,
if
there
was
something
like,
we
tried
using
zipkin
as
well,
and
it
just
didn't
just
another
thing
and
it
kind
of
it
didn't
become
widespread
in
our
production
environment
so
but
open
telemetry
might
make
it
easier
to
use
that
sort
of
thing
if,
if
you're
collecting
the
logs
and
the
spans
and
the
traces
all
with
one
tool,
yeah.
A
B
A
The
idea
is,
you
collect
it
using
one
set
of
libraries
in
a
uniform
way
and
you
collect
it
in
a
in
a
vendor
agnostic
way
right,
regardless
of
what
your
choice
of
the
back
end
is
you
collect
it
in
the
same
way
and
the
libraries,
the
open
planetary
libraries
have
exporters
for
vendor
specific
formats
or
the
the
data
goes
through
the
collector,
which
contains
a
lot
more
supports
a
lot
more
formats
and
that
the
entire
open
telemetry
basically
is
vendor
independent.
From
that
perspective,
right.
B
A
From
the
from
the
from
the
customer's
perspective,
basically,
you
are
the
the
locking
is
vendor,
locking,
is
removed
from
the
picture
and
then
they,
whatever
vendor
they
choose
on
the
back
end.
They
just
reconfigure
most
likely
just
the
collector
if
they
you're
using
a
collector
or
if
they
are
sending
directly
to
the
back
end.
They
will
just
again
need
to
reconfigure
the
the
application,
the
the
sdk
on
the
planetary
sdk
in
the
application.
A
So
this
is,
I
guess,
the
primary
promise
of
open
telemetry.
The
removal
of
the
dependence
of
on
the
on
the
particular
vendors
vendor
provided
solution
on
the
on
the
customer
side
of
things.
B
Right
so,
in
terms
of
like
I
know,
this
discussion
doesn't
have
anything
to
do
with
the
language
right.
It's
just
about
like
the
terms
and
the
capabilities
of
open
telemetry
for
logging.
A
Almost
there
is,
there
is
going
to
be
at
least
as
an
as
an
example
as
an
illustration,
an
implementation
for
one
popular
language,
most
likely
for
java.
We
will
implement
an
open,
telemetry
sdk
for
for
java
logging
libraries,
which
will
allow
you
to
let's
say
you're,
using
block
4g
in
your
java
application,
together
with
open,
telemetry
sdk
for
logging.
You
can
then
make
sure
that
the
logs
that
your
java
application
is
meeting
are
of
inflammatory
compliant
compliant,
meaning
that
they
are
written
in
a
format
that
they
are.
A
B
A
That
means
that
you
have
the
trace
id
and
span
id
in
your
works,
and
then
it
also
provides
means
for
sending
logs
from
the
application
from
the
login
library
directly
to
the
backend
or
to
the
collector,
using
open
clan
to
protocol
instead
of
writing
to
a
file
and
then
taping
the
file
parsing.
The
file.
Basically
by
passing
this
file
as
an
intermediate
meeting
for
storing
the
personal
world.
B
A
But
but
you're
right
in
that,
we
are
primarily
concerned
about
defining
at
least
right
now
defining
what
it
means
to
be
open,
telemetry
compliance.
So
that's
why
the
data
model
defining
how
the
applications
are
supposed
to
write
logs
if,
if
they
are
writing
blocks
to
files
right
in
a
way
that
is
understandable
for
for
a
logging
agent,
which
expects
the
logs
to
be
the
logs
to
be
in
open,
telemetry
compliant
structure
so
that
it
can
later
read
this.
This
talks
extract
the
logs.
B
A
Yes,
we
will
be
adding
support
for
different
vendor
formats
in
the
collectors
so
that
at
the
minimum
in
the
in
the
logging,
libraries,
we
don't
expect
a
lot
of
vendor-specific
exporters,
but
at
the
minimum
it
will
support.
Sending
to
the
open,
telemetry
collector
there
in
open
temperature,
collector,
it
can
be
translated,
will
be
translated
to
a
vendor
specific
format.
A
Let's
say,
for
example,
you
want
to
send
to
splunk
and
there
will
be
a
splunk
hack
protocol
exporter
in
the
collector,
which
you
can
configure
and
basically
send
you,
your
java
application
logs
that
you
were
sending
from
the
login
library
to
the
collector
and
collector
sending
to
the
back
end.
This
is
this
is
exactly
the
same
model
for
traces
and
metrics.
A
A
Telemetry
is
the
case.
We
support
typically
three
four,
maybe
four
months
like
it's,
the
open,
telemetry
protocol,
open,
sensors,
jager
zipkin
prometheus.
I
think
that's
it
yeah
collector.
We
have
dozens
of
different
formats.
So
if
you,
if
you
want
to
send
to
let's
say
elastic,
you
can
use
the
collector
to
receive
from
the
application
using
inflammatory
format
and
then
translate
this
to
elastics
format
and
then
send
to
your
elastic
server.
A
B
Like
the
severity
number,
I
guess
has
to
be
a
number
or
something
like
that
looking
at
the
stuff
here,
but
then
there's
also
some
sort
of
arbitrary
data
that
you're.
Actually,
yes,
that
constitutes
the
log.
Yes,.
A
B
B
Yeah
yeah,
I
was
curious
because
we
we
do
a
lot
of
we
work
with
a
lot
of
security
data,
and
I
was
just
it's
probably
outside
of
the
realm
of
the
sorts
of
things
that,
like
correlating
different
events
by
other
ids.
That's
that's
something
that
you
could
do
on
your
own
just
by
adding
that
data
into
your
logs.
If
you
really
want
it.
B
A
A
Yeah,
okay,
so
if
you
saw
that,
if
you're
interested,
please
comment
have
a
look
at
that,
so
that's
the
thing
that
I
was
I've
been
working
recently
on
some
other
regulars.
David
he's
been
working
on
the
java
solution.
B
So,
what's
your
background
with
regards
to
open
telemetry.
A
I
have
been
with
open
telemetry
for
more
than
a
year
already.
B
B
I
noticed
that
there's
so
many
things
in
the
industry
where
it's
there's
just
no,
no
one
ever
agreed
on
it.
So
it's
just
you
know
it's
just
crazy
and
I'm
sure
you've
experienced
that
anyone
that
deals
with
blogs
experiences
that
it's
just
like.
How
do
you?
A
I
mean,
I
think
so
far
we
we've
been
quite
successful
in
coming
to
an
agreement
around
things,
at
least
for
traces
and
metrics.
We
were-
and
I
think
we
will
also
be
successful
for
the
vlogs
you're
right-
that
there
is
a
lot
of
different
opinions
around
how
things
should
be
working.
B
Right
right,
yeah-
and
I
think
I
think,
people
that
are
paying
attention
and
have
experienced
the
pain,
will
see
something
like
this
and
and
quickly.
You
know
at
the
very
least,
give
it
a
shot
because
they're
just
so
tired
of
the
mayhem
of
trying
to
support
one
thing
after
the
next
okay
cool.
Well
yeah.
I
think
I
better
check
out
more
about
this
collector.
I
think
I'm
curious
about
that
so
other
than
that.
B
I
don't
really
have
much
to
say,
but
I
will
check
out
the
pull
request
you
put
out
and
just
see
what
you
guys
are.
What
the
latest
is.
B
Okay,
okay,
all
right,
it
was
good
to
meet
you.
What's
your
name.