►
From YouTube: Keptn Documentation Office Hours - Jun 22, 2023
Description
Meeting notes: https://docs.google.com/document/d/156pqw4r_O6O7O-YDm9DfuuBUS3GvAXmvjjDyFEpbwUY/edit
Learn more: https://keptn.sh
Get started with tutorials: https://tutorials.keptn.sh
Join us in Slack: https://slack.keptn.sh
Star us on Github: https://github.com/keptn/keptn
Follow us on Twitter: https://twitter.com/keptnProject
Sign up to our newsletter: https://bit.ly/KeptnNews
B
A
So
I
was
unable
to
attend
the
last
mid
since
due
to
kcd
Mumbai,
but
yeah
I'm
on
it
right
now,.
A
Yes,
I
I
had
to
go
into
the
issue
section
and
came
up,
came
across
track.
Klt
migration
guide
so.
B
B
And
I'm
going
to
we
finally
made
zero
eight
zero
yesterday,
just
before
the
community
meeting
so
and
I
I
found
one
thing,
I've
got
to
put
on
that
and
then
I'm
gonna
try
to
get
the
the
basic
structure
of
the
migration
guide
merge.
So
we
can
work
against
that
for
the
migration
guide
sure
there
and
that.
A
B
B
A
So
I
I
went
through
the
whole
issue
and
the
issue
clearly
I
can
showcase
or
what
are
the
problems
that
we
are
facing
regarding
the
migration
one
so
over
there
I
think
there
are
five
issues,
six
issues
so
I
I
can
start
contributing
to
each
of
them.
Okay,.
B
Yeah
and.
C
B
C
B
B
B
B
Shoot
we've
got
to
get
you,
it
seems
like
if
you
put
a
we've,
got
it
when
you're
part.
When
we
get
your.
What?
How
far
are
you
from
applying
for
membership
in
the
com
in
the
community.
A
Four
four
of
my
PRS
has
been
merged
and
Android,
and
you
had
recently
to
will
be
able
to
sponsor
me.
If
I
can
apply.
Yes,
yeah
so
should
should
I
apply
for
the
member
post.
Why.
B
Don't
you
and
in
the
meantime,
if
you
comment,
it
looks
like
if
you
comment
an
issue
and
say
I
want
to
work
on
this,
then
we
can
assign
it
to
you
and
once
you're
a
Community
member.
Of
course
we
can,
but
but
if
you'd
like
it
we'd
like
to
have
you
definitely.
A
A
Sure
we'll
make
up
issue
on
the
community
section
of
Captain
yeah.
A
So
for
since
the
upper
two
issue,
since
you
were
working,
I
went
through
that
the
so
I
decided
to
start
working
from
the
migrating
one.
B
You're
gonna
have
to
get
creative
here.
Let's
see,
let's
go
over
and
looks
at
the
V1.
A
A
B
B
B
We
should
make
that
as
a
okay,
because
what
we
have,
let's
see,
didn't
make
it
separate
one
I
should
make
a
let's
do
right
now.
C
B
Your
data
sources,
that's
going
to
be
like
Prometheus,
datadog
Dynamics,
are
associated
as
Integrations.
B
C
B
And
so
here,
oh,
are
they
gonna
I,
wonder
if
it's
yeah
they're
gonna
be
okay,
so
there
well
there's:
let's
go
for
Prometheus.
A
B
Right
well,
this
is
now
Captain
V1,
it's
not
a
crd
in
Captain
B1,
where
they're
converting
from
that's
migration.
It's
tough,
it's
they're,
gonna,
have
to
install
this
Prometheus
server
and-
and
it
looks
like
it's
already
saying
what
port
is
on,
etc,
etc.
B
Now
for
life
cycle
toolkit
in
a
way,
I
think
it's
actually
simpler,
but
the
migration
will
be
a
little
bit
tricky,
but
but
I
think
it
it'll
be
I.
Think
it'll
be
an
easier
first
thing
for
you
to
work
on
yeah.
B
You
will
go
ahead,
you
will
install
Prometheus
and,
and
it's
well
set
up
and
the
people
you
know
with
klt.
You
can
have
three
Prometheus
servers
up
there
for
different
purposes
and
two
or
three
data
you
have
all
you
want
and
you
can.
They
can
be
a
mix
of
all
of
them
and
then
for
each
one
like
each
of
your
Prometheus
servers
and
you're
gonna
have
to
have
one
Prometheus
server,
because
open
Telemetry
needs
it.
B
Now
what
I
don't
know
the
first
question
I
have,
but
we
you
can
ask
the
engineers
can
I
use
that
Prometheus
for
anything
else
or
does
that
have
to
be
dedicated?
Only
for
open
Telemetry
that
I
don't
know
one
of
these,
where
you
would
give
it
a
name,
and
so
this
would
be
like
QA
Prometheus.
Maybe
it's
the
Prometheus
that's
going
to
take
stuff.
B
That's
gonna
do
metrics
for
the
QA
deployment
or
something,
but
you
give
it
a
unique
name
and
say
what
name
space
it's
in
and
then
say
type,
whether
it's
Prometheus
or
whatever,
and
specify
the
URL.
For
that.
A
B
Yeah
yeah
I.
B
But
I
think
I
think
that's
fairly
straightforward,
yeah
and
I.
Don't
know
somebody.
Maybe
some
buddies
are
working
on
additional
metric
providers
and,
at
the
same
time,
instructions
for
how
you
create
a
new
one.
I.
A
B
Man
on
the
correct
page,
but
there's
a
missing
something
here
and
I
will
show
you
and
it's
not
a
wonderful
little
chapter
but
under
implementing.
We
have
a
Guide
chapter
about
Captain,
metrics,
okay,
where
a
lot
of
it
is
stuff
that
somebody
else
had
written
I,
just
pulled
it
together
and
wrote
a
little
bit
of
up
front
and
we've
got
some
other
stuff
so
feel
free.
This
is
Open
Season
Two,
while
you're
working
on
like
reporting
this.
You
might
have
information
to
add
here,
oh
and
another
goal.
B
C
B
In
you
know,
for
the
when
you're
talking
about
porting
say
for
each
data
provider
that
you
want,
you
need
to
set
up
a
captain,
metrics
provider
and
make
that
a
link
to
this
reference
page.
So
they
go
over
there
for
all
the
dirty
details
and
I
think
even
there
we
give
them
yeah.
Here
we
give
them
an
example
of
the
naming
stuff
that
you
might
have
Dev
Prometheus
for
the
Prometheus.
It
monitors
the
development
deployment
and
QA
Prometheus
Etc.
B
So
so
you
know
what
I'm
so
a
lot
of
a
lot
of
the
migrating
stuff
should
be.
It
may
end
up
being
pretty
short
because
it's
going
to
be
like
fill
out
one
of
these
and
do
one
of
these
and
that
sort
of
stuff.
C
B
So,
but
look
into
it
and
talk
to
it,
you
know
right
and
yeah,
because
it
would
be
wonderful
it'll
when
we
have
the
instructions
up
as
I
understand
it.
Creating
like
doing
an
integration
in
Captain
V1
for
a
new
service
was
a
lot
of
work
and
for
klt
as
I
understand
it.
It's
like
20
lines
of
code
or
something
it's
much
simpler
and
more
straightforward.
B
Oh
one
of
the
big
I
need
to
write
up,
but
one
of
the
big,
probably
I,
think
I.
Think
people
will
now
think
it
was
a
little
bit
of
a
mistake.
I
don't
know,
but
a
big
Paradigm
switch
is
that
Captain
V1
did
their
own
configuration
files
for
everything
they
didn't
use
the
resource.
C
C
B
And
just
so,
you
know
and
I
I'm,
not
sure
how
this
is
going
to
work
out.
If
you
do
this
one,
the
next
one
will
give
you
is
observability
and
then
because
one
thing
they've
done
is
like:
if
you
have
a
metric,
let's
say
you
have
a
metric,
that's
reporting
how
many
CPUs
are
are
alive.
B
If
you
want
to
do
an
evaluation,
you
will
use
that
same
metric.
So
your.
B
Will
be
if
number
of
CPUs
is
greater
than
eight
or
whatever,
then
we
pass
so
so
that
we
started.
If
you
look
and
see
the
first
at
the
very
beginning,
they
had
a
separate
evaluation
provider.
So
you
had
something
like
this
and
it's
like
that's
nuts,
we'll
just
use
the
metrics
for
everything.
B
B
Got
too
many
tabs
I
don't
know
where
I
am,
but
then
there's
your
metric,
and
so
here
you
give
your
metric
a
name
and
the
application,
the
namespace
it's
in
and
which
provider
that
name
that
you
assigned
in
the
provider.
And
then
here
you
put
your
query
as
it
would
come
from
Prometheus
or
dynatrace
or
whatever,
and
you
specify
how
often
you
want
them
to
fetch
that
information.
B
B
Something
else
that
you've
been
working
with
these.
Do
you
know
what
the
API
references.
B
A
B
B
But
in
here
we
see
okay,
so
I
want
to
see
Captain
metrics
provider.
So
here
are
my
fields
and
then
they
start
that
what
I
said
the
the
engineers
thought
that
this
would
be
fine
for
users.
This
was
all
they
needed
and
I
thought
they
needed
something
else
yeah
once
you
understand
it,
it's
full
of
information,
so
okay
version
string.
We
saw
that
now,
but
for
metadata,
so
we've
got
something
else
here:
another
one.
That's
called
object,
metadata.
B
Over
here-
and
in
this
case,
that
is
a
kubernetes
one
I
my
I
have
and
the
spec
here
kept
in
metrics
provider,
spec
I
think
that's
going
to
be
one
of
ours,
let's
see
yeah
and
there
you
have,
and
that
was
what
I
just
came.
I
was
just
over
here.
Looking
I
thought
you
know
and
then
so
I'm
looking
I,
don't
see
this
optional
thing
that
that
Nikki
just
did
for
this,
but
then
I
go
into
secret
key
selector,
so
you
just
have
to
keep
drilling
down.
A
B
A
B
I
found
the
optional
it's
a
Boolean
and,
and
it
has
now
for
the
kubernetes
they're
done.
The
the
commentary
that
is
on
the
captain
ones
is
something
that
we
write.
The
engineers
are
supposed
to
write
it.
Sometimes
they
do
and
sometimes
they
don't,
and
sometimes
it's
better
than
others.
I
just
saw
one.
That's
like
no
and
I've
gotta
I
forgot
the
path
name.
I
need
to
get
the
path
name
again
for
where
you
go,
but
any
of
these.
B
C
C
C
C
C
B
B
B
A
B
B
Okay
and
and
you're
gonna
go
ahead
and
apply
for
Community
membership
right.
A
Yeah,
yes,
yes,
we
will
also
talk
to
we'll
make
a
prpi
regarding
that
and
the
community
section
yeah.
B
You
do
the
pr
and
propose
it,
and
what
do
you
need?
Two
sponsors.
A
You
are
the
one
of
the
sponsors
and
otherwise
are
Argent
I.
Think:
okay,
yes,
also.