►
From YouTube: Keptn Community & Developer Meeting - Mar 22, 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
A
First
things:
first
roadmap,
so
we
close
quite
severing
items
recently,
in
particular
the
installation
of
of
the
KLTV
Helm,
the
generic
metric
server,
and
also
the
consolidation
of
examples
in
a
separate
Repository,
plus
some
documentation
and
some
refactoring
our
code
base.
And
so
the
next
items
for
oops
for
us
to
work
on
are
the
following
work
on
the
automatic
application.
Discovery
then
start
to
work
towards
enhancement,
proposal
for
the
context,
propagation
across
different
stages
and
some
documentation
about
how
to
install
the
klt
and
a
bit
of
architecture,
foreign.
A
So
for
Captain
D1,
we
are
about
to
cook
the
release,
1.3
everybody.
So
it's
in
preparation
here
we
have
several
but
fixes
or
small
features,
also
added,
but
most
importantly,
we
bumped
a
lot
of
dependencies
because
recently
there
were
quite
some
vulnerability
in
the
go
standard.
Library
we
make
sure
to
use
the
latest
version
of
all
the
go
libraries
to
avoid
these
problems
so
expect
to
release
soonish
or
klt.
We
just
released
on
Friday
last
week
version
0.7.0.
We
didn't
make
an
announcement
yet
because
only
the
Manifest
is
available.
B
A
Yes,
nice,
then,
just
to
refresh
the
memory
about
this
Docker
Hub
problem.
Recently
Docker
announced
that
they're
changing
their
subscription
model.
We
currently
have
a
free
tier
and
we
applied
for
the
open
source
theorem,
but
based
on,
we
never
know
when
that
will
say
yes
or
no
to
us.
We
decided
to
move
towards
thanks
Morris
for
linking
the
proper
issue.
A
So
in
the
short
time
we
agreed
to
move
everything
to
the
Google,
the
GitHub
registry,
and
use
that
as
the
default
one
anyway.
I
want
to
remind
everyone
that
we
are
publishing
everything
in
three
different
registries:
the
GitHub
container
registry
Kai
and
Docker
Hub.
So
images
are
available
on
all
of
these
freeze,
we're
just
changing
the
default
registry
and,
if
you're,
a
maintainer
or
contributor
of
one
of
our
repositories,
in
particular,
for
the
contribute
and
the
sandbox
org.
A
A
A
D
A
So
I
don't
see
any
new
face
here
compared
to
yesterday
meeting
on
the
maintainer
track
for
discussing
the
refinement
of
the
tickets
for
discussing
the
documentation.
All
it
should
look
like
so
yesterday
we
discussed
this
PR
from
Mac
and
we
agree
that
we
should
move
things
instead
of
having
this
more
user-facing.
A
D
A
D
A
C
And
but
but
in
in
I'm,
not
sure
if
we
are
maintaining
information
twice
here
and
the
the
biggest
problem
I
see
there
is
who
maintains
it.
D
Need
yeah
the
duplication
right
now
it
is
I'm
using
as
my
source
a
lot
of
the
stuff.
That's
in
Concepts,
yes,
but
when
we,
if
we
get
this
done,
we
will
get
rid
of
it'll
replace
what's
in
the
concepts.
So
no,
we
will
not
be
double
dishing
on
the
information
and
yeah.
It's
primarily
I
mean
I.
I,
always
appreciate
somebody
telling
me
hey,
I
broke
this
or
whatever,
oh
you
know,
I
will
need
some
of
that
help,
but
overwhelming
for
the
writing
effort.
D
That
will
be
a
writing
effort
and
you
can
you're
free
to
throw
an
issue
at
me
that
such
and
such
needs
replacing
or
this
needs
to
be
added-
or
this
is
just
wrong
whatever,
but
you
don't
have
to
do
with
the
work.
Thank.
C
You
Mike
yes,
but
what
I
just
wanted
to
say,
the
reference
is
all
is,
is
already
old
now,
because
I
saw
now
that
we
are
heading
towards
an
old
API
version
as.
C
Yes,
but
what
this
is
also
not
not
against
something
or
whatever,
but
we
should
be
very,
very
careful
with
this
with
this
kind
of
documentation,
because
this,
these
things
might
change
very
often
in
the
next
few
months,
I
think
I
think
we
we
switched
API
versions,
at
least
in
the
in
the
last
three
versions
two
times,
and
someone
has
to
take
care
of
it,
even
if
we
don't,
even
if,
if
the
development
team
doesn't
take
care
of
it,.
C
Yeah,
so
we
are
thinking
normally
of
the
examples
and
we
are.
We
are
automatically
generating
the
CD
reference,
but
I'm
not
sure
if
we
want
to
to
have
to
if
we
want
to
have
another
point
where
we
have
to
take
such
things
up,
but
that's
only
my
my
take
on
it
so
I'm
not
totally
against
it.
But
we
have
to
be
very
cautious
with
the
information
we
maintained
by
by
hand,
especially
in
such
things,.
D
Right
and
one
thing
I
need
to
document
is
now:
what
I
don't
know
is
if
I'm
upgrading
from
an
old
klt
release
to
the
new
one-
and
there
is
a
new
library,
do
I
have
to
update
my
yaml
files
for
the
new
library
or
can
I
use
the
old
one
or
you
know?
Is
it
a
mix?
Is
it
if
there's
new
features
in
the
new
one
I
have
to
upgrade,
but
otherwise
the
old
one
might
work.
D
F
F
D
To
it-
and
there
is
a
place
for
examples,
I
started
looking
at
the
examples
trying
to
find
this
and
I
don't
know
what
I'm
looking
at
so
I
will
get
better
reading
the
guide,
the
source
code
too,
because
you
guys,
if.
D
F
D
Yes,
exactly
under
yeah
the
examples
we
can
put
in
an
annotated,
I,
don't
know,
Andy
keeps
saying
Andy
one
and
and
examples
in
there
and
I'm
not
sure
that
I'm
completely
I
think
I'm
more
comfortable
linking
to
the
others,
although
I'm
realizing
just
you
know,
a
link
to
the
top
of
examples
isn't
very
useful.
We've
got
to
link
them
into
where
they'll
see
what
they're
looking
for
so
I.
D
A
With
the
commenting
things
and
see
where
we
are
going
towards,
because
we
are
talking
always
about
problems
of
the
documentation,
we
always
postponing
writing
documentation
and
I.
Think,
let's
start
with
the
script
description
of
the
fields
and
slowly
improving,
based
also
on
the
feedback
of
the
readers.
A
D
B
D
D
And
it
may
just
be
I
mean
I
need
to
make
a
checklist
of
what
I
need
to
do
at
new
releases,
and
it
may
be
that
I
need
to
if
I
know
the
right
place
to
look
in
the
code.
I'm
just
going
to
have
to
run
through
all
of
these
and
compare
what
they've
got
with
what's
in
the
current
examples
and
make
sure
there's
the
match
and
if
not
update
and
record
that
there's
a
difference
between
versions
there.
So.
D
D
I
think
what
we
should
do
is
move
like.
Some
of
these
I've
got
a
lot
of
questions
embedded
here,
but
if
we're
going
to
go
forward
with
this,
I
will
take
out
a
lot
of
those
and
we'll
have
them
in
the
background,
but
to
get
something
up
there
fairly
quickly
and
then
it
can
grow.
A
Okay,
so
bottom
line
documentation
looks
very
good.
Let's
start
to
add
more
CRTs
there,
okay
and
I
see
that
I
don't
know
joined
the
meeting.
So
we
can
go
back
to
the
previous
point
and
maybe
let's
start
the
discussion
today.
There
is
also
Thomas,
so
maybe
Thomas
do
you
want
to
start
or
around?
Do
you
want.
E
E
Should
we
go
into
detail
about
like
kind
of
what
at
Brand
watch,
we
were
looking
for
more
or
more
like
I,
don't
know
like
I'm
more
like
I
just
have
the
ideas.
I
I
think
it'd
be
good
to
have
your
insights
on
this
weekend.
C
I
took
a
look
on
on
the
recording
yesterday
to
be
honest,
not
in
not
too
too
deep
into
deep,
but
I
have
some
takes
on
it.
So
there's
also
some
questions.
So
for
me
it
was
not
clear
if
you
want
to
take
actions
based
on
the
on
the
previous
versions
and
so
on,
and
if
it's
only
about
storing
the
version
somewhere
on
the
on
the
Caster,
where
the
life
cycle
toolkit
runs
or
if
you
want
to
have
a
Central
Central
place
where
they
should
be
stored.
C
Because
at
the
moment
and
I
think
this,
this
was
a
design
decision
we
take,
we
took
for
the
for
the
livestock
toolkit,
we
don't
have
any
kind
of
database
behind
it.
We,
the
only
thing
we
have
are
kubernetes
objects
and,
at
some
point,
I'm
totally
fine
with
storing
storing
information.
If
you
need
this
objects
as
long
as
long
as
we
can
manage
it,
and
as
long
as
it
makes
sense.
E
No
I
definitely
think
this
should
stay
stateless,
just
as
how
we
did
it
with
our
in-house
Tool.
For
example,
a
lot
of
image
repositories
have
search.
Functionality
like
artifactory
has
their.
What
is
it
called?
Jfrog
query
language
or
something
like
that,
and
so,
if,
if
you're
talking
about
Docker
images
right.
C
D
C
This
this
is
a
thing
which
is
pretty
hard
to
accomplish,
in
my
opinion,
without
the
control
blame
and
at
the
moment
we
also
have
no
information
about
the
images
themselves.
So
we
know
about
about
Vehicles.
We
know
about
applications
and
about
the
things
which
are
happening
with
them,
but
we
in
fact
we
are
not
really
aware
of
of
images,
which
does
not
mean
that
we
could
not
get
aware
of
this.
But
this
is
a
decision
we
have
to
take
carefully
I.
E
Yeah
some
providers
will
have
the
information
like
you,
can
query
Argo
CD
in
our
case
right
and
and
they
will
tell
you
what's
deployed
so
you
don't
need
to
store
it.
But
then
that's
yeah
you'll
need
a
more
generic
solution
which
I
can
see
like
you.
Don't
want
to
be
querying
the
cluster
directly
to
fill
in
this
information.
I.
Don't
think
that
would
really
make
sense.
C
But
even
if
ago
has
the
information
we
cannot
assume
that
everyone
uses
ago.
C
So-
and
this
was
also
the
the
reason
why
I
wrote
yesterday-
that
it
would
be
a
good
idea
to
oops,
to
put
your
thing,
your
idea
into
words,
because
I
I
also
think
I
don't
even
understand,
I
understood
everything
you
mentioned
there.
E
Sure
well,
I
can
give
like
a
quick,
high
level
right,
like
essentially
I,
think,
there's
a
bit
of
a
philosophy,
difference
between
like
iterating
over
artifacts
and
using
rollouts
and
rollbacks
and
Cannery
deployments,
and
really
focusing
on
an
existing
flow
right
where
you
either
go
back
or
you
continue
moving
it
forward
with
more
percentage
or
promotion
into
your
live
environment
right.
This
is
one
way
of
doing
things.
A
lot
of
organizations
do
this
pretty
much
a
state
of
the
art
and
I
totally
get
it.
E
However,
at
Brand
watch,
one
of
the
use
cases
that
we'll
run
into
fairly
often
is
the
need
to
roll
back
several
artifacts
previously
and
be
able
to
see
what's
there
right,
and
so
in
that
way,
we
just
have
kind
of
what
you
were
saying
earlier
right
like
just
a
table
which
shows
what
can
be
deployed
and
what
is
deployed
in
one
environments,
and
so
this.
This
really
is
kind
of
that
glue
right
between
these
things,
which
is
where
we
originally
were
thinking.
E
Captain
would
be
perfect
for
that,
but
also
yeah
I
I
can
see
now
what
you
mean
like
with
storing
the
deployment
information.
It's
tedious
to
keep
it
stateless.
C
So,
at
least
in
a
way
we
are
doing
it
in
in
it
at
the
moment.
So
at
the
moment
in
the
livestock
toolkit,
we
don't
know,
we
know
about
One
environment,
so
we
have
the
information
about
what's
what's
deployed
or
which
web
contents
and
applications
are
deployed
on
this
one
custom,
if
you
want
to
go
a
step
forward,
going,
have
stage,
promotion
and
so
on.
C
This
information
will
be,
will
not
be
there,
at
least
at
the
current
state
of
the
lifecycle
toolkit
and
therefore
it's
a
bit
hard
to
find
such
things
out.
The
second
thing
we
have
is
we
we
decided
not
to
take
actions
based
on
something.
So
if
you
want,
we
can
store
some
such
information
as
what
was
the
last
working
information,
last
working
version
and
so
on,
but
the
life
cycle
took
it
by
itself,
will
never
do
a
rollback
principles.
C
So
we
could.
The
only
thing
we
could
do
is
we
could
change
git
repositories,
but
in
fact,
on
the
Caster
we
won't
do
a
rollout.
We
won't
do
a
rollback,
because
then
then
we
would.
C
We
would
make
some
changes
on
the
Caster.
The
eclipse
controller
on
the
other
side
would
find
out
something
changed
and
I
have
to
I
have
to
get
my.
My
I
have
to
get
back
to
the
to
my
desired
State
yeah.
A
C
This
will
get.
This
will
get
the
never-ending
story,
and
this
was
also
the
reason
why
we
decided
not
to
not
to
do
some
some
Roblox
or
whatever,
but
we
won't
hold
someone
back
from
writing
actions
where
to
earn
to
do
this.
So.
E
Yeah
I
think
that
was
what
we
were
aiming
to
do
right
was
a
rollback
would
consist
of
triggering
a
GitHub
action
which
would
then
make
the
changes
in
the
git
Ops
repository
and
effectively.
That
would
be
the
the
rollback
action,
but
again
I
can
see
where
you're
coming
from
right,
with
assuming
now
that
you're
using
this
type
of
automation
right
but
but
the
functionality
is
there
right.
Captain
can
definitely
do
this.
It's
it's
not.
You
just
have
to
call
the
trigger
the
whatever
mechanism,
the
thing
doer
that
you're
doing
to
modify
the
repo.
C
We
in
the
meanwhile
also
wrote
the
wrote
a
task
to
a
key
Club
Action,
also
kitlab
a
pipeline,
so
this
would
definitely
work
yes,
the
only
thing
which
is
important,
which
would
be
important
for
us,
would
be
what
use
case
you
exactly
have
and
how
how
you
you
propose
this
to
be
to
be
solved
in
the
end.
Given
this
information
you
have
now
so
I'm,
not
I'm,
not
sure
and
I
think
they
are
Florian
or
the
other.
The
other
developers
might
have
more
insights,
I.
C
Think
at
some
point
in
time
we
start
the
last
the
last
inform
version
of
a
of
an
application
or
a
vehicle.
I'm.
Not
really
sure
is
this.
Is
this
still
true.
C
Yes,
but
I
think
and
I
think
I
was
the
I
was
the
I
was
the
one
who
did
this,
but
I
think
we
in
in
the
airport
in
the
workload
definition,
we
also
store
the
last
version.
C
So
the
previous
version,
because
from
that
because
from
that
we
calculate
the
metric
between
the
two
and
then
we
would
have
the
previous,
then
we
would
have
some
kind
of
a
previous
version
for
tool
to
do
some
rollbacks,
but
to
at
least
to
give
you
the
possibility
to
parameterize
the
tasks.
C
And
it
would
be,
it
would
also
be
interesting
if
we
have
this
in
the
workload.
Another
thing
this
is
the
same,
so
we
should
have
the
same
in
the
workload
instances
as
far
as
I
know,.
C
It
yes,
there
is
also
a
previous
okay
there.
It
is
so
given
this
information,
at
least
you
could
you
could
trigger
a
rollback
if
something?
Well,
you
have
the
information
about
the
previous
version.
The
only
thing
we
we
would
need
to
do
in
this
case
is
to
store
somewhere.
If
this
was
successful
or
not,.
E
E
C
E
E
E
I
think
that's
a
good
idea,
because
there's
a
need
for
this
kind
of
tooling
in
the
CI
cdcncf
space
right
and
that's
kind
of
where
we
ended
up
here
in
the
first
places.
You
know
we're
not
the
only
company
building
these
in-house
tools
right
and
I
I
think.
Maybe
if
you
wanted
to
appeal
to
a
broader
audience
with
Captain,
then
these
kind
of
features
would
go
there.
I'm,
not
sure
you
know
what
what's
the
driving
all
of
this.
It
seems
a
lot
more
about
validation
and
testing
right.
E
A
Okay,
then
just
friendly
reminder
Google
summer
of
no.
This
is
called
Google
season
of
talks
not
summer
of
code.
Do
we
have
any
project
that
we
like
to
work
on?
This
question
is
so
what
you
Meg.
D
D
One
that
I
do
have
in
mind
is
to
take
on
the
Aesthetics
of
the
captain.shell
landing
page,
the
Aesthetics
and
the
content.
There's
old
content
Simon's
doing
a
bunch
moving
stuff
over
to
the
communities,
but
you
know:
there's
all
I,
don't
like
the
looks
of
the
place,
but
that
may
just
be
me-
and
it's
not
my
call
I'm,
not
a
designer
but
the
stuff
in
the
resources
and
the
fact
that
anytime
I'm
in
the
captain,
V1
docs
I've
got
a
link
to
get
to
klt.
D
But
as
soon
as
I
go
into
klt
I
lose
most
of
the
links
that
are
in
the
top.
You
know,
there's
just
a
whole
bunch
of
crappy
stuff
there
and
it
might
be.
It
would
be
something
that
it
would
be
a
good
project
for
somebody
who's
interested
in
sort
of
web
documentation,
and
it
would
not
involve
getting
you
know,
extremely
deep
understanding
of
the
product.
D
Call
it
the
and
I
need
to
go
back
there.
We
had
back
way
a
long
time
ago.
There
were
some
notes.
I
had
there
were
a
couple
of
other
things
that
we
had
in
mind
to
so
yeah.
If,
if
we're
interested
in
doing
this
and
I've
been
trying
to
figure
out,
if
we
were
actually
going
to
go
through
with
it,
and
we
should
Define
a
couple
of
real
projects,
I
I'm
actually
happy
to
Mentor
any
of
these,
but
I
need
somebody
else.
D
D
A
G
Well,
the
code
is
more
likely
ready,
there's
still
the
discussion
regarding
membership
and
how
membership
is
planned
and
how
this
will
change
or
affect
those
kind
of
things
there
isn't.
There's
a
issue
open
in
the
community
repository
wait
a
second
where
it's
over,
it's
opened
by
Thomas,
it's
it's
number
240
in
the
community
Repository
and
the
question
is
how
we
so
what?
What
is
the
outcome?
Because
this
is
actually
somewhat
blocking
the
way?
G
There
are
two
different
approaches
we
can,
we
can
merge,
we
can
merge
paribos
and
then
act
actually
on
that
or
what
the
how
we
proceed.
It's
more
like
the
question
is
how
we
want
to
proceed
here.
It's
not
like
I'm,
not
sure
what
is
blocking,
what
or
how
we
want
to
proceed,
because
this
is
somehow
we
have
a
change
in
structure
of
permissions
more
likely,
because
it's
not
just
one
project.
G
A
G
Yeah
we
can,
we
can.
The
pull
request
is
ready,
I.
Think
there's
just
one
repository
in
the
meantime
for
the
captain
organization,
which
was
created
in
the
meantime,
which
is
not
part
of
that.
The
talks,
tooling
repository
and
the
rest
of
it
is
actually
I
think
virtually
so.
Somebody
need
to
put
need
to
have
another
check,
but
I
also
think
we
I'm
not
sure
who
has
the
permission
to
merge
anything
in
the
community
report
right
now,
because
the
owner
file
also
changed
recently.
A
B
A
G
I
will
just
share
my
screen
shortly.
If
that's
okay,
I
will
just
share
where
I'm
currently
at
and
the
question
is
more
likely
than
other
this
will.
This
will
be
now
a
little
bit
slower
because
the
a
lot
of
things
are
happening
in
the
community
repository
and
we
need
to
wait
for
the
governance
board
to
approve
those
things
because
of
the
code
owner
file.
This
is
maybe
also
something
where
we
might
want
to
think
about.
G
Maybe
if
we
want
to
change
a
little
bit
about
the
code
on
a
file
and
Define
clear
section
which
are
a
where
we
do
need
the
approval
of
the
governance
board,
but
other
section,
for
example
like
the
gsoc
part,
for
example,
the
mentorship
guidance.
We
do
not
need
governance,
approval
for
changing
Pro,
adding
project
ideas,
or
things
like
that.
So
currently,
I
have
this
draft
pull
request
open,
which
already
generates
this
nice
preview
or
I,
took
the
oh,
no
I
took
the.
G
G
But
there
are
a
lot
of
changes
coming
up
actually
because,
as
I
took
a
look
at
it,
I
refactored
that
we
can
even
display
all
our
who
will
sum
of
code
information
and
things
like
that-
and
this
is
just
a
little
bit
of
changes
which
need
to
be
which
I
need
to
add
and
to
fix-
and
this
is
just
something
which
will
be
done
in
the
near
future.
And
with
that
the
question
is:
do
we
also
want
to
have
a
netly
five
build
actually
on
the
community
repository?
G
Because
this
is
now
just
my
community
repository
with
the
Community
page,
so
people
can
actually
validate
during
pull
requests
if
their
changes
are
correct
or
not
correct
and
second
question
to
that.
If
we
want
to
have,
for
example,
a
domain
like
community.cap.sh,
where
we
are
representing
those
changes.
A
No
we're
still
thinking
domain
I.
Don't
think
we
need
that.
B
I
agree:
that's
maybe
a
little
lower.
The
top
I.
Don't
think.
We
need
to
kind
of
over
engineer
that
yeah.
G
G
Mission
yeah,
but
then,
and
then
in
the
end
we
can
people
can
take
a
look
at
the
documentation,
as
I
can
take
a
look
at
the
files
more
easily
and
we
I
think
it's
also
easier
to
maintain
them.
If
you
have
a
web
page
rather
than
just
markdown
files,
so
also
to
have
a
web
page
to
look
at
it
rather
than
make
it
tab
overview.
B
G
At
a
later
stage
that
would
be
is
a
I
would
I
would.
My
suggestion
is:
take
an
atli-fi
built
as
there
anyways,
because
we
have
then
this
overview.
We
have
this.
We
can.
We
can
take
a
look
at
pull
request,
changes
immediately
in
a
nice
rendered
way,
and
also
we
can
take
a
look
as
an
order
we
will
see
later
on
when
we
we
can
start
version
the
community
repository
with
our
changes.
So
it's
something
new
finished
up
and
the
include
will
then
be
better
within
our
Pages.
Where
is
this
page
now
up
here?
G
G
Okay,
now
that's
it
for
myself.
Actually,
yes,
just
my
two
questions:
I
will
create
tickets
for
that,
and
we
can
just
talk
about
this,
and
the
second
question
is
I
shortly
talked
with
Giovanni
about
setting
up
a
one
password
account
for
Captain
an
open
source
account.
The
question
is:
where
should
I
create
this
issue?
Should
it
should
I
create
it
in
the
and
Community
repository
and
the
proposals
and.
G
G
Yeah,
the
one
free
one
for
one
password
you
I'll
normally
only
need
to
open
a
pull
request
and
they
provide
you.
Then
you
say
I'm
this
GitHub
I'm,
this
open
source
organization
and
then
you
can
write
who
should
have
access
to
that
I.
Think
it's
handled
with
a
pool
request.
Last
time.
I
did
this
is
two
years
ago.
I
need
to
check,
but
it's
not
the
that
hard.
G
If
there
is
something
from
the
cncf,
maybe
we
use
the
cntf,
because
it's
also
getting
I'm
not
sure
if
you
real,
as
a
lot
of
you
most
likely,
have
received
the
email
from
GitHub
regarding
two-factor
Authentication.
G
B
All
on
a
2fa
I
have
access
and
I
think
triovanni
has
access
yeah.
A
B
A
Right,
if
you
see
here
in
order
to
get
into
the
organization
you
need
to
have
two-factor
authentication,
otherwise
I
cannot
send
you
an
invite
all
right
and
I.
Think
since
is
the
cncf
that
is
owning
our
organization,
they
set
that
up
most
likely,
but
someone
I
was
looking
here
in
the
CCF
tooling
that
they
provide
us
is
a
zoom
account,
which
is
the
one
that
you're
currently
using,
but
also
LastPass
one
password
and
key
base
to.
D
A
B
B
G
D
D
And
does
it
need
to
be
I'm
having
without
doing
details
and
I
could
do
details
I'm?
Finding
that
I'm,
not
sure
I
agree
with
all
of
the
mark,
all
that
the
markdown
linger
is
doing
and
I
know
we
can
control
some
of
that,
but
I
think
the
the
one
that
just
got
me
is
I.
Think
I
got
busted
on
having
a
space
between
list
items
in
the
bullet
list
and
I've.
D
You
know
most
of
the
markdown
documentation
sort
of
recommends
that
you
do
that
when
you've
got
long
items
to
make
the
source
file
more
readable,
I'm,
not
100
sure
that
that
was
what
the
problem
was,
but
I,
but
but
I
just
I
and
some
of
these
things
I
don't
know
you
know.
Obviously,
if
you
have
a
past
TCO
you're
not
going
to
merge
but
Simon,
which
of
these
are
are
any
of
the
doc
tests
just
advisory
at
this
point,
are
they
all
blockers.
G
G
For
example,
and
the
the
the
the
extra
checks
are
blocking,
because
we
do
not
want
to
have
malicious
right
what's
in
there
and
the
other
part
is
for
example,
here
we
have
markdown
rules,
no
multiple
planks,
no
inline
HTML.
G
Those
are,
for
example,
we
we
do
not
want
to
have
people
writing
custom
HTML,
because
they
with
that
they
could
style
whatever
they
want.
If
there
is
a
reason
we
need
to
take
a
proper
look
at
it
right
and
the
other
part
is,
for
example,
multiple
consecutive
blank
lines.
One
blank
line
is
fine,
but
multiple
ones
are
it's
the
question,
so
it
does
not
complain
about
one.
It
complains
about
multiple
ones.
G
You
can
always
run
make
markdown
link
minus
fix
locally,
and
this
will
take
care
of
all
those
errors.
There's
no
of
most
of
those
errors.
There
are
some
errors
which
are
fixable
so
blank
lines.
Markdown
link
can
fix
HTML
elements
inline,
it
cannot
fix
because
it
does
not
know
what
to
do
or
how
to
replace
them
or
what
else
to
put
in
there.
Okay,
but
make
markdown
lint
Dash
fix,
will
fix
all
kind
of
markdown
link
issues
actually
and
with.
D
G
This
markdown
link,
minus
fix,
is
the
that's
the
command,
and
when,
when
this
is
run,
it
will
try
to
fix
as
many
problems
as
possible
what
it
can
fix,
and
we
also
have
a
second
pull
request
open,
which
also
adds
additional
rules
like
we
had
this
issue
with
the
git
GitHub
admonitions
doing
that.
So
we
have
a
custom
rule,
then
fixing
those
and
putting
them
into
proper
format
that
we
have
the
that
are
running
possibilities
and
also
other
the
maximum
of
one
sentence
per
line.
D
D
Cool
and
those
are
the
two
big
ones
that
we've
got
right
and
for
the
record
shall
we
add
that
there
is
work
ahead
so
that
the
doc
checks
won't
run
on
software
PRS
and
the
pr
checks
or
the
software
P
checks
won't.
F
D
G
Is
done
already,
this
is
in
place.
This
is
done,
but
there
is
there
are.
There
are
some
some
some
some
some
some
yeah?
How
to
say
is
git
specifics,
which
you
need
to
be
aware
of,
because
as
soon
as
you
do
a
merge,
for
example,
at
Magic
Main
in
there,
somebody
has
changes
in
the
software
repository
done,
and
this
is
part
of
your
merch.
G
Your
change
set
of
the
pull
request,
although
it's
not
visible
within
the
podcast,
your
change
that
contains
changes
in
the
in
the
in
the
software
files
and
that's
why
it's
running
so
there
are
the
instead
of
doing
a
merge.
A
rebase
is
a
little
bit
better
because
then
you
do
not
have
those
changes
in
there
and
see
I
will
not
be
triggered
for
those
because
of.
D
D
G
Checked
there
was
no
I
I
searched.
There
was
no
nothing
from
mentioned
from,
as
I
did
as
I
searched
for
the
git
command.
Git,
merge
and
I
did
not
find
it
in
our
contribution
the
last
time
so,
but
this
is
something
very
which
so,
instead
of
keep
merge,
it
replays
is
most
likely
better
because
you
do
not
get
this
change
set.
G
D
G
G
Based
on
the
documentation
from
Giovanni
or
what
you
wrote
in
here,
we
should
maybe
add
in
in
as
a
side
note
that
people
should
be
aware
of
doing
git
push
force
with
leads
if
they
do
a
rebase
most
likely,
so
that
they're,
not
as
if
there
are
changes
that
they
do
not
do
get
push
for
us.
They
should
do
a
git
push
force
with
lease
so
that
they
do
not
change
or
have
any
kind
of
side
effects.
They
are
not
aware
of.
A
G
G
On
a
on
a
side
note,
which
would
be
actually
maybe
quite
helpful,
I
think
I'm
not
sure
if
it's
up
set
up
for
the
kill.
T
repository
that
you
have
your
update
branch
configuration
that
you
have
on
a
pull
request
when
you're
out
of
date,
I'm,
not
sure
Kenny
I
did
not
check.
Maybe
you
can
change
it
that
this
is
not
a
merged
but
a
squash
pull
request,
but
the
squad
as
a
rebase
in
the
settings.
G
G
G
Allow
suggest
updating,
pull
requests
actually
there.
It
is,
and
with
that
people
at
least
do
not
have
to
do
it
manually
on
their
own.
But
when
they're
in
the
pull
request,
they
we
just
need
to
make
the
people
aware
that
they
click
replays
and
not
merge.
So
when
you
go
down,
you
know
this
is
not
the
is
this
outdated.
A
A
But
I
I
saw
it
already
there.
It.