►
From YouTube: Cloud Foundry Community Advisory Call [November 2020]
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).
B
Good
to
go:
okay,
welcome
to
cloud
foundry
community
advisory
board
meeting
a
short
one
today,
I
think.
For
november
the
18th,
a
small
group
of
people
together
welcome
everyone.
C
No,
nothing
in
particular
chip
is
here.
I
don't
know
if
you
have
anything.
D
Yeah
yeah,
so
we
just
shared
earlier
in
the
week
that
there's
a
kind
of
a
redesign
of
the
technical
governance
for
the
the
purpose
of
like
how
decisions
are
made
in
the
projects.
What
projects
are
formed.
D
All
that
good
stuff
happening,
there's
been
a
small
working
group
from
the
four
the
four
member
companies
who
represent
about
95
percent
of
the
the
contributions
to
the
project.
That's
been
kind
of
working
over
the
last
couple
of
months.
D
That
said,
it's
a
bit
of
a
straw
man
and
starting
in
two
weeks,
because
here
in
the
u.s
there's
you
know
we're
going
to
try
to
thread
the
needle
a
little
bit
and
have
the
first
version
of
this
meeting
that
is
open
to
anyone
with
any
interest
whatsoever
to
come
and
join.
So
if
you
look
at
the
community
calendar,
let's
see
two
fridays
from
now.
D
D
B
Thank
you.
I'm
going
to
drop
the
link
in
the
agenda
doc
so
that
people
can
find
it
and
yeah
just
connect
that
to
what
I
was
saying
earlier.
We
can,
after
the
updates.
I
want
to
talk
a
little
bit
about
conformance-based,
certification
and
tie
that
into
I
want
to
get
some
more
people
some
more
eyes
on
it
from
a
smaller
group
before
we
send
it
out
to
the
whole
organization.
D
A
Look
chris
we're
doing
the
hands-on,
hands-on
lab
in
in
mark
in
december.
C
Yeah
for
cf
protect
yeah,
I
believe
james
is
on
vacation
at
the
moment.
He
was
going
to
get
us
some
copy
for
that,
but
we
have
a
tentative
date
which
is
not
fresh
in
my
mind,
but,
yes,
we
will
be
running
the
cf
protect
hands-on
lab
in
december.
At
some
point.
B
E
Yeah,
I'm
happy
to
do
that,
troy.
So,
on
the
reliant
side,
they've
had
a
couple
of
quick
releases
of
cf
deployment
to
deal
with
some
certificate
regeneration
issues
that
I
believe
the
latest
version
of
golang
is
requiring
in
terms
of
how
it
verifies
certificates
so
watch
out
for
those
I
don't.
I
don't
think
that
they
should
introduce
any
major
complications,
we're
just
making
sure
that
those
certificates
are
appropriately
generated
for
going
to
interact
with
them
and
then
on
the
cf4k
side.
E
They've
been
continuing
work
on
both
track
for
some
more
standardized
image
management
and
more
sophisticated
management
of
secrets
in
the
configuration
there
and
then
qcf
has
had
another
minor
version
release
of
2.6
and
I
believe,
they're
investigating
what
would
be
like
to
integrate
some
of
the
kpac
based
staging
mechanisms
that
we've
had
in
cfr
gates
into
qcf
as
well,
which
sounds
great
on
some
of
the
component
team
fronts.
E
I
know
the
cappy
team
and
the
services
api
team
have
requested
comment
on
what
is
the
final
stretch
of
the
v3
api
endpoints
and
corresponding
v7
cfcli
commands.
E
These
are
all
for
various
service
management
and
service,
binding
interactions,
arena's
also
continued
working
on
support
for
sidecar
processes
and
they're
continuing
work
on
their
lrp,
crd
and
they're
also
streamlining
and
simplifying
some
of
their
templating,
which
I
think
they're
planning
in
preparation
for
a
2.0
version,
because
there
might
be
some
breaking
changes
to
the
configuration
there
and
then
networking
is
continuing
on
track
of
work
around
more
flexible
plugability
of
other
ingress
routing
and
networking
systems
focusing
on
contour
as
a
lighter
weight
option
alternative
to
issue,
and
then
logging
in
metrics
has
been
continuing
work
on
improving
log
cache
performance
and
scalability.
E
B
E
B
Okay,
no,
I
thought
was
it
was
it
this
span
is,
is
jules
is
still
leading
it.
E
B
E
Jules
yeah,
no
actually
here
jules,
is
now
the
project
lead.
That's
right
for
the
team,
yeah.
B
Yeah,
I
I
don't
know
if
everyone
was
aware
of
that,
but
that
that
we've
changed
jules's
for
irene
right,
yeah
boshbmc
kevin.
Do
you
have
any
updates?
B
Sorry,
I
didn't
mean
to
cut
off
questions
by
the
way
anything
else
for
eric.
B
That's
all
right
I'll
chase
up
some
posh
updates
on
slack.
If
we
don't
have
anyone
here
and
for
extensions,
the
only
update
I
have
I
will
have
more
after
the
next
extensions
pmc
meeting,
but
yeah
stratos
4.3
release,
mostly
bug
fixes
and
some
back
end
work.
Let
me
quickly
call
up
the
release,
notes.
B
To
see
what
that
was,
I
should
know
it
off
the
top
of
my
head.
I
don't
think
it
was
a
lot
of
huge
stuff.
B
List
filters
some
work
on
the
helm,
chart
list
some
styling
and
coloring
thing
for
the
the
dark
theme.
I
think,
and
a
bunch
of
bug
fixes
so
congrats
to
that
team.
B
But
that's
the
only
thing
I
have
for
open
discussion.
I
just
want
to
introduce
this
document
I
started
at.
You
is
well
not
at
use
request.
It's
something!
I've
been
talking
with
yui
about
for
a
while,
and
she
said
you
know
you
really
got
to
get
this
started
and
because
no
one
else
is
going
to
float
this,
and
so
I'm
going
to
drop
a
link
here
in
the
google
doc
link
for
something
I
will
share
once
I've
had
some
more
eyes
on
it
with
cf
dev
conformance-based
cf
certification.
B
So
I
talked
a
little
bit
about
this
in
philadelphia
in
the
the
cf
kubernetes
panel
that
we,
as
we
move
away
from
the
traditional
vm
based
cf
deployment
release,
and
we
have
now
two
different
kubernetes
implementations:
the
ship,
this
code
model
of
the
certification
process
for
for
cloud,
foundry
certified
release
distributions
is
going
to
be
more
difficult
because
we
would
just
have
to
keep
adding
different
optional
pieces
of
code.
B
In
order
to
to
to
sort
of
guarantee
this
roughly
the
same
user
experience,
it
would
be
much
better
to
define
that
user
experience
by
way
of
something
that
tests
the
api.
Like
cloud
foundry
acceptance
tests,
so
what
I'm
proposing
here
is
that
we
start
with
cats.
B
The
current
cloud
foundry
acceptance
test
suite
as
a
point
for
if,
if
a
distribution
of
cloud
foundry
passes
either
all
of
the
cloud
foundry
acceptance
tests
or
a
designated
subset
or
a
designated
subset,
plus
some
additional
tests
that
it,
it
should
be
considered
certifiable,
so
that
those
are
the
that
it
would
pass
the
technical
requirements.
B
Now.
This
has
some
interesting
side
effects
and
it
it
it
maybe,
is
different
from
the
original
goal
of
cloud
foundry
certification
and
feel
free.
Anybody
to
jump
in
and
correct
me
or
offer
a
different
point
of
view,
but
I
believe
the
something
I
can
I
refer
to
as
the
exact
software
requirement,
which
is
just
borrowing
some
of
the
wording
from
the
certification
requirements.
B
I
believe
the
exact
software
requirements
were
designed
to
encourage
people
to
all
contribute
to
the
same
code
base
so
that
maintenance
load
is
was
shared
rather
than
having
each
company
working
on
forks
of
of
cloud
foundry.
So
it
was
designed
to
bring
bring
the
companies
together
to
actually
like,
initially
or
certainly
historically,
to
pair
on
on
the
software,
but
also
to
make
sure
that
the
code
base
was
was
homogeneous.
B
I
think
now
that
we
already
have
two
converging
diverging
parallel.
I
will
say
projects
for
cf
on
kubernetes,
it's
time
to
really
look
at
opening
things
much
wider.
B
That
is,
if
we
were
to
to
go
towards
an
api
driven
conformance
test
like
kubernetes
has
we
could
have
certain
components
of
cf,
be
radically
implemented
and
still
provide
the
same
interface
to
end
users
and
potentially,
if
we
want
to
define
this
provide
the
same
interfaces
internally
within
cloud
foundry
on
a
component
by
component
basis.
B
Now
that
part
of
this
there's
a
lot
of
to
do's
in
my
in
my
document
here,
so
I'm
please
understand
I'm
sharing
a
very
early
version
of
this
in
hopes
that
other
people
might
contribute.
But
this
would
be
a
different.
The
end
result
might
be
a
bit
different.
We
might
have
many
implementations
growing
for
either
components
or
entire
distributions
of
of
cloud
foundry,
but
I
I
thought
I'd
put
this
out
there
from
susa's
perspective.
We
would.
B
We
would
really
like
to
have
the
freedom
to
to
innovate
and
and
still
have
a
distribution
of
cloud
foundry,
which
is
which
is
certifiable.
So
that's
that's
the
motivation
from
me
and
from
some
of
my
colleagues
open
it
up
for
thoughts
generally
on
a
conformance
based
more
similar
to
the
kubernetes
model
test
framework
for
for.
B
Certification
chip-
I
saw
you
mentioning
something
in
the
the
the
github
issue
where
I
raised
this.
D
Yeah
that
was
well.
I
actually
brought
you
in
to
raise
it.
The
the
first
thing
I
mentioned
was
that
you
know
we
want
to
rework
the
the
way
technical
governance
operates
right
and
so
the
the
straw
man
proposal
is
that
you
sort
of
reconstitute
instead
of
having
the
pmc
council
with
pmcs
with
projects
it
kind
of
gets
reshuffled
around
and
it
becomes
a
toc
with
a
bunch
of
working
groups.
D
So
my
my
suggestion
in
order
of
operations
here
is
we
don't
need
to
rush
to
do
an
update
for
2021.,
let's
sort
out
who's,
how
we're
going
to
make
decisions
as
a
community,
but
and
then,
if
that
straw,
man
is
roughly
what's
going
to
happen.
D
Let's
stand
that
thing
up
and
then,
let's
make
deciding
about
exactly
you
know,
maybe
maybe
approving
the
final
approach
to
to
the
way
cert
should
work
going
forward
in
in
that
yes
eoc
body
right.
So
that
was
the
right
repo
to
toss
that
that
idea
into
so
guillaume
from
orange
added
it
there,
which
is
awesome,
and
then
I
think
that
issue
basically
ends
up
being
one
of
the
early
things
that
that
atsc
would
take
on.
D
In
in
early
2021,
so
at
the
same
time
the
fact
that
you're
sharing
this
now
is
awesome,
troy
because
I
do
think
people
should
start
thinking
about
it.
If,
if
we
wait
to
start
to
formulate
plans
or
approach,
then
it
means
that
won't
be.
That
won't
actually
be
the
thing
that
changes
until
you
know
the
summer
or
later.
So
it's
good
to
do
some
things
in
parallel.
B
Yeah
and
and
to
be
to
be
clear,
this
is
very
much
a
work
in
progress
and
I'm
looking
to
get
eyes
outside
of
souza
on
this
as
well,
so
that
we
have
something
fairly,
full
fully
fleshed
out
to
present
to
the
new
governance
body
to
to
make
a
decision
on.
So
we've
got
some
consensus
in
the
community
already
about
what
what
what
it
means
to
be
a
cloud
foundry
distribution,
what
it!
What
cloud
foundry
is
a
lot
credit
to,
and
I'm
just
blanking
on
his
name.
B
The
stark
and
wayne
unconference
presentation
cloud
foundry
is
dead.
Long
live
cloud
foundry
james
james
yeah,
who
was
on
the
last
cab,
call
as
well.
His
presentation
sort
of
kicked
off
like
okay.
Other
people
feel
this
way
that
that
what
cloud
foundry
really
is
is
the
experience,
and
especially
because
it's
changing
so
much
right
now
and
it's
it's
changed
over
time.
B
The
thing
that
is
most
important
is
is
preserving
the
amazing
feeling
you
get
when
you
use
cloud
foundry
and
and
somewhat
of
the
the
workflow
that
has
been
established
at
so
many
companies
of
people
who
who
use
this
day-to-day
and
then
allow
for
more
freedom
on
the
implementation
side,
because
there's
been
a
lot
of
inven
innovation.
B
I
know
from
other
companies
that
have
re-implemented
parts
of
cloud
foundry
and
they
just
sit
in
a
private
fork
somewhere,
because
it
doesn't
really
make
sense
for
them
to
to
bring
it
into
the
cloud
foundry
community
because
it
would
never
fit
in
a
certified
distribution.
So
part
of
this
is,
is
me
thinking
about
revitalizing,
also
the
extensions
pmc
and
if
there's
some
overlap
there,
but
but
I
I
really
think
this
is
the
way
to
go.
I
know
a
lot
of
my
sousa
colleagues
feel
the
same
way.
B
So
any
of
you
on
this
call,
please
ping
me
on
slack,
and
I
can
add
you
as
an
editor
to
this
document.
Anyone
can
comment
on
it
so
feel
free
to
do
that.
B
But
if
you
want
to
help
me
work
through
these
ideas,
just
just
hit
me
up
on
slack
and
we'll
we'll
open
it
up
any
any
questions
about
it.
A
I
I
think
our
experience
has
been
using.
We,
we
use
the
cat
stuff
to
help
validate
some
of
our
deployments
of
code
as
well
for
clients,
and
but
I
don't
think
it's
at
the
level
that
that
I
would
call
a
a
certification
level
at
all.
Yeah.
B
This
is
feedback,
I've,
I've
gotten
from
scissor
developers
as
well.
So
we
would
need
to
replace
some
of
cats
or
use
that
as
a
starting
point,
but
it
would
require
in
some
areas
more
rigor
and
and
in
other
areas
we
would
just
leave
out
parts
that
are
very
specific
to
certain
internal
implementations.
I
know
there's
certainly
diego
specific
cats,
which
wouldn't
really
apply
to
irani.
B
I
think
this
was
some
of
the
reason
they
turned
off
some
of
those
tests
so
yeah,
and
that's
where
I
really
need
help,
because
I
I'm
not
familiar
enough
with
the
cloud
foundry
acceptance
tests
running
them
myself
to
know
what
they're
exercising
so
I've
got
some
help
on
the
suicide
to
to
go
through
these,
but
anyone
else
who's
interested
to
identify,
because
that's
where
the
hard
work
is
going
to
be
is
actually
making
the
test
suite
and
creating
something.
That's
a
viable
conformance
suite
of
tests.
So.
A
I
think
the
major
lacking
is.
I
look
at
the
cats
as
being
more
of
a
unit
testing
more
than
anything
else.
Yep
from
that
standpoint
there
are
some
integration
tests,
but
it's
not.
It
doesn't
take
it
from
the
angle
of
the
end
user
or
the
user.
If
you
want
to
do
certification,
you
want
to
make
sure
and
you're
trying
to
validate
the
user
experience.
A
Yeah,
it
exercises
the
commands,
but
it
doesn't
exercise
the
commands
in
the
order
of
things
how
how
things
get
accomplished.
It's
kind
of
like
a
single
stack.
He
said.
Okay,
this
worked,
but
you
don't
have
anything
that
does
you
got
stuff
that
goes
depth,
but
nothing
in
the
breath
angles,
and
so
there's
just
not
enough
testing
in
there.
A
In
my
opinion
that
you
know,
does
it
doesn't
combine
together
segmentation
groups
along
with
trying
to
make
sure
routing
works
and
all
this
other
stuff
everything
seems
to
be
isolated
and
you
can
run
stuff,
and
you
really
don't
know
exactly
how
all
the
pieces
fit,
and
so
I
think,
there's
a
different
way
of
looking
at
how
the
tests
should
run
and
then
I
think
that
will
allow
you
to
change
the
tests
to
to
be
more
certain
certified.
Do
you
know
that
you
can
run
a
thousand
users
or
stuff
like
that?
A
E
E
E
Rights,
I
I
think
maybe
it
is
useful
to
look
at
the
scope
of
conformance
testing
that
that
kubernetes
has
and
what
they
found
useful
in
their
community.
You
know
is
that
primarily
functional
testing?
Is
there
scale
or
performance
testing
there
as
well?
And
you
know
these
could
could
all
be
different
areas.
You
know
we
don't
have
to
do
it
all
at
once
right,
but
from
my
perspective,
starting
from
cats,
as
still
representing
that
end-to-end
flow
for
a
particular
user.
B
Something
else
that
just
occurred
to
me
and
I
don't
know
how
much
work
it
would
well
probably
a
lot
of
work
to
backing
up
the
cloud
foundry
certified
developer
exam
is
based
on
a
bunch
of
code
that
defines
how
a
cloud
foundry
developer
should
use,
or
you
know
what
a
cloud
foundry
developer
should
know
to
use
the
platform.
D
They
are
less
specific
than
I
even
think
that
the
the
cats
suite
is
right.
So
remember
the
approach
for
how
do
you
sort
of
how
do
you
verify
the
skills
of
an
individual?
Is
that
you
think
about
what
are
the?
What
are
the
points
of
knowledge
that
someone
has
to
have
in
their
head
right,
easily
accessible
in
order
to
define
them
as
being
an
expert
in
using
something?
I
think
that's
a
very
different
intention
from
how
does
a
you
know?
D
How
do
you
look
at
conformance
of
api
implementation
relative
to
two
downstream
distributions
right,
because
we
might,
there
are
a
ton
of
features
that
are
not
tested
for
in
the
developer
certification
and
they
shouldn't
be
because
they're
less
commonly
used,
but
they
might
very
easily
be
worthwhile
to
check
for
conformance
between
distros.
D
B
Oh,
my
goodness,
I
totally
didn't
think
of
that.
This
is
why
I'm
not
a
teacher
okay.
What
about
the
idea
of
different
categories
or
tiers
of
of
conformance
that
is,
you
could
get
a
conformance,
so
it's
not
a
pass
or
fail.
It's
a
percentage
conformance
or
it's
a
or-
and
this
is
what
I'm
thinking
is.
We
have
a
core
set
which
is
pass
or
fail,
and
then
other
areas,
for
example,
routing.
B
If,
if
a
distribution
didn't
want
to
implement
tcp
routing
or
wanted
to
implement
it
differently,
could
that
be
outside
of
the
core?
What
what's
called
the
core
functionality.
C
So
I
I'm
a
hundred
percent
behind
a
core
which
should
be
centered
around
the
ex
developer
experience
and
then
a
you
know,
sort
of
like
a
a
feature
group
or
ver
like
vertical
of
features.
Like
you
know,
if
you
care
about
this
stuff,
then
here's
the
the
scores
in
the
respective
departments,
but
they're
not
necessary
to
satisfy
what
is
the
core
experience
of
cloud
foundry.
So
you
know
the
core
experience.
The
cloud
foundry
should
obviously
be
the
cfcli
based
stuff
plus
build
packs.
C
Okay,
you
know
score
right,
which,
unfortunately,
right
now
cf
for
kate's
would
probably
not
score
too
well
in
that
area,
but
they're
coming
along
nicely,
for
example
right
like
so
that
I
would
say
I
am
100
behind
that
approach
of
the
the
core
they
nailed.
The
experience
plus,
if
you
care
about
these
verticals
or
whatever
you're
going
to
use
feature
set
groups
like.
B
Yeah-
and
there
are
some
fairly
well
understood
categories
of
that-
I
think
there's
there's
a
container
to
container
networking,
not
everyone
does
or
not
everyone
uses.
I
should
say.
B
What
else
the
build
taxes
there's
a
lot
of
divergence
in
build
packs
between
distributions.
C
B
B
And
I
do
think
it
was
smart
too,
to
leave
the
build
packs
out
of
of
the
current
certification
for
that.
For
that
reason,
because
people
do
have
very
specific
requirements
and
often
fork
build
packs
to
to
do
something.
C
B
Yeah
and
now
now
do
you
think
that's
gonna
carry
on
or
even
there'll
be
more
proliferation.
C
Honestly,
yes,
I
feel
it's
going
to
not
only
carry
on
for
for
needing
that
feature,
but
I
think
the
like
today,
we
can
have
turn
on
for
cloud
foundry,
the
you
know
either
use
build
packs
approach
or
run
docker
container
images
approach,
and
I
think
that,
but
but
still
can
maintain
the
cf
experience
developer
experience
in
the
future.
I
foresee
you
know,
especially
on
top
of
kubernetes
like
right.
Like
being
you
know,
supporting
both
directions
is
going
to
be
very
important,
but
the
bring
your
own
is
the
most
critical
thing.
G
B
No,
that's
excellent
excellent
feedback.
D
Yeah
and
the
score
actually
had
a
nice
side
effect.
If
you
can
sort
of
say
here
is
the
the
core
of
of
cloud
foundry.
You
know
these
five
items
out
of
ten
here.
Is
you
know,
team-based
cloud
foundry,
six
items
other
ten
here
is,
you
know
for
lack
of
a
better
word
enterprise
cloud,
foundry,
nine
out
of
ten,
because
I've
always
been
advocating.
D
You
know
smaller
footprint,
so
being
able
to
say
you
can
still
be
conformed
without
feature
a
b
and
c.
You
know
maybe
a
a
lightweight
component
swapped
in
this.
B
I
I
it's
something
we're
looking
at
as
well,
and
something
that's
been
on
my
mind,
something
the
reason
I
really
like
cf
local
is
because
there
is
a
great
use
case
for
developers,
so
just
using
pac
or
or
on
its
own
or
or
cf
local
allows
you
to
check
that
your
staging
is
going
to
work
locally.
So
the
the
tightest
loop
of
the
the
iterative
process
with
your
app
can
happen
on
your
own
machine
and
once
that's
nailed
down.
B
Then
you
move
to
pushing
it
to
to
a
real
pass
wherever
that
might
be
on
your
local
server
or
you're,
even
in
a
small
footprint
cf
on
your
local
machine,
but
but
that
that
ability
to
scale
down
the
essence
of
cf
for
a
developer,
to
work
with
locally
would
be
great
and
that
what
that
would
mean
is
that
would
be
maybe
an
oslis
or
a
an
orgless
cf,
or
something
with
just
a
single
user
cf.
B
And
that
would
there
be
a
whole
chunk
of
the
cats
that
would
not
pass
for
that,
and
that
would
have
to
be
a
a
special
case
and
maybe
that
kind
of
thing
doesn't
have
to
be
certified.
That's
maybe
we
could
leave
that
out
of
the
certification
process
and
that's
just
you
know
a
nice
tool
that
you
use
to
to
check
that
your
stuff
is
going
to
deploy
okay
to
cf.
C
C
These
are
the
things
I
care
about
rate
and
give
me
a
score
right
or
what's
the
score
for
these
things,
and
then
they
can
go
based
on
their
business
needs
versus
just
in
general,
because
if
we
have
a
one
system,
that's
not
going
to
work
so
well,
because,
oh
it
scored
60,
except
that
you
don't
know
that
yeah
percentage
is
all
you
care
about.
So
you're
good.
B
Yeah
we
so
the
is
something
that
was
floated
around
us
is
the
idea
of
a
matrix
where
you
can
actually
just
so.
This
is
this.
Is
the
sweets
these
and
we
wouldn't
do
it
test.
We
could
you
know
you
could
always
expose
test
by
test
what
passes
and
fails.
You
could
have
like
a
score
sheet
that
people
could
real,
but
I
think
people
need
groups
like
some
logical
groupings
of
those
tests,
so.
B
B
No,
you
caught
it
right,
you
know
as
we're
getting
into
it,
so
it's
kind
of
where
we're
going
with
this.
A
Troy
it's
almost
like
a
behavioral
test
suite
in
some
ways
what
you're
looking
for
you're
looking
for
this
behavior.
B
Yeah
I
mean
we
do
want.
We
do
want
the
correct
response
to
an
api
call
so
that
we
know
that
if
you
use
the
client
it's
going
to
behave
the
right
way
or,
if
you're
implementing
a
new
client,
it's
going
to
behave
the
right
way
do
we
include
both
v2
and
v3
is
another
thing
that
I'm
curious
about.
C
So,
due
to
where
we
are
at
in
that's
a
loaded
question-
and
I
I
normally
I
would
say
no
just-
do
b3
focus
on
the
future
moving
forward.
This
is
a
new
effort.
Just
sort
of
a
new
change
just
go
with
that.
The
problem
is,
I
don't
think.
I
think
that
we're
in
a
state,
whereas
v2
and
v3
is
like
incomplete.
Is
it
not
like?
We've
got
some
some
drift
between
us.
You
know
what
works
on
kevin.
Isn't
it
isn't
it
something
you
guys
discovered
like
we've
got
some
drift
on.
C
F
A
B
Better
with
the
latest
releases,
so
you
can
go
back
and
forth
more
easily.
Now,
okay,
that's
that's
we
because
we
found
that
as
well,
but
the
the
very
latest
releases
of
both
v6
and
v7
seem
to
not
not
be
so
bad.
I
haven't
given
it
done
at
all.
I
just
know
that
there's
a
lot
of
people
in
the
world
that
are
still
on
the
v2
api
using
the
the
v6
client.
They
have
a
lot
of
tooling
in
place
that
they
don't
want
to
necessarily
move
right
away.
C
B
C
A
D
Yeah,
if
I
remember
correctly,
there's
a
few
things
done
in
v3
a
little
bit
differently
than
in
v2.
So
it's
not
it's
sort
of
a
change
in
the
workflow
under
under,
but
maybe
it
would
be
just
simple
to
have
in
the
matrix.
You
know
v3
compatibility
and
then
another
one
for
v2,
optional,
v2
compatibility.
F
I
can
give
some
feedback
from
from
sap
side.
We
have
actually
gone
through
this
exercise
to
switch
from
cfcli
version,
six
to
version
seven
for
all
our
internal
installation
or,
let's
say
for
a
big,
the
bigger
part
of
it,
and
that
was
actually
not
a
big
deal.
I
mean,
if
somebody,
let's
say
programs
against
the
cli.
It's
anyway
broken,
but
those
scripted
stuff
was
was
was
not
so
much
of
a
problem.
F
Those
colleagues
who
were
really
implement
against
the
v2
api
and
now
have
to
migrate
to
v3.
They
have
a
some
some
more
work
to
do,
especially
where
synchronous
calls
switch
to
other
chronos
calls
and
the
biggest
problems
that
we
have
is
that,
let's
say
the
v3
api
still
has
here
and
there's
some
bugs,
and
there
are
some
really
nasty
bugs
inside
so
race
conditions,
and
I
guess
it
will
take
another
half
a
year
or
so
until
all
these
things
are
gone
actually
and
you
get
the
stability
back.
F
F
Of
the
open
and
copy
team
is
actually
looking
into
it,
but
these
are
tricky
things:
it's
not
the
easy
bugs
they
are
all
found
and
and
fixed,
probably
by
cats
and
and
other
things.
But
let's
see
the
delicate
things
so
yeah.
B
B
Also,
I
think
I
think
we
should
push
to
to
get
these
bugs
fixed
faster
than
six
months
right,
especially
if
they're
things
people
are
going
to
hit
in
the
wild
yeah.
You've
probably
done
a
very
exhaustive
testing
to
find
these
things.
Hopefully,
they're
they're
more
edge
cases,
even
if
they're
nasty
but.
F
C
Troy,
I
was
just
reminded
in
the
background
that
effectively
the
the
thing
that
I
was
coming
to
mind
for
me
is
that
we,
our
last
round
of
evaluating
cf
for
cates.
It
really
only
was
supporting
the
v7
cli
in
api,
basically,
and
no
support
any
longer
for
cappy
v2
calls
and
and
what
the
struggle
there
for
us
was
that
that
the
current
implementation
of
stratos
would
not
work.
B
C
F
Yeah,
okay,
definitely,
I
think
it's
a
way
forward.
Whatever
can
you
do?
What
would
help,
for
instance,
is
also
when
you
bring
out
a
real
deprecation
plan
for
v2?
That's
one
of
the
most
asked
questions
that
I
get
is
v2
still
available
or
not,
and
then
you
say
well
yeah,
it's
not
yet
officially
deprecated,
but
the
resources
have
already
in
the
documentation
and
application
mark.
You
should
use
v3.
We
can
recommend
v3
cfcli
version.
F
6
is
deprecated
at
least
on
our
landscapes,
because
we
have
recency
of
deployment
versions
inside
where
it's
not
supported
officially
anymore,
it's
still
working
and
sometimes
even
better
than
the
cf7.
So
that's
too
much
talking
I
mean
if
you
have
here
a
clear
statement
mid
of
next
year.
V2
will
be
gone.
Time
is
ticking.
Please
migrate,
that's
something
that
everybody
understands
and
where
then
also
people
will
start
to
work.
If
you
explain
that
much
of
stories-
and
everybody
will
say,
okay
got
the
message-
I
will
wait.
A
A
B
Okay,
has
anyone
else
got
any
topics
for
discussion
today
that
otherwise
we
can,
we
can
call
it
early
and
get
back
20
minutes
or
so
just
wait
is
it?
Does
anyone
want
to
help
with
this
document?
Does
anyone
want
edit
rights?
B
Okay,
thanks
norm,
I'll
ping,
you
sorry
anyone
else.
G
This
is,
I
just
wanted
to.
Let
folks
know
that
there's
a
couple
of
hands-on
labs,
the
priciest
one,
which
is
very
introductory
scheduled
for
early
december,
one
in
the
first
week
and
one
in
the
second
week.
So
one
of
the
things
that
we
are
trying
to
do
is
schedule
this
in
time
slots
that
are
more
friendly
and
in
the
eu.
Slash
africa,
slash
middle
east
regions.
So,
okay,
if.
G
That
you
know
who
are
in
these
regions,
who
might
be
interested
in
an
introductory,
see
a
friend
something
thing.
D
Europe,
asia
friendly
times
and
then
we've
also
got
a.
B
So
that's
tri-cf
and
have
you
ram,
have
you
put
something
in
cf,
dev
about
that
or
cf
the
cf
users
mailing
list
or
something
where
can
people
find
more
information.
G
B
Okay,
I'll
add
it
to
the
the
notes
here.
So
everyone,
if
you
know
someone
in
those
time
zones
those
regions,
please
evangelize
this
and
make
it
known
that
there's
some
tri-cf
courses
in
first
and
second
week
of
december
thanks.
B
Thanks
very
much
for
joining
this
episode.
Hopefully
we'll
have
some
I'll
round
up
some
presentations,
but
I
I
thought
it
was
really
good
to
have
this
discussion
really
really
interesting.
Some
interesting
information
on
v2
versus
v3.
Thank
you.
Thank
you.
Stefan
I
may
reach
out
to
you
with
some
questions.
Later.