►
From YouTube: Keptn Community & Developer Meeting - Mar 15, 2023
Description
Meeting notes: https://docs.google.com/document/d/1y7a6uaN8fwFJ7IRnvtxSfgz-OGFq6u7bKN6F7NDxKPg/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
A
B
Hi
sure
thank
you.
Yeah
I'm,
Aaron
mascot
I'm,
with
the
developer
platform
team
at
brandwatch.
We've
recently
done
some
exploration
into
considering
Captain
as
a
viable
alternative
for
our
in-house
tool,
which
accomplishes
similar
goals
and
I
figured.
It
would
be
a
good
idea
to
attend
this
meeting
to
meet
some
of
the
guys
and
and
gals
making
the
magic
behind
the
scenes,
as
well
as
to
give
my
feedback
about
what
our
first
impressions
were
with
Captain
and
how
it
what
our
resolve
was
with
things.
A
So
welcome
thanks
for
providing
the
feedback
I'm.
Looking
forward
to
the
discussion
later
I
see
you
put
some
agenda
items
so
I
like
to
kick
it
off
with
a
small
announcement
about
the
public
roadmap
I
work
or
already
send
the
last
community
meeting
that
I
work
on
making
this
roadmap
a
bit
more
visible
for
everyone.
A
So
Community
feel
free
to
look
into
this
project.
It's
called
project
or
Capital
roadmap.
It's
pinned
on
top
of
the
organization
and
the
different
repositories.
So
should
be
fine
and
easy
to
look
for
that,
then
also
some
news
that
we'll
discuss
a
bit
more
in
details
later
with
Moritz
about
a
Docker
Hub
migration.
We
got
an
email
that
Docker
will
delete
our
repository
or
our
organization
in
30
days,
because
we
are
backed
by
a
company
and
they
don't
consider
us
an
open
source
project
by
default
or
they
are
getting
rid
of
their
free
tier.
A
A
A
We
work
a
lot
in
bumping
all
the
different
dependencies
and
being
get
rid
of
all
the
smaller
vulnerabilities
found
in
the
standard
core
Library,
and
we
will
most
likely
release
this
next
week
and
you
can
already
see
a
small
change
log
preview
about
features
and
Bug
fixes.
You
will
not
see
the
dependency
bumps
here
and
with
that
I
think
we
are
over
with
the
news
or
is
there
any
questions
from
the
community.
C
Yeah
you
already
tested
anyways
yeah
tokahap
is
sunsetting.
The
free
teams
plan
that
we're
currently
using.
So
that's
basically
a
plan
where
you
have
a
maximum
of
three
reusers
in
your
team
and
we're
using
that
for
the
cap
and
Docker
Hub
registry,
and
also
for
the
captain
F1
that
we
use
for
development
yeah,
as
Johnny
said
already,
we
are
gonna
I
think
apply
again
for
open
source.
We
tried
already
yeah
for
an
open
source
open
source
plan.
C
C
D
C
D
C
But
for
existing
users
that
have
deployed
already,
that's
that's
going
to
be
something
that
they
need
to
change,
so
we're
gonna.
Do
announcements,
we're
also
gonna
open
some
issues
for
all
the
community
maintained
services,
so
that
people
get
at
least
a
notification
for
that
and
can
update
on
their
services.
Yeah.
D
C
C
A
Yeah-
and
we
will
also
announce
this
in
the
announcement
channel
in
our
Slack.
C
Not
really
I'm,
already
working
on
it,
I
have
a
draft
PR
open
to
exchange
lots
of
variables
and
stuff
and
I'm
still
working
on
kind
of
a
small
script
to
fetch
or
the
most
active
contributors
of
all
the
rappers.
A
D
B
Sure,
thank
you
yeah,
and
thanks
for
keeping
this
form
public
and
open
I
really
appreciate
that
it's
nice
to
see
the
activity
going
on
yeah
so
before
I
jump
in
I'll
also
introduce
my
colleague
who's
also
joined
Lynn.
He
came
in
just
a
moment
ago.
We
kind
of
explored
this
together,
so
give
them
a
warm
hello,
hi
nice
to
meet
you
guys,
all
right.
B
So
essentially,
we
built
an
in-house
collection
of
tools
as
part
of
our
internal
developer
platform,
which
fills
gaps
in
environment
management,
mostly
visibility,
but
especially
with
deployment
management,
and
we
have
certain
paradigms
that
we've
been
following
that
largely
just
work
out
of
convenience
or
by
what
naturally
came
as
a
byproduct
of
the
tools.
We're
using,
which
specifically
were
kept
them
would
come
in
was
that
we
have
GitHub
actions
creating
the
artifacts
and
Argo
CD
continuously
deploying
them,
but
this
is
where
our
in-house
tool
comes
in.
B
B
B
Breaker
was
the
part
like
well
just
to
give
like
kind
of
like
our
trade-off
is
essentially
considering
if
the
benefits
included
by
Captain
can
outweigh
the
work
needed
to
replace
our
in-house
tool,
but
where
I
think
like
the
most
critical
thing
for
us
came
in,
was
in
this
philosophy
of
going
hard
with
the
deployments
and
the
rollouts
and
the
Cannery
type
style,
and
not
having
that
visibility
on
the
artifacts
would
kind
of
force
people
to
change,
how
they
think
of
things.
B
B
But
I
also
wanted
to
hear,
maybe
from
this
forum,
if
our
assessment
was
incorrect
or
if
maybe
we
missed
some
kind
of
an
opportunity
about
how
the
tool
can
be
used
because
it
really
seemed
like
this,
this
would
have
been
possible,
but
unfortunately,
at
least
we
didn't
figure
out
how
to
do
it.
So
that
was
kind
of
like
where
we
we
figured.
B
This
might
not
really
be
a
fair
thing
to
do,
but
then
also
it
kind
of
creates
some
warning
signs
when
we
see
that
there's
some
signs
of
really
like
this
isn't
like
early
stage
project.
You
know
it's
really
nice
that
you
got
the
incubation
stage
from
cncf.
B
However,
like
you
know,
I,
and-
and
this
is
not
like
no
offense
to
anybody
here-
you
guys
are
all
you
know,
making
some
great
stuff
happen,
but
what
we
did
see
was
that
there's
just
like
you
know,
this
extensions
and
stuff
haven't
been
maintained
for
over
a
year
in
a
lot
of
cases,
as
well
as
the
life
cycle
toolkit.
B
Is
it's
so
fresh
that
you
know
it
just
seemed,
maybe
a
little
bit
naive
to
think
that
we
could
take
something
so
early
and
use
it
as
a
replacement
for
hundreds
of
Engineers
right
so
anyways
I'll
leave
it
at
that.
I
would
love
to
hear
some
feedback
from
you
guys
as
to
kind
of
what
whether
our
thinking
is
clear
on
this.
A
So
or
to
start
I'll
say
yes,
it's
true
on
the
integration
page,
many
of
them
are
not
maintained.
Unfortunately,
it's
the
said
reality
of
Open
Source
projects.
A
Someone
come
up
with
a
very
nice
idea
how
to
integrate
with
their
own
tool,
and
then
they
forget
about
that
when
they
move
from
fancy
tool,
one
to
fancy
tool,
two,
we
are
still
keeping
them
out
in
the
wild
for
a
reason,
because,
since
tools,
the
tool
landscape
evolves,
you
might
want
to
provide
your
own
integration
for
your
own
tool
that
you
are
currently
using,
and
so
you
can
use
the
existing
code
base
as
a
reference
to
look
how
you
could
integrate
your
own
tool
into
captain.
A
For
the
terminology,
yes,
I
agree
that
sometimes
we
have
some
terms
that
are
a
bit
not
aligned
with
the
standard
understanding
of
that
specific
term
like
project.
What
is
a
captain
project?
How
do
you
relate
to
that
is
an
application
is
not
service.
Is
the
kubernetes
service
or
is
a
micro
service
in
terms
of
an
artifact
yeah
I
see
there
might
be
a
bit
of
a
learning
curve,
basically
for
the
life
cycle
toolkit
we
started
six
months
ago.
A
I
cannot
say
you
should
use
it
in
production
at
the
moment,
because,
depending
on
what
do
you
want
to
use
it
for,
because
some
functionalities,
such
as
bringing
observability
out
of
the
box
to
your
deployments,
that
I
would
say
you
can
use
the
lifecycle
toolkit
in
production
for
that's
very
well
tested
it's
the
first
thing
we
start
to
work
on
and
I
would
feel
confident
running
that
in
my
production
environment,
other
things
such
as
the
metric
server
to
scale
your
workloads
up
and
down.
A
It's
a
really
new
feature
that
we
just
recently
added
few
weeks
ago.
That
might
be
not
good
for
your
production
environment,
but
for
the
last
cycle
took
it,
we
wanted
to
have
multiple
different
things.
That's
why
the
name
toolkit
and
you
can
decide
how
to
use
it.
You
can
start
with
observability
and
just
use
that
possibly
to
see
what's
happening
in
your
deployment.
A
Then,
when
you
consider
the
maturity
of
the
life
cycle
handling
to
also
automatically
deploy
things
into
your
community
cluster
and
support
your
deployment,
then
you
might
want
to
change
it
and
started
up
this
new
feature.
When
you
see
that
also
the
metric
server
to
scale
your
workloads
automatically
works
fine,
you
can
start
to
add
that.
So
that's
why
we
use
the
name
toolkit,
because
you
can
pick
the
feature
and
start
to
use
that
specific
feature
dependently
from
the
rest.
B
That's
very
helpful
and
you've
addressed
most
of
kind
of
what
I
mentioned
it's
nice
to
to
have
that
and
I
guess
the
the
last
one
that
I
wasn't
addressed
would
kind
of
be
something
I'd
like
to
press
a
little
further
was
that
idea
of
the
promotion
flow
with
the
artifact
view
as
a
contrast
to
this
functional
philosophy
around
the
deployments?
Has
this
been
something
that's
been
considered,
or
it
kind
of
looked
like
I
was
seeing
your
roadmap
here
as
some
of
these
yeah
exactly
kind
of
there.
A
A
So
this
one
is
mostly
about
propagating
the
trace
ID.
So
sorry,
maybe
a
bit
of
context.
The
lifecycle
toolkit
Works
in
a
github's
fashion.
So
you
have
your
repository
with
from
manifest
I've,
been
deploying
to
the
kubernetes
cluster,
and
then
lifecycle
took
it
watch
for
them
and
create
traces
and
metrics
of
your
deployment
and
support
with
running
some
pre-task
or
post
task.
A
When
things
happens
inside
your
deployment
and
we
create
a
trace
for
this
deployment,
but
then
this
is
maybe
just
a
Dev
stage,
and
then
this
one
will
be
promoted
into
the
hardening
stage
and
in
the
hardening
stage,
would
be
nice
to
have
a
link
that
points
to
the
trace
from
the
dev
stage,
and
this
announcement
proposal
is
to
propagate
the
trace
ID
context.
So
you
can
have
open
Telemetry
that
monitors
every
single
stage
of
your
promotion
flow
from
one
stage.
A
E
A
No,
so
the
idea
here
is
exactly
about
this
promotion
part
well,
you're,
telling.
Okay.
If
everything
went
fine
in
your
previous
deployment,
with
the
concept
of
evaluation
that
we
have
also
in
that
lifecycle
toolkit,
which
is
similar
to
the
Quality
gate
in
Captain
B1.
If
every
lsli
and
SLO
are
fine,
then
you
want
to
promote
your
artifact
to
the
next
stage,
for
instance
from
Dev.
You
want
to
promote
it
to
staging,
and
then
you
can
specify
which
repository
Branch
path
or
the
Manifest
file
and
anything
else
to
run
this
promotion
in
agitops
fashion.
B
Yeah
nice
build
on
creating
more
like
options
and
functionality
around
the
promotion,
flow,
yeah,
so
and
I
guess
like
what
I
would
wonder
is
if
there's
a
way
to
have
like
or
anything
planned
for,
are
sort
of
like
like
a
table
view
of
like
a
history
or
like
previous
artifacts
that
could
be
reintroduced
into
the
promotion
flow,
for
example,
or
as
a
way
of
having
Captain
be
used
as
like,
like
you,
could
trigger
a
manual
step
that
would
initiate
a
rollback,
for
example,
that
isn't
part
of
the
active
promotion
like
not
just
a
rollback
to
the
previous
state,
but
maybe
multiple
States
is
this
something
that
could
be
possible.
A
We
do
not
have
this
yet
in
both
version
of
Captain,
the
lifecycle
toolkit
and
captain
captain.
That's
an
interesting
approach.
B
B
Fair
enough,
fair
enough
and
I'm
curious.
How
many
people
here
are
with
dynatrace.
D
B
B
And
you
know
what
what
I
do
want
to
say
is
a
possibility
on
the
table.
Is
that
what
would
be
worth
for
us
is
if,
in
the
future,
we
can
find
something
that
achieves
parity
right,
then
we
would
rather
make
contributions
to
a
project
like
Captain,
then
to
just
isolate
ourselves
with
our
own
internal
tool
right,
but
with
where
we
are
now.
B
You
know
that
we're
gonna
have
to
keep
going
with
with
what
we're
already
working
on,
but
okay,
but
that's
really
cool
I'm,
happy
that
at
least
we
could
kind
of
bring
up
this
kind
of
an
idea,
because
I
think
this
was
really
like
the
the
key
thing
about
having
that
visibility
on
the
artifacts
right.
B
If
Captain
could
be
used
as
a
way,
maybe
some
sort
of
using
artifactory
as
an
input
or
kind
of
like,
like
a
way
of
looking
at
previous
States
right
that
that
would
that
would
kind
of
tie
it
all
together.
But
okay,
anyways
I,
don't
want
to
take
too
much
of
the
the
time
on
the
agenda
here.
I
know:
there's.
A
A
That's
also
the
promotion
aerobic,
it's
a
very
interesting
concept.
I
would
suggest
maybe
to
have
a
read
that
this
Captain
management
proposal
and
provide
your
two
cents
on
this
topic.
B
Okay,
cool
yeah,
I'll,
I'll
message:
Thomas
I
was
actually
kind
of
hoping
he'd
be
here
today.
No.
A
Fortunately,
cannot
it
won't
be
able
to
be
here
today,
but
we
have
also
community
meeting
I,
don't
know
where
you're
based,
but
next
Wednesday,
then
in
the
morning
Central
European
Time.
B
Oh
okay,
where.
D
B
Okay,
but
so
sorry,
what
was
the
the
meeting
you're
proposing
here.
B
Okay,
interesting
yeah,
okay,
well,
I'll
I'll
see
what
I
can
do
I'll
message
Thomas
first
and
see
what
makes
the
most
sense
from
there,
but
I
I
definitely
wanted
to
to
be
able
to
just
yeah
bring
these
ideas.
You
know,
and
maybe,
if,
if
planting
the
seed
there
now
can
help,
you
guys
out
a
lot
in
the
future
to
get
more
companies
to
onboard
captain
and
potentially
us
in
the
future.
Then
this
is
a
huge
win
for
everybody.
Yep.
B
A
E
I'm,
just
curious
and
I
just
want
to
ask
and
I
use
this
community
meeting
regarding
the
peribolo
status.
If,
if
we
have
any
news
from
the
governance
board
regarding
continuous
or
changing
our
community
management
or
things
like
that
or.
A
I
didn't
check
yet
I
didn't
heard
anything
from
the.
E
Community,
there
are
two
pull
requests
open,
currently
the
one
where
we
apply
paribolos
and
then
there
is
the
other
order,
and
then
there
is
the
issue
where
we
are
talking
about
it's
linked
in
the
adding
cerebral
as
functionality.
You
will
find
the
other
ticket
you
further
down.
I
think
is
the
link
where
Thomas
said
he
linked
it.
I
had
an
rfe
established,
maintainer
approval
person,
project
and
now
those
seem
to
be
a
little
bit
stale
and
I'm
curious
because
the
longer
we
wait,
the.
A
E
A
No,
but
I
will
bring
this
up
with
a
governance
board.
I
will
try
to
Ping
them
directly
and
see
the
status
and
try
to
push
it
a
bit
further.