►
From YouTube: 2022-02-02 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
All
right,
so
the
first
thing
that
I'm
going
to
discuss
is
our
open,
pull
request.
We
have
two
pull
requests
open
right
now.
I
merged
tmos
this
morning
with
the
environment
variable
refactor,
so
we
have
one
for
making
it
possible
to
pass
the
endpoint
url
and
headers
the
constructor
of
the
otlp
hdp
exporter
right.
It
looks
like
you
are
working
on
this
one
with
this
person
and.
C
A
A
A
Oh
yeah,
yeah
yeah,
you
mentioned
I'd,
prefer
any
direct
configuration
of
an
exporter
to
be
the
first
argument
from
the
constructor.
D
A
C
Yeah
sure
so
I
had
to
talk
with
timo
regarding
this,
as
only
so,
the
thing
is
like
like
we
wanted
to
introduce
the
metrics
and
traces
and
some
feedback
type
of
loops,
which
will
help
to
monitor
the
library,
the
open,
the
open,
telemetry
php
module.
So,
like
team
only
suggested
me
to
use
the
event
listener
mechanism
from
what
both
symphony.
E
C
Ladder
will
use
this
blue
user.
So
presently
this
pr
deals
with
the
I
have
using
this
pi
created
the
api,
the
event,
the
event
listener
and
dispatcher
dispatch
api
through
which
so
which
we
can
register
the
events
in
our
code,
the
points
of
code.
We
want
to
record
the
events
and
the
listener
could
be
configured
according
to
it
could
be
automate.
Some
listeners
could
be
automatically
given
to
automatically
fixed
to
the
library,
and
some
listeners
can
be
given
to
choice
of
users.
C
A
Excellent,
this
is
this
pr
looks
like
it's
ready
for
review
and
merge.
Do
you
agree
with
that?.
C
No
like
this
is
just
the
initial
api
and
the
design
there
has
to
be.
There
are
more
subtle
changes
to
be
done,
like
which
points
we
need
to
open
the
event
and
what
type
of
listeners,
but
first.
A
A
A
I'm
not
yet
doing
a
good
job,
I'm
trying
to
do
a
better
job
of
being
a
little
bit
more
of
a
project
manager
for
this
project,
even
though
I
don't
want
to
be
this
board
has
all
the
open
tasks
that
we
need
to
do
for
tracing
for
our
tracing
to
be
spec
compliant.
As
far
as
I
can
understand,
the
way
that
this
board
is
broken
down,
items
that
are
in
the
to-do
column
are
things
that
are
still
left
to
do.
They
may
or
may
not
have
an
owner
assigned
with
them,
but
they
can.
A
A
C
C
A
A
Me
or
send
it
in
the
channel
or
whatever
I
just
wanna,
try
to
do
a
better
job
to
keep
track
of
all
the
things
that
we're
working
on
so
that
other
people
can
be
aware
of
what
we're
working
on.
D
D
Yep,
no
pressure
was
just
a
question.
Okay
and
I
think
some
did
you
close
some
issues.
D
Okay,
because
I
think
maybe
not
the
last
time,
but
a
while
ago,
when
I
looked,
there
was
still
the
symphony
and
issue
which
I
closed.
But
oh.
A
D
D
A
D
I'm
just
kidding
yeah
okay,
so
those
ones
are
not
there
anymore.
Okay,
maybe
we
have
to
assign
to
from
one
yeah
I'm
guessing.
Maybe
some
of
the
issues
might
need
to
be
de-assigned,
a
specific
word
or
that's
the
that's
the
opposite.
D
A
D
Yeah,
I
refuse
to
do
a
lot
of
I'm
not
really
a
big
fan
of
jira.
A
You
and
me
both,
but
you
know
it's
a
necessary
evil.
I
think
I
will
unassign
the
span
context,
one
because,
oh
sorry,
that
one's
completed,
so
we
can
put
that
in
done
then
yep.
That's.
I
think
that
all
of
the
one
I
could
probably
I'm
going
to
re-review
the
ones
and
to
do.
I
think,
I
think,
technically,
all
the
ones
that
are
assigned
to
people
are
assigned
to
people
that
should
be
working
on
them,
but
I
don't
necessarily
know
that
they're
going
to
so
I'm
gonna
have
to
follow
up
with
them
or
something.
A
A
A
D
D
So
at
the
moment
we
only
have
aws
stuff
there
and
symphony
things.
However,
there
is
also,
of
course,
an
issue
board,
and
the
thing
is
so
are
my
idea
or
a
proposal:
let's
put
it
in
that
way
that
we
move
all
the
issues
we
have
there
into
the
main
repository,
so
they
get
more
exposure
and
on
the
other
side
they
are
easier
to
handle.
D
So
that's
just
an
idea,
so
we
don't
have
to
do
this,
but
I
don't
know
how
many
people
are
actually
looking
into
this
repository
or
even
are
even
aware
that
there's
another
repository,
because
I
don't
think
it's
anywhere
linked
on
our
main
people.
A
D
Yeah,
especially
since
we
have
this
at
least
rudimentary
symphony
bundle
now,
so
it
could
be
added
to
at
the
bottom
of
the
region.
So
people
also
can
look
there
also
if
they
for
some
reason,
need
to
use
aws.
A
Long
term
you're
proposing
that
we
keep
track
of
all
the
issues
in
the
open,
telemetry,
php
repository
and
just
maybe
I
don't
know,
is
there
some
way
that
you
want
to
correlate
them
back
to
their
contributory
pen.
Maybe
we
could
add
a
specific
label
for
them.
A
Yeah
that
seems
reasonable
to
me,
I'm
trying
to
I'm
trying
to
best
balance
the
creation
of
an
issue
versus
like
discoverability
of
an
issue.
A
A
D
Okay,
well,
we
could
for
once
we
could
add,
like
a
you
know,
auto
automatically
command
thing
there,
or
I
think
you
can
even
probably
automatically
move
the
issues
to
the
because
if
you
click.
C
D
You
can
you
can
just
move
them
to
another
repo
and
I
guess
you
can
do
the
same
with
the
apis.
A
Yeah,
I'm
looking
to
see
if
there's
a
way
for
me
to
just
remove
the
issues
yeah
or
yeah.
C
D
A
Transfer
issue
got
it,
I
understand,
so
so,
shall
we
transfer
all
the
issues
that
are
on
here
into
the
open,
symmetry
php
repo
and
then
remove
the
issues
to
have
to
make
a
mention
in
the
readme.
If
you
have
an
issue
to
open
it
in
the
optometry
php
repo.
B
A
D
Okay,
cool
and
then
yeah.
Maybe
you
can
introduce
like
a
prefix
or
something
or
automate
things
so,
but
I
think
we'll
give
this
country
the
repo
a
little
bit
a
better
explosion.
D
And
the
same
goes
with
all
the
package
stuff.
D
D
But
then
this
reminds
me
of
another
idea
or
thought
I
had
if
we
actually.
I
think
we
briefly
talked
about
this
and
it's
like
if
we
want
to
move
questions
from
the
issue
board
into
the
discussion
board
automatically
or
not
automatically
or
encourage
people
to
use
this
discussion
boards.
I
don't
know
if
that's
better
or
not,
that
was
was
just
an
idea,
so
maybe.
D
So
my
point
is
that
you
know
like
we
had
a
few
questions
or
something
in
the
worst
couple
of
weeks
or
days,
the
problem
or
the
thing
with
questions
on
the
issue.
Borders
always
you
know,
then
can
they
be
closed
or
should
they
be
close
and
otherwise,
if
they
closed
you
know,
sometimes
the
questions
would
be
valuable
for
other
people.
You
know
who
are
looking
for
something,
and
so.
A
It
is
good
historical
context
to
be
able
to
see
like
the
decision
entry
behind
a
decision
rather
than
just
the
issue
and
upgrade
closes
us
because
we,
you
know
xyz.
D
D
I
never
power
used
this
discussion
board
before
so,
but
since
it's
there
and
for
now
it's
relatively
empty,
maybe-
and
I
always
had-
but
I
always
had
a
little
bit
of
problems
with
questions
being
in
the
issue
reports,
because
there
are
not
really
issues
and
you
know
when
will
they
be
closed.
A
Sounds
like
a
good.
I
think
that
maybe
we
should
maybe
for
now
we
should
say
we
can
encourage
people
to
actively
use
issues
for
discussions
and
we
can
see
like
sort
of
see
how
it
works,
and
if
we
find
that
the
discussions
feature
in
github
is
really
helpful
for
us,
we
can
continue
to
use
that
otherwise,
we'll
use
a
combination
of
both,
but
I
can
see
where
that
also
would
have
some
decision
fatigue.
So
I
don't
know.
Maybe
we
can
give
it
a
go
and
see
how
it
goes.
We
can
yes
later.
D
E
No
offense
taken
just
thinking
about
the
the
discussions
I
I
was
unaware
that
there
was
even
a
discussion
in
there.
So
there's
clearly
it
you
have
to
go
searching
at
the
moment,
but
I
wonder
whether
we
could
have
whatever
process
is
sending
a
slack
message
to
us
when
a
new
issue
is
created.
E
A
D
You
have
admin
rights
for
our
channel.
A
D
Yeah,
well,
I
think
this
on
or
off
decision.
If
an
app
is
I
mean
if
it's
already
well.
Actually
I
don't
know
I've
played
around
with
a
lot
of
stuff,
it's
like
in
the
past,
but
yeah.
I
don't
really
know,
remember
how
the
github
stuff
worked,
but
I
think
you
have
to
have
access
to
the
app
on
on
github.
A
I
did
just
do
a
curse.
I
did
just
do
a
cursory
search,
but
it
doesn't
look
like
the
github
integration
has.
It
doesn't
look
like
it
has
discussions
as
a
feature
enabled
for
it.
Yet
you
can
just
subscribe
based
on
a
repo,
and
I
think
it
just
performs
all
the
actions
based
on
that
repost
subscription
and
I'm
sure
that
they
just
probably
haven't
added
discussions
as
a
talking
point.
Yet.
D
Well,
I
think,
there's
probably.
D
Yeah,
if
it
doesn't
work,
there's
probably
a
way
to
do
this.
I've
seen
that
this
keyboard
is
enabled
for
all
slack.
So
at
the
end
of
the
day,
you
can
just
write
a
little
bot
so.
D
If
you
go
yeah,
if
you
have
a
look,
you
can
have
a
look
of
what
kind
of
I
don't
know
what
they
call
plugins
are
enabled
for
this
legs,
or
this
keyboard.
It's
basically
yeah
a
spot
where
you
can
create
little
bots
for
ruto
stuff,
and
it's
like.
D
Yeah,
but
the
you
can
it's
not
so
hard
for
to
create
an
app
for
select,
but
then
you
have
to
reduce
it
with
cncf
and
stuff.
Like
this,
I
don't
know
how
much
yeah
pre-bureaucracy
this
will
create.
A
That
would
be
I'm
sure
that
that
would
not
be
the
easiest
thing
on
the
planet
so
anyway,
all
right.
So
that's
that
was
that
I
got
the
only
other
thing
that
I
had
on
the
agenda
was
teemo
talking
about
your
kubernetes
discussion,
but
I
want
to
give
you
extra
time
for
that.
If
you
need
it
and
I
wanted
to
check,
does
anybody
else
have
anything
that
they
wanted
to
bring
up
on
the
agenda
today?.
E
Just
to
note
that
the
the
work
else
you
know
I'm
getting
protobuf
working
for
eight
one
is
slowly
limping
to
completion
yeah.
My
tests
are
green,
just
waiting
for
more
feedback
or
more
activity
from
someone
from
google
yeah.
D
Cool
those
google
guys
so
and
there
are
people
working
for
them
and
they
can't
even
move
their
buttocks
yeah.
A
A
D
Okay,
no,
I
actually
wanted
that's
the
question.
I
wanted
to
show
you
guys
first,
something
some
more
security
relevant
stuff
in
k3.
Yes,
I
don't
know
about
you.
Brett
is
already
open.
We
can.
We
could
schedule
a
different
date
for
this,
maybe
a
little
bit
later
on
another
day.
So,
if
that's
better,
because.
D
That's
a
question
because
I
can
show
you
stuff,
but
it
will
need
some
yeah.
It
will
be
some
heavy
input
for
you
guys
so
sure.
A
Yeah
no
problem,
I
can
schedule
some
other
time.
That's
more
connected.
D
D
D
The
other
stuff,
when
everything
is
said
and
done,
then
I
can
show
the
stuff
which
is
not
relevant
to
everybody
here,
but
there's
yeah,
basically,
all
they
set
up
and
stuff.
That
would
be
not
very
intelligent
to
put
it
on
recording
on
youtube,
but
then
yeah
the
the
stuff
when
this
is
basically
just
the
setup
for
kubernetes.
Who
will
talk
about
them
when
we
actually
have
the
other
stuff
in
place
and
there's
no
problem
of
showing
this
here,
because
that's
not
really
security
relevant
or
something
like
this.
D
D
D
I
know
by
the
time
I
could
have
created
this
the
10
repositories
and
have
them
registered
on
packages
and
stuff
like
this.
But
since
I'm
lazy
I
rather
work
20
hours
to
have
this
automatic
and
be
reusable,
and
it's
a
little
bit
more
fun
for
me.
So
I
showed
you
told
I
think
brett
and
bob
about
this
temporal
stuff,
right,
yep
and
that's
actually
really
cool.
I
had
to.
I
had
a
little
bit
of
a
look
into
it
a
few
months
ago,
but
I
never
played
around
with
it.
D
Where
is
it?
You
can
just
link
it,
and
if
anybody
is
interested,
you
can
look
into
it.
So
it's
basically
it
sounds
very
simple
and
as
if
it's
nothing
very
cool,
so
it's
a
fault,
tolerant,
workflow
automation.
D
D
Yeah
calling
packages
to
publish
it
there
and
there's
some
other
stuff
or,
for
example,
things
you
may
want
to
do
when
there
is
a
release
or
things
and
then
synchronizing
the
repositories
and
the
cool
thing
about
temporal
is
basically
you
can
set
up
workflows
there
and
even
if
something
crashes,
it
will
pick
up
from
the
exact
point
where,
basically,
the
crash
happens,
and
you
can
do
stuff
like
sleep
for
40
days
or
even
one
year.
You
know
imagine
you
have
a
process
and
that's
sleeping
for
a
year.
D
So
that's
basically
normally
you
would
never
do
something
like
this
and
you
can
really
do
very
cool
stuff
like
with
it
like
this.
So
that's
yeah,
very
nice
cool
tool
for
doing
automation,
stuff
and
they
have
because
for
the
roadrunner
guys
who
did
this
app
server
roadrunner?
They
actually
worked
with
the
temporal
guys
to
have
integrations
now,
so
it
works
very
nicely
with
php
together,
it's
a
cool
tool
to
do
stuff
like
this
long
story
short.
D
D
D
Yeah
anyway,
so
this
is
one
point,
and
this
is
one
thing
we
will
do
the
kubernetes
cluster
and
then
of
course,
or
also
enter
interests,
postings
and
demos
stuff,
and
things
like
that,
so
it's
nearly
all
set
up.
When
I
will
show
bob
and
brett
what's
up
there,
then
we
yeah,
I
will
ask
for
some
feedback
and
then
bob
vollen
volunteered
to
create
some
super
fancy
dashboards
for
us.
D
C
D
D
Well,
there's
like
I
said
I
set
it
up
at
the
moment.
We
have
the,
I
don't
know
about
grafana,
but
it's
probably
the
same.
The
prometers
is
just
a
custom
resource
definition,
so
you
can
just
set
up
this
all
the
times
as
many
as
you
want.
So
as
long
as
space
along
the
customs.
D
D
D
E
D
D
E
D
C
A
C
A
D
Which
yeah?
They
look
very
very.
E
D
Yeah
I
had
this
for
a
long
time.
Actually
I
started
this
at
the
beginning
of
december
because
I
just
created
all
these
constants.
You
know
because
I've
seen
all
the
environment
variables
used
everywhere
and
I
get
kind
of
in
my
real
life,
I'm
not
the
most,
what
you
say
not
ordinary.
If
everything
is
in
the
right
place,
I
cannot
come
up
with
the
word.
D
A
D
Nervous,
you
know
if
I
see
too
many
references
and
stuff
like
this,
so
this
is
just
well,
so
I
went
through
the
documentation
and
just
picked
everything
up
there
and
put
it
in
constants
and
there
were-
and
this
I
think
one
resolver
thing
and
then
was
nearly
on
my
hard
drive
for
two
months,
and
so
I
just
picked
it
up
and
cleaned
it
up
a
little
bit.
So.
E
D
And
now
I
could
remove
also
some
stuff
from
you
know
this
header,
passing
from
from
the
exporters
and
some
stuff
which
actually
the
defaults
are
all,
and
I
didn't
use
it
everywhere,
because
you
could
actually
get
rid
of
a
lot
of
defaults
but
yeah.
We
can
still
work
on
this.
You
know
what
I
mean,
because
the
defaults
are
all
in
one
consonant
or
one
of
these
interfaces
and
they
are
actually
called
when
there's
no
default.
So
that's
how
it
works.
Do
you
know
understand
what
I
mean?
Yes,
yeah
yeah
yeah
yeah?
Okay.
D
So
if
you
don't
put
the
first,
if
you
don't
put
a
real
user
default,
it
will
look
for
the
the
one
which
is
in
the
interface
basically,
but
of
course
not
every
environment
variable
is
a
default,
just
just
a
bunch
of
them,
but
yeah.
Yes,.
D
And
then
maybe
we
can,
if
we
do
work
on
the
config
stuff
like
this,
we
can
maybe
move
well
actually
yeah.
We
can
move
the
axes
completely
out
of
all
the
exporter,
stuff
and
stuff
like
this.
You
know
and
have
basically
just
construct
your
arguments
or
something
like
this
and
then
have
to
have
the
factory
or
whatever
part,
take
care
of
buying
unfair
rules.
So.
E
C
E
Just
I
was
just
gonna
note
something
he
said
there,
oh,
so
if
exporters
are
in
contrary,
but
the
default
configuration
for
the
exporters
is
in
core.
Is
that
backwards
dependency?
E
What
you
can,
can
you
ask
again?
I
didn't
really
yeah.
So
our
exporters,
the
code
for
the
export,
is
losing
the
control,
but
if
the
default
configuration
for
the
exporters
is
in
core,
are
we
trying
to
reduce
or
remove
instances
of
that
so
that
the
there
is
not
a
dependency.
D
No,
no,
what
I
mean
is
that
in
the
otlp
exporters
or
something
they
know,
they
know
about
what
the
environment
variables
you
know,
because
they
they
just
pull
stuff
out
of
and
and
vars,
are
basically
global
scopes.
So
from
my
experience
is
always
cleaner
to
have
something
it
doesn't
matter
where
it
is,
it
could
be
a
dedicated
factory
for
all
the
for
every
exporter
or
some
other
place
or
configuration
or
dependency
ejection
or
whatever,
but
some
place
out
of
outside
of
the
exporters
which
actually
pulls
the
values
from
the
environment
variables.
D
You
know
what
I
mean
absolutely
yes,
so
then
we
so
that
the
class
just
has
constructor
arguments
or
whatever
doesn't
matter
but
really
doesn't
know
what
the
infor
is
and
then
some
some
somebody
else.
Another
class
basically
just
takes
care
of
it,
and
I
think
that
will
be
easier
to
handle
than
also
when
how
we
can
the
configuration
when
we
want
to
mix
user
input
and
and
files
and
defaults
and
all
the
stuff.
You
know
we
have
a
central
place
where
we
can
understand.
E
Yes,
although
it's
being
merged
or
gone
make
sure
I
understand
that
that
pr.
D
I
can
explain
it
to
you,
it's
not
so
it's!
It
looks
very
basic,
it's
better!
If
you
look
at
in
the
code,
which
is
much
not
in
the
in
the
pr
because
yeah
I
think
so.
D
Yeah,
basically
take
a
look
at
this.
What
you
might
call
it
the
and
I
wrote
this
class,
and
I
can
remember
what
it's
called
my
brain:
there
are
not
many
classes
in
there
resolver.
This
resolver
class
basically
takes
care
of.
You
know,
picking
up
the
defaults
and
looking
at
this
stuff.
So
that's
basically
the
the
where
all
the
strings
go
together:
okay
and
the
other
one.
This
accessor
is
basically
just
on
top,
so
that
you
can
get
the
specific.
D
And
then
there
is
four
parsers,
but
basically
you
know
one
for
boolean
one
for
the
map,
thingies
one
for
lists
and
another
one
for
yeah.
I
forgot,
so
that's
basically
all
and
then
I
just
a
lot
of
the
other
stuff
is
basically
just
you
know,
using
all
the
stuff
in
the
other
classes
and
cleaning
up
a
little
bit.
E
Yeah,
I
think
I
get
about
halfway
through
it
last
time.
I
think
I'd
seen
all
of
the
you
know
all
of
the
most
interesting
bits
already.
A
D
E
E
D
Then
everybody
have
a
nice
day
morning
afternoon
night,
wherever
you
are
in
the
world
and
we
will
talk
at
least
next
week.
Hopefully.