►
From YouTube: 2022-01-13 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
Hey
timo
good
morning,
good
afternoon,.
C
So
my
my
appointment
got
changed,
so
I
can
stay
for
the
whole
meeting
now
and
I
could
facilitate.
C
Oh,
it's
just
like
that
would
be
like
being
the
meeting
leader
like
talking.
A
C
Mc
title
that
sounds
not
that
prefix
always
is
funny
to
me,
because,
because
of
mailchimp.
A
C
D
Oh,
hey:
everyone:
hey
siddique!
How
are
you
hello,
yeah
doing?
Okay,
I've
been
trying
to
figure
out
how
sessions
and
php
work
or
how
we've
done,
and
it's
not
been
going
well
well,.
C
A
And
well
usually,
if
you
use
a
framework,
something
like
this,
they
will
handle
this
out
of
the
box,
but
the
most
important
gotcha
is
don't
send
anything
to
the
output
before
you
started
the
session
and
then
what
what
were
the
problems
you
had.
D
D
Yeah
like
so,
we
use
like
cake
for
most
of
the
applications
and-
and
I
found
we
have
some
config-
that
points
to
like
a
database
table.
But
I
don't
see
the
things
I
expect
to
see
there
and
then,
like
I,
looked
at
the
php
documentation
and
found
we
have
this
like
the
default
set,
so
that
if
we
were
using
the
php
one
would
be
going
to
the
accessions
folder
thing,
but
not
seeing
anything.
There.
A
D
D
Yeah
yeah,
I
think
I
just
haven't,
found
any
code
that
would
mention
that
yeah.
D
D
If
I'm
looking
backwards,
okay,
yeah,
I
see
a
lot
of
that
in
the
config
file.
Now,
okay,
cool
yeah
check
that
next
yeah
it's
been
it's
been
an
interesting
morning.
A
A
It's
usually
you
can
mess
up
stuff,
but
usually,
if
you
use
like
the
framework
library,
it's
really
handled
by
them,
because
it's
such
a
low
level
thing
that
people
normally
don't
want
to
bother
with.
A
D
Is
one
of
our
principal
engineers
made
a
brand
new
like
identity
service,
and
I
was
told
yesterday
to
like
validate
it
from
a
security
standpoint,
and
so
these
are
like
two
things:
I've
never
done
before
so
trying
to
flying
by
the
seat
of
my
pants
at
the
moment.
C
It's
it's
a
really
great
way
for
you
to
gain
more
exposure.
C
So
I
guess
we
can.
I
guess
we
can
get
going.
I'm
not
expecting
anybody
else
today,
how's
everybody
doing
we're
doing
all
right.
Anybody
have
any
blockers
or
interrupts
right
now
before
we
get
started
on
open
discussions.
D
I
think
the
only
thing
for
me
on
that
pr
that
I
revamped
for
sakti.
This
is
like
what
to
do
with
the
code
that,
like
is
dead
code,
but
I'm
not
sure
if
it's
like
relevant
in
the
future
and
kind
of
things
so
yeah.
I
think.
C
D
C
Yeah
I'd
say:
do
what
did
I
say
I
said:
make
sure
you
could
just
you
definitely
should
be
able
to
delete
that
codec
utility
and
leave
a
second
block
comment
to
that.
If
somebody
needs
to
use
it,
they'll
fix
it
back
up,
but
as
long
as
the
test
pass
and
you're
making
progress
in
the
direction
you
are,
I
don't
think
that
that's
really
a
huge
boulder.
D
Okay,
cool
yeah
I'll
go
ahead
and
do
that
then
I
think
brett
also
said
he
was
able
to
try
out
the
brand
functionally
somehow
and
he
said
it
was
working
okay
as
far
as
he
could
tell.
So.
D
A
Already
made
the
button
clean,
clean,
green
for
us.
A
D
C
C
D
C
D
It
yeah
definitely
yeah
huge
props
to
sakti
for
doing
all
the
like
main
work.
I
guess
so
yeah
I
haven't
been
able
to
get
a
hold
of
him,
but
yeah.
A
It
doesn't
have
to
be
perfect
from
the
first
pr.
D
A
D
And
I
think,
for
what
I
want
to
try
next
is,
is
getting
that
one
file
with
the
commented
out
code
up
to
spec
compliance,
so
just
kind
of
going
through
the
other
languages
comparing
to
the
spec
as
well,
and
then,
probably
after
that,
there
should
just
be
one
or
two
pr's.
I
think
to
get
the
http
support
and
I
don't
think
that
should
be
a
huge
lift
because
timo
you
already
like
made
a
generic
way
to
do
that.
So
do
what
a
generic
way
to
have
like
an
http
exporter
right.
A
All
the
stuff
at
the
moment,
but
nearly
finished,
and
actually
I
was
a
little
not
waiting
on
this,
but
this
will,
when
this
prsmerge
will
help
me
with
figuring
out
what
the
best
way
in
abstracting
things,
because
then
I
got
everything
one
code
base
and
I
can
look
into
you
know
what
could
be
an
interface.
What
could
be
a
generic
class
with
you
know
extracting
stuff
and
refactoring,
so
it's
gets
a
little
bit
better
maintainable
and
yeah.
So.
A
I
always
forgot
how
we
call
these
classes,
the
ones
a
converter.
I
remember
and
stuff
like
this
yeah,
but
I'm
not
quite
sure,
and
I
will
create
a
pr
and
people
can
tell
what
they
think
about
another
solution.
A
Another
thing
once
this
is
merged,
I
will,
when
I
have
time
to
look
into
extracting
off
the
thrift
and
the
protobuf
files.
So,
as
we
talked
about.
D
A
Week,
yeah
so
that
we
have
them,
maybe
in
a
different
repository
or
at
least
get
have
them
generated
automatically.
But
this,
as
I
mentioned
last
week,
will
be
easier
if
they
are
in
a
separate
repository,
because
that
we
don't
need
to
the
goal
is
to
not
have
any
one
of
us
need
to
change
anything
there.
So
it
just
gets
checked
for
updates
once
a
day
and
then
it
automatically.
A
We
publish
the
package
with
this
one
and
because
we
are
aiming
if
we
publish
all
the
packages
from
the
contract
repository
and
the
main
repository
then,
and
the
goal
is
to
have
all
those
packages
with
the
same
version
number.
So
they
require
always
the
same
version
and
that's
easier
to
handle
a
lot
of
projects.
Do
it
like
this
and
they
would
collide
a
little
bit
with
the
generated
packages
because
they
would
have
a
version
number
of
of
the
upstream
repository
and
would
be
much
easier
if
they
would
be
separated.
A
B
Yes,
in
the
last
segment,
we
discussed
that
we
will
go
through
usage,
use
a
research
document
and
see
if
anything
is
outdated
or
anything
has
changed.
My
issue
is
225.
B
It
was
part
of
like
it
was
somebody
raised
it
from
this
user
research
feedback
only
so
currently
I
am
blogged
on.
I
like.
I
don't
clearly
understand
the
scope
of
this
issue
so
till
now
I
just
set
up
the
open
elementary
agent
with
the
laravel
and
yeah.
I
just
set
it
up
now.
I
wanted
to
make
changes
for
this
issue
and
I'm
clear.
A
All
right,
so
did
you
sorry.
C
225.,
okay,
so
have
you
see
if
you
read
through
the
spec
for
the
auto
instrumentation
pieces?
I
think
that's
what
this
is.
Yes,
yes
and
they're
much
if
the
specs
different,
then
we
should
definitely
follow
this
back
rather
than
whatever's
in
this
ticket.
C
C
Sure
yeah,
if
you
said
the
spec,
is,
has
a
little
bit
different
of
an
output
than
the
user
research
document.
Could
you
potentially
use
the
spec
to
try
and
accomplish
this?
This
issue
sure
actually.
B
I
am
facing
issue,
I
understood
the
specs,
but
I'm
just
having
a
difficult
time
implementing
them
in
the
code
and
like
probably,
if
any
other,
open
telemetry
agent.
If
there
is
any
example
from
that
like
python
or
go
or
any
other
agent,
if
any
pr
is
raised
or
pr
that
I
can
take
it
as
a
reference.
C
I
don't
so
I
don't
think
we,
I
don't
think
we
have
an
agent.
Currently
we
just
have
our.
We
just
have
the
open,
telemetry,
php
instrumentation
and
then
the
collector
is
acting
as
our
agent.
We
have,
we
don't
have
a
separate
standalone
daemon.
Yet
I
don't
know.
If
that's
I
don't
know
where
that
is
in
our
priority
list,
but
I
know
that
that's
not
currently
a
thing.
B
A
A
Yeah
regarding
the
auto
instrumentation,
I
actually
have
something
in
the
work
which
I
was
I'm
tinkering
around
since
over
a
month
but
yeah
I
will
publish
it
when
once
it's
more
or
less
worked
out
how
this
will
work,
but
in
essence
I
mentioned
this
and
the
ticket
for
creating
an
extension.
So
there's
this
view,
ops
extension
we
could
use.
So
we
could
basically
monkey
patch
other
libraries.
So
we
don't
have
to
create
a
complicated
solution
for
that,
but
yeah.
A
I
will
show
you
and
show
you
guys
once
it's
a
little
bit
more
ready
to
be
shown.
A
No
problem:
this
is
my
fun
part
of
coding,
not
so
much
coding
against
specifications.
E
Yeah
actually
298,
actually
I'm
sorry
yeah
yeah.
So
like
I
had
some
talk
with
team
teemo
about
it
so
issue
like
we.
It
was
basically
deals
with
the
creation
of
event
handlers
so
that
we
could
gain
information
about
some
extra
metrics
and
traces.
The
specifics
like
it
was
discussed
that
specifics
have
not
been
decided.
Tell
us
yet,
but
we
like,
I
started
working
on
the
api
part
and
how
do
we
get,
but
my
contention
point
was
like
we
merged
the
logging
api
for
10
to
15
or
two
weeks
back
right.
E
A
Yeah
your
question
is
clear,
but
the
answer
is
difficult
because,
let's
say
the
specification
is
not
very
verbose
on
this
point:
what's
actually
to
be
exposed
there
or
what's
to
be
handled
this
on
this
mindset.
A
E
A
A
E
A
Yeah
self-diagnostic,
so
I'm
not
quite
sure,
so
I
guess
the
idea
is
that
we
use
the
tools
we
have
to
to
get
metrics
about
the
library
itself.
So
that's
what's
meant
for.
C
Yeah
yeah,
I
believe
that's
me
correct.
A
I've
read
the
the
specification
on
this
part
over
and
over
again,
it's
not
really
clear
to
what's
actually
meant
about
this,
so
I
think
we
can
just
try.
This
is
rather
complicated
issue
because
using
stuff
you
know
it's
like
writing
a
programming
language
in
the
language.
You're
writing
it.
You
know
writing
c
and
c,
or
something
like
this.
So
I'm
it's
not
really
clear
and
there
could
be
like
feedback
loops
which
could
be
created,
which
you
know
this.
A
C
Yeah,
the
metrics,
the
metrics
that
the
metric
spec
is
about.
I
think
they're
planning
on
going
stable
relatively
soon.
I
know
they're
in
beta
now
and
they're.
They
said
they
were
planning
on
doing
a
monthly
release
cadence.
So
hopefully
something
good
will
come
in
january
or
february,
but
I
think
that
it's
not
imminent.
A
E
A
E
Also,
one
point
to
mention
that
it
would
be
getting
the
information
or
the
metrics
about
the
library
itself,
isn't
that
the
same
logging
api
also
doing
it
right.
A
D
A
E
A
Yes,
yeah
they're,
both
quite
related,
I
would
say
so
yeah
but
yeah
just
give
it
a
try
and
then
we
can
get
some
feedback
from
us.
So
we
can
see
how
it
works
out.
So
I
would
not
expect
any
anybody
to
come
up
with
the
perfect
solution.
A
First
of
all,
because
this
is
also
quite
opinionated,
because
there's
no
because
the
specification
is
not
really
clear.
So
probably
people
have
different
opinions
and
stuff
or
we
can
just
see
how
it
works
out
at
the
end
of
something
you
know,
people
actually
use
it
if
is
actually
useful
and
so
on.
So
just
give
it
a
try
and
then
we
can
see
what
we
can
do
with.
C
Cool
we're
we're
also
very
happy
to
like
review
your
prs
and
give
you
relevant
input
if
it,
if
that
will
make
you
feel
more
comfortable,
trying
to
come
up
with
a
solution
like
you
must.
I
think
this
is
like
unbelie,
it's
unbelievably
opinionated,
so
we
all
have
opinions
and
we're
happy
to
give
them.
C
A
Okay,
well,
I
can
come
off
with
more
stuff,
but
no,
I
wasn't
really
prepared
for
the
stuff,
so
the
issue
I'm
working
on
is
still
the
kubernetes
setup
and
k3s.
So
I
will
give
you
an
update
on
that
front
is
okay.
I
don't
know
if
front
is
a
little
correct.
A
Yep,
okay,
so
the
basic
setup
for
k3s
is
done
and
it's
can
be
repeated
with
the
script,
so
we
can
could
just
recreate
the
whole
server
and
set
it
up
again.
However,
they
have
been
while
I
was
going.
I
had
have
been
some
gotchas,
so
I
had
to
work
on
some
stuff
and
try
and
fix
it.
A
So
the
first
thing
was
I
mentioned
in
our
chat
that
I
was
aiming
to
use
linker
d,
so
just
to
give
a
general
context,
people
don't
know
what
I'm
talking
about
so
link
id
and,
for
example,
istio
are
service
meshes
and
the
interesting
part
for
us
not
to
go
into
detail.
What
the
service
mesh
is.
That
you
can
probably
find.
A
Is
it's
not
easy
to
answer
because
they
do
a
lot
of
stuff,
but
the
interesting
part
for
us
is
that
they
usually
offer
a
way
to
start
tracing
and
trace
stuff
through
different
micro
services,
for
example.
So
that
would
be
a
way
for
us
to
test
stuff
like
this.
They
have
other
things
you
can
traffic
splitting
and
all
the
stuff.
You
can
look
up,
but
this
would
be
the
most
interesting
part
since
we're
using
ks3,
which
is
at
the
moment,
which
is
like
a
lightweight
kubernetes
distribution.
A
Point
where
the
crest
from
outside
come
into
the
cluster
and
then
we'll
get
distributed
to
certain
services,
so
it
basically,
it
usually
has
a
lot
balancer
and
then,
with
that,
will
forward
a
certain
request
for
whatever
patterns
you
have
to
certain
services.
A
A
To
get
to
my
point
so
link
id
is
a
service
mesh
which
doesn't
have
an
opinion
on
an
english
controller
and
since
k4s
comes
with
traffic
as
an
inverse
contour.
That
would
be
a
good
solution
to
use
this
one.
However,
it
didn't
really.
It
was
very
complicated
to
get
this
working
because
we
have
to
use
a
certain
application
or
actually
it's
a
network
solution
which
is
called
calico
for
joining
two
servers
and
they
didn't
work
really
good
together.
So
basically,
the
services
couldn't
be
created
because
there
wasn't.
A
But
I
might
just
kick
out
traffic
for
that
reason,
and
just
use
traffic
has
some
very
interesting
parts
which
you
can
maybe,
if
you're
interested
look
into,
because
you
can
create
routes
and
stuff
like
this
in
a
lot
easier
and
better
way
than
usually
within
this
controller.
However,
so
I
maybe
just
reworked
that
so
there's
still
some
gotchas.
A
Certificates
and
all
the
stuff
and
the
test
will
test
service
which
just
you
know,
puts
out
in
static
action
alpha
and
which
gets
routed
and
all
this
is
set
up,
but
there's
still
some
gotchas.
I
have
to
work
on
also
there's
a
little
bit
of
a
problem
with
actually
joining
another
k3s
worker
to
the
master.
I
know
master's,
not
a
word
which
should
be
used
anymore,
but
that's
how
what
it's
called
yeah.
So
there
are
still
some
gotchas.
I
have
to
work
on,
and
but
most
of
the
stuff
is
really
done.
A
So
that's
mostly
getting
some
things
of
the
way,
and
that's
often
not
that
I
do
so
much
setup
of
kubernetes
in
in
my
work.
But
privately
I
will
do
and
once
you
add
more
and
more
stuff,
then
you
will
figure
out
what
things
don't
really
work
with
others
and
since
kubernetes
doesn't
really
have
a
good
package
manager,
helm,
it's
not
really
like
a
pickup
launcher
and
that
it
can,
you
know,
can
handle
conflicts
and
stuff
like
this.
So
it's
always
a
lot
of
trial
and
error.
A
I
hope
to
be
here
be
done
with
that.
At
the
end
of
the
week,
so
then
we
have
a
setup
with
all
the
tracing
stuff.
You
need
so
different
tracing
collectors,
back-end
for
apms
application
process,
monitoring
and
so
on.
So
then
we
can
actually
use
this
stuff
to
test
our
library
to
do
end-to-end
tests
to
the
performance
test,
to
host
some
demo
applications
and
so
on.
A
Yes,
it
is
so
it's
not
really.
I
like
this
stuff
so
because
the
box
I
wanted
to
say
devops
is
usually
more
my
hobby.
So
it's
not,
though
I
use
my
hobby
stuff
and
work
as
well.
So
it's
not
that
it's
you
know
something
I
wouldn't
be
bothered
to
do.
So
that's
a
reason.
A
E
A
A
Can
figure
out
an
appointment
with
brett
and
you
so
I
can
walk
you
to,
but
first
there's
some
things.
I
don't
want
to
do
open
because
I'm
maybe
a
little
bit
security
related.
But
after
that
is
done
we
can.
I
can
do
a
presentation
here
on
the
meeting
so
where
not
some
security
related
stuff
is
involved,
another
general
setup
artwork,
so
you
can
expect
if
you
want
to
use
to
assault
him
as
a
serious
person.
A
Yes,
but
that's
all
this
stuff
is
the
reason
why
I
couldn't
be
so
involved
in
the
last
week
into
anything,
really
php,
library
and
stuff,
so
I
had
to
take
a
step
back
and
could
not
walk
into
pr.
So
at
that
point,
I'm
not
really.
A
Yeah
also,
the
provider
also
metal
metal,
phoenix
metal
is
kind
of.
I've
never
worked
with
that
before,
so
they
have
there's
some
gotchas
in
there
as
well,
which
I
have
to
learn,
but
yeah,
that's
kind
of
interesting
and
it's
actually
quite
interesting.
A
nice
provider
for
kind
of
dynamic,
bare
metal
servers.
That's
interesting.
C
All
right,
no
news
is
good
news.
Give
some
people
some
time
back.
Well,
thanks
for
everybody
for
attending
and
we'll
see
you
on
the
internet.
Let
us
know
if
you
have
any
questions
or
concerns
or
have
prs
to
share.