►
From YouTube: Retrospective: Regressions in Jenkins 2.204.3...2.204.5
Description
We had serious regressions in the 2.204.x LTS baseline starting from 2.204.3 which was released on Feb 28 (changelog). We held a retrospective to to discuss what we could do better in the next releases to prevent such issues from happening again.
Discussion: https://groups.google.com/forum/#!topic/jenkinsci-dev/d9lgjbruRno . Retrospective document: https://docs.google.com/document/d/1NzR1XtkCfk6MDSD1jRq5H-iPSjqKvP-pfUO0b9lVZ9k/edit?usp=sharing
A
Okay,
really
just
in
case
you
watch
this
recording
this
prospective
for
recently
Jenkins
LTS
releases.
You
might
have
seen
that
between
merging
is
to
go
to
told
to
go
to
three
and
2.5.
We
had
a
number
of
progressions
and
that's
about
to
be
a
discussion
in
the
developer.
Man
increased
and
agreed
what
we
want
to
review
our
processes
and
to
discuss
what
winter
long
until
how
we
could
improve
and
how
we
open
for
quality
of
the
next
releases.
The
interested
you
can
not
find
with
conversations
worries.
C
A
A
Okay,
so
the
first
item
might
be
good
and
improve
quickly
raises
qualities
or
the
tribulations
don't
get
hits
the
analogous
and
the
last
minute.
So
it
comes
from
the
history
because
the
many
regressions
which
led
to
the
instability
we
actually
integrated
to
only
in
the
weekly
cycle,
but
we
didn't
notice
them
until
very
late
faces
and
ALCS
based
on,
for
example,.
A
Yeah
fair
point,
so
you
know
what
let
me
rephrase
I
mean
that
we
did
basically
what
you
said.
We
didn't
act
on
one,
so
two
regressions
were
basically
introduced
in
version
2.2
or
five,
and
whatever
happened
is
the
back
porting
process.
We
didn't
really
work
on
them
until
February,
so
it
was
quite
late
and
it
we
fixed
them
earlier.
We
wouldn't
have
had
this
problem
and
for
that
yeah
I
think
that
we
should
just
double
down
on
weekly
release
quality.
A
A
B
Three,
it
feels
challenging
to
actually
motivate
these
people
to
be
our
characters
right
because
everybody,
let's
face
everybody's,
got
an
interesting,
more
interesting
project
to
work
on
and
we're
all
busy.
So
it's
very
hard
to
get
somebody's
share
of
time
on
this,
and
getting
people
allocated
solely
on
this
I
mean
he's
very
challenging
to
keep
the
team
motivated
doing
their
job
in
a
long
term
as
I
see
it
I.
A
Totally
agree
with
you,
so
what
I
was
thinking
about
that
that
we
could
have
just
one
contributor
but
on
annotation
basis,
so
directly?
One
year
ago
we
didn't
have
so
many
people
contributing
to
Jenkins
code,
but
they
changed
so
explain
the
Jenkins
coach
team
and
now
we
could
probably
agree
with
his
leaving
this
team.
But
somebody
just
monitors
and
comment
wishes
for
one
week
than
somebody
else,
and
this
is
how
we
could
balance
the
load.
E
A
So
I
was
monitoring
all
incoming
wishes
during
the
timeframe,
object,
200
and
not
a
big
changes.
We
had
a
couple
of
years
ago.
I
really
do
think
it
takes
too
much
time
to
monitor
them.
But
having
one
person
takes,
look,
let's
say
every
day
would
be
really
helpful
and
the
next
step
would
be
to
actually
follow
up
and
to
define
an
action
plan,
because
initial
response
is
one
thing,
but
we
also
need
to
plan
a
fix,
but
even
initially
response,
and
maybe
some
highlighting
for
LTS
would
be
nice.
A
D
Well,
that
would
be
especially
in
this
case.
The
problem
is
LTS.
Candidate
is
for
things
that
are
expected
to
be
back
ported
and
the
problems
we
encountered
were
reported
in
the
weekly
release,
just
after
the
last
LTS
baseline.
So
for
its
first
three
months.
This
would
not
be
a
backporting
consideration
at
all,
because
there
were
like
18
weekly
releases
during
which
a
fix
would
be
integrated
into
the
weekly
and
would
go
into
the
next
LTS
baseline,
like
that.
So
that
is
not
really
helpful.
D
A
D
A
Obviously
miss
who
would
fix
the
defect-
maybe
this
next
one,
the
Jenkins
called
him-
have
a
better
situation
now
on
this
day.
In
the
case
of
these
issues,
when
I
highlighted
them,
JC
fixed
this
one,
almost
immediately
this
one,
this
was
basically
fixed
by
dependable,
but
we
also
did
some
analysis
and
testing
quickly
and
it
happened
only
one.
These
issues
were
highlighted
in
the
discussions.
A
A
A
Okay,
so
the
next
item
is
about
discovering
quiches.
So
one
of
the
reasons
why
we
factor
these
issues
is
what's
important.
It
was
related
to
the
portion
of
57
a
3/8,
so
it
was
a
relatively
minor
progression.
Well,
I,
don't
know,
can
see
B
minor,
but
it
was
reported
long
to
go.
I
mean
we
didn't
have
so
many
words,
but
to
be
back
ported
to
eat
and
imperil
the
bed.
We
also
posted
the
entire
instant
weight.
So
that
was
a
comment
that
will
market
sots
can
be
big,
but
it
rarely
Winterson
but
14.
A
Where
did
whatever
digression
and
5.6
so
basically,
instead
of
custom
installation,
the
entire
Winston
was
but
posted
and
again
there
was
no
blame
at
all
because
they
know
multiple
relievers
who
are
supposed
to
take
a
look,
and
they
everyone
I
missed
that.
So
it's
something
we
just
need
to
renew
and
here
so
what
would
help
to
discover
such
commands.
B
Yeah,
it
was
absolutely
misunderstanding
on
my
side
when
I
reread
it
after
he
pointed
me
to
daddy,
was
absolutely
clear.
What
you
find
out,
I
misunderstood,
that
at
5.6
is
sufficient
and
I
guess
was
the
poor
didn't
be
haven't
bumped
as
far
as
we
as
we
need
it
to
so
yeah.
There's
a
misunderstanding
on
my
side:
yeah!
B
D
A
A
B
B
So
how
about
getting
the
the
owner
or
the
assignee
of
the
issue
that
is
being
reported
to
actually
have
it
reviewed
I
mean
I
understand
is
you
know,
puts
the
burden
on
the
other
people
that
they're
supposed
to
review
the
back
board,
but
since
they
understand
it
the
best
and
they
are
understandable-
the
implications
and
things
that
might
be
very
hard
to
share.
In
plain
English.
A
It
would
be
definitely
important
so
and
in
in
this
case,
is
definitely
mistake
on
my
side,
because
I
was
the
one
who
submitted
a
pull
request
to
pump
out
the
dependency
I
was
the
one
Celsius
candidate
and
I
received
notification
when
it
was
a
bit
ported
and
basically
I
was
busy
with
other
stuff.
So
I
didn't
really
review
that.
So
if
we
could
ensure
that
there
is
additional
review
notice
and
that
we
ensure
that
I.
A
Was
thinking
that
the
best
way
for
that
is
just
do
it
through
pull
request
and
this
your
comments
here
are
basically
the
same,
so
have
working
through
public
ways.
The
question
is
how
to
do
that.
Yes,
your
preference
is
to
have
one
bulk,
who
requests
which
does
initial
reporting
and
then
a
number
of
minus
ones.
If
we
need
to
pick
something.
B
Yeah,
my
proposal
really
was
that
we
would,
as
a
community,
to
cultivate
the
content
that
was
soon
before
class
and
then,
when
we
are
all
okay
with
what's
in
there,
who
just
merge
it
as
it
is
so,
basically,
the
work
in
progress
would
be
eliminating
isolated
in
the
single
or
request.
So
that's
what
we're
working
on
right
now
and
at
the
time
you
merge
it,
which
actually
represents
the
agreement
that
this
is
what
goes
into
the
next
release.
Obviously,
if
we
forget
something
or
discover
something
later,
it
would
just
create
a
follow-up.
D
So
well,
the
current
state
is
directly
committing
to
the
branch
after
locally
applied
backwards,
but
I
think
this
is
a
false
comparison,
because
another
option
would
be
to
actually
have
per
issue
per
feature
pr's
into
LTS
next
to
the
other
ones.
Since
we
introduced
the
label
into
all
TS,
which
we
can
always
rename,
it
would
be
easy
enough
to
distinguish.
Lt
is
related
back
port
pull
requests
from
the
others.
D
I,
don't
really
see
the
benefit
of
having
one
pull
request
for
benefit,
although
Oliver
makes
a
good
point
that
it
would
allow
us
to
identically
already
have
the
LTS
in
a
separate
branch,
because
this
would
probably
be
an
original
PR
anyway,
based
on
an
origin
branch,
and
so
we
have
the
actual
final
LT
s
branch
and
we
deliver
from
one
other
branch.
So
I
can
see
it
going
both
ways.
A
Another
advantage
should
this
approach
a
that.
Basically
it
requires
less
time
because
if
we
start
plating
at
multiple
parallel
pull
more
request,
then
just
Li
you
want
the
hook
to
resolve
potential,
merge
conflicts.
If
they
happen,
then
whomever
does
the
backcourt
anchor
has
to
spend
more
time
to
submit
the
two
requests
separately
so
I.
It
includes
some
overhead
yeah
I.
D
And
I
mean,
depending
on
how
our
liver
has
time
or
whoever
opened
up
or
requests
me
basic,
there's,
there's
a
fair
likelihood
that
we
have
most
of
the
stuff
in
toward
the
beginning
of
the
backcourt.
We
know
right
which
would
allow
someone
interested
in
reviewing
to
submit
a
review
even
after
it
was
merged.
So
we
can
always
amend
branches
needed.
B
Yeah
I
don't
mind
being
the
one
creating
the
creating
the
branch
both
initially
so
we
test.
How
does
this
little
girl,
or
even
they
do
long
term
I
mean
I
presume
this
all
becomes
semi
automated.
So
even
the
poor
request
creation
with
some
template-
and
you
know
mentioning
necessary
people
and
labelling
or
whatnot-
can
be
done
with
the
automation,
so
I
don't
expect.
That
would
be
that
much
of
a
problem
and
I
I
love
not
like
I
agree
to
you
know
start
doing
this.
A
This
current
korppoo
request
three
years
will
define
to
dedicate
some
time
to
review
in
courteous
and
we
can
request
reviews
from
Homer
contributed
fixes,
so
I
believe
that
the
most
of
patches
come
from
people
who're
members
of
the
Harper
organization.
So
everybody
can
just
request
reviews
directly.
B
Yeah,
that's
right.
Perhaps
a
little
bit
of
a
challenge
would
be
that
we
basically
backcourt
based
on
zeros,
but
we
have
the
JIRA
IDs.
We
have
access
to
the
comments
inside,
but
actually
identifying
the
particular
and
github
handles
for
the
users.
There
will
be
some
some
mapping
here,
some
challenge
to
overcome,
but
yeah.
A
So
right
now
what
we
are
doing.
So
if
you
go
to
the
Jenkins
your
repository,
then
you
can
see
that,
for
example,
for
weekly
llamo,
we
updated
how
to
make
to
include
in
the
bottom.
So
ultralink
includes
authors
of
political
eyes
right
now.
You
just
don't
expose
it
there
and
they
change
look,
but
we
can
use
this
metadata,
for
example
in
backporting
or
anywhere
else
to
generate
a
list
of
review
requests.
B
C
A
F
A
A
F
A
F
A
B
A
B
B
Well,
I,
don't
think
so,
I
guess
at
the
time
they
merged
the
poor
request.
Somebody
would
have
to
go
and
change
the
labels
from
candidate
to
to
fixed
in
Britain
Club
version,
which
I
guess
would
be
where
I
integrates
that
again.
This
is
under
OTS,
later
I'm,
fine
doing
that
I've
been
doing
this
before
so
it
just
slightly
changes
the
time
when
this
is
being
done,
but
yeah
yeah.
B
Actually
you
got
me
thinking
that
perhaps
it
would
be
good
to
somehow
indicate
energy
has
been
put
into
the
request
right,
so
owners
and
original
contributor
and
I
don't
know
people
are
watching,
for
that
would
be
a
very.
We
are
considering
that
and
can
be
part
of
the
discussion
because
it
would
be.
D
D
D
B
Yeah
I
was
referring
to
that
so
yeah
I
think
my
suggestion
to
this
was
I
mean
probably
not
very
systemic
one,
but
suggesting
that
there
will
be
more
eyes
on
the
process.
So
these
kind
of
issues
is
less
likely
to
slip
and
if
we
I
mean
I,
don't
have
a
strong
opinion
on
this
I
sort
of
suspect
that,
even
if
we
are
this
label,
it
doesn't
necessarily
make
sure
that
whoever
does
the
back
poor
thing
doesn't
miss
an
important
piece
of
information
somewhere
in
there
right
so
I
guess
it.
B
This
will
be
very,
very
structured
like
if
backporting
a/b
needs
to
leave
the
port
is
B.
Perhaps
you
can
somehow
encode
that
right,
I,
don't
know
either
in
labels
or
links
or
whatever,
and
then
we
can
actually
have
an
automation
that
would
keep
us
accountable
right.
I
mean
when
I
send
the
announcement.
There
is
enough
automation
to
actually
verify
that
the
are
Seabees
are
deployed
before
I
send
an
email,
and
there
is
no
way
that
email
would
go
out
if
the
base
are
not
already
there
and
I
guess.
B
We
can
do
something
like
that
to
actually
make
sure
that
if
we
say
well,
okay,
if
a
baby
needs
to
be
a
backported
as
well,
so
we
can
actually,
you
know,
get
some
other
automation
around
that
it
would
actually
verify
that
these
kind
of
preconditions
are
all
Matt.
That's
something
I
can
imagine,
but
if
we
I
say
that
this
is
non-trivial
and
care
needs
to
be
taken,
we
again
are
back
in
relying
that
humans
would
do
the
right
thing.
Yeah.
D
D
A
Imagine
we
didn't
have
all
these
issues
music,
instant,
but
we
just
had
a
decision
which
is
still
there.
A
request
for
version
bumper
from
instant
I
would
say
that,
even
if
there
was
no
report,
officious,
miss
Winston,
if
would
have
been
to
go
back
for
think
because
yeah,
but
of
for
Denise's
jump,
which
includes
something
like
eight
or
nine
objective
releases.
So
even
without
report
Tisha's
it
with
equal
additional
care.
When
you
doing
such
a
back
course.
D
If
you
look
at
the
original
issue,
rather
than
what
we
discovered
late
and
what
went
wrong
into
two
or
four
point
three,
it
introduced
a
regression
in
through
use
of
Winstone
five
point:
eight.
Instead
of
five
point,
nine
or
five
point
six,
instead
of
five
point:
seven,
yes,
this
one
and
that
could
easily
have
been
a
linked
issue
in
JIRA.
That
basically
says
this
issue
cause
the
other
one
and
I
mean
I,
mean
I'm,
not
saying
that
the
label
is
wrong,
but
it
seems
like
we
introduced
a
less
useful.
D
Alternative
to
linking
issues
specifically
for
this
purpose
now,
maybe
this
is
what
we
have
to
make
it
super
visible,
but
Oliver
I,
don't
know
what
the
state
of
your
backporting
automation
is
or
whether
you
do
everything
manually,
but
I
would
say
as
soon
as
there's
a
linked
issue.
If
you
have
a
script
or
something
that
would
be
surfaced
and
would
need
separate
acknowledgement,
props.
B
Right
currently,
when
it
comes
to
links
linked
issues,
we
rely
basically
on
the
fact
that
they
don't
spot
this
partially,
because
not
all
the
causes
or
I
mean
all
the
link
types
seems
to
have
a
different
semantics
to
what
we're
discussing
right.
None
of
them
explicitly
says:
if
a
is
that
port
EP
needs
to
be
reported
as
well,
so
I
like
this
proposal,
but
I
would
probably
vote
for
having
a
different
link
type.
So
it
can
be.
B
B
How
old
is
it
and
stuff
like
that,
and
perhaps
it
definitely
can
be
extended
in
a
way
that
it
says
okay,
but
if
you
for
days
there
are
these
other
issues
it
was
caused
by
that
and
there
needs
to
be
exported
as
well,
but
basically
that
we
require
a
manual
again
manual
review
of
these
issues
to
see
because
they
can
be
in
a
no.
They
not
everything
that
is
linked
in
the
causes.
Link
type
does
necessarily
needs
to
be
reported.
In
my
opinion,
right.
D
But
you
will
need
a
manual
review
anyway,
even
if
you
have
the
non-trivial
backporting
label
right,
the
non-trivial
backporting
label
tells
you
hey
something's
going
on
here.
You
need
to
be
very
careful
when
puck
cording
this.
While
the
causes
relationship
says
hey,
you
need
to
be
careful
when
by
parting.
This
look
at
this
other
issue
to
figure
out
what's
going
on,
so
the
second
has
a
higher
fidelity.
D
C
D
To
clarify
I'm
not
saying
we
shouldn't
do
non-trivial
backporting
ID
as
an
additional
indicator
that
might
might
still
be
useful.
What
I'm
saying
is
that
in
each
link
or
process
we
apply
during
the
back
porting
process
should
look
not
just
for
the
label
but
also
add
linked
issues.
Now,
obviously,
the
semantics,
or
sometimes
not
great,
but
even
if
you
have
a
back
court
candidate,
it
has
a
link.
Issue
probably
means
you
should
take
a
look
at
that
link
relationship
and
the
link
issue.
A
D
Right
there
are
two
parts
to
this.
Ideally,
we
look
at
when
we
investigate
issues
we
create
the
causal
relationships,
that's
the
one
half
of
it.
The
other
half
is
that
Oliver
who
or
whoever
else
is
doing
the
back
porting-
considers
these
linked
relationships
as
important
data,
as
potentially
important
metadata
to
inform
how
how
things
should
be
back
ported
or
whether
things
should
be
back
ported,
because
even
LTS
candidate
is
marked
cause
an
unresolved
issue.
That's
probably
going
to
be
a
rejection
right.
There.
A
A
D
A
D
Mean
it
convert
to
expose
release
candidates
more
publicly
and
not
just
rely
on
the
limited
audience
of
devilÃs
to
get
testing
in
the
problem,
specifically
with
the
the
issues
we
experienced
here
is
they
are
four
very
specific,
setups
and
basically
impossible
to
catch
in
what
I
would
call
test
testing
setups
one
was
extremely
long.
Configuration
forms,
typically
the
cutout
configuration,
even
if
you
have
a
clock
configuration
with
five
different
loud
templates
configured.
D
You
won't
hit
this.
You
need
dawson's,
many
many
dozens.
You
need
to
click
really
many
buttons
to
trigger
this
issue
and
the
other
issues
were
the
wild
card
certificates
also,
probably
not
something
that
your
test
environment
has
and
a
very
narrow
and
not
recommended
reverse
proxy
configuration.
So
maybe
I'm
under
estimating
the
fidelity
of
our
complexity
of
test
environments,
but.
F
D
Specifically,
in
the
reverse
proxy
host
error
case
right,
if
you
configure
the
host
headers
in
one
order,
it's
broken
in
the
other
order.
It's
fine
that
was
was
an
annoying
ones
and
I
mean
in
the
totality
of
all
jetty
users
that
still
occurred
plenty,
but
I'm,
not
sure
how
that
would
help
here.
So
obviously
the
plan,
but
the
idea
is
good
and
recommendable
and
we
should
be
doing
it,
but
we
shouldn't
expect
miracles.
That's
that's
what
I'm
trying
to
say
here,
yeah.
A
A
A
A
Download
page
also
I
was
thinking
just
about
tech
with
github
release,
because
right
now,
if
you
subscribe
to
Jenkins
air
Jenkins
and
in
you
get
identification
and
there
is
the
number
914,
so
this
is
a
number
of
github
users
or
guaranteed
to
receive
notification,
and
actually
this
number
doesn't
include
ones
who
just
watches
releases.
So
I'm
not
sure
what
is
the
exact
number,
but
if
we
just
push
it,
let's
say
it
through:
github
produces
additional
notifications,
so
go
out.
B
F
B
The
Alexis
suggestion
was
that
they
would
publish,
didn't,
have
a
release
in
order
to
get
all
the
people
that
are
subscribed
to
notification,
so
we
informed
them
that
there
is
a
release
candidate
for
them
to
test,
and
my
concern
was
that
we
would
have
to
you
know
basically
just
create
this.
Release
is
for
TRC.
I
was
afraid
that
people
suggesting
using
since,
like
a
draft
releases
visually,
doesn't
send
notifications,
etcetera,
so
yeah,
I,
guess
with
this.
What
is
it
alpha
releases
or
whatever
I
call
it
pre-releases?
This
would
not
be
an
issue.
D
D
A
So
it's
just
a
best
effort
to
get
highlights,
work,
intra,
post,
a
link
from
downloads.
It
makes
sense
because
we
already
have
incremental
releases,
so
we
could
just
add,
align
or
whatever
your
transom
various
candidates
and
other
things,
but
right
now
somebody's
willing
to
do
that.
It'd
be
great
right.
Now
we
don't
so,
for
example,
here
on
past
releases,
so
you
just
release
candidates.
D
A
I'm
sure
that
Jenkins
LTS,
our
testing
timeframes,
don't
have
too
much
overlap,
so
I
think
it
was
one
of
the
root
causes,
for
example,
for
me,
Kosovo,
which
was
definitely
a
root
cause.
We
had
never
to
get
a
lot
of
changes
towards
the
LCS
baseline,
including
system
review
missions,
including
manage
permissions
for
some
web
UI
things
plus,
but
for
some
other
things,
and
although
we
historically
say
that
the
release
is
expected
to
be
several
weeks
old,
so
it
should
be
here.
The
baseline
attorney
steps
technically
between
two
to
five
three
weeks.
A
B
Let
me
see
if
I
understand
and
understand
your
concerns,
so
yeah
I
agree.
Did
we
have
sort
of
drifted
off
from
this
two
to
five
weeks,
because
there
was
a
lot
of
pressure.
You
know
obviously
I
mean
the
developers
would
like
to
have
latest
features
in
in
the
OTS
which
I
understand.
So
even
you
would
benefit
if
we
would
be
more
conservative
in
choosing
the
the
baselines.
D
How
that
would
change
anything?
Because
if
you,
because
the
problem
is
we
have
the
LTS
cycle
so
that
every
four
weeks
we
work
so
that
it
is
on
a
four
week
cycle
and
activities
related
to
LTS,
repeat
every
four
weeks
and
it's
worse
for
the
point
one
because
for
LTS
users,
that's
a
major
version
increase,
but
the
problem
here
was
actually
the
point:
three
pepper
preparation,
rather
the
point
than
the
point,
one
preparation.
D
So
exactly
so,
if
we've
just
said,
for
example-
and
you
said
four
weeks,
if
we
just
moved
the
LTS
a
cutoff
to
four
weeks
earlier,
then
you
would
have
been
busy
with
the
LTS
preparation
or
the
baseline.
Getting
your
fixes
into
the
upcoming
baseline
four
weeks
earlier
and
would
not
have
had
time
to
review
and
test
LTS
point
to
rather
than
LTS
point
three.
So
I
don't
see
how
that
would
help
anything
else
and
not
whatever
pose
so.
A
Right
now,
this
situation
in
a
bit,
so
the
critical
part,
is
he
between
week,
two
and
week,
four,
so
what's
LTC
testing
and
his
release
and
next
Doge's
baseline
selection.
So
if
you
say
that
we
follow
how
the
documentation
on
our
side,
then
the
weekly
emerge
window
would
be
some
way
here,
so
during
back
porch
and
core
even
before,
but
not
during
the
testing
side
right
now.
So
what
we
have?
A
We
have
weekly
inertia
russian
between
two
and
week,
two
and
four
just
getting
the
last
minute
changes
in
place
and
at
the
same
time
you
will
have
LTS
release,
but
basically
people
do
not
spend
too
much
time
on
looking
at
that.
So
for
me,
what
would
what
would
be
the
preference
that
week?
Zero
is
LCS,
but
portal
continues
then
yeah
whatever
weekly,
and
then
we
have
time
for
testing.
F
A
D
D
A
Just
try
it
at
four
so
that
there
is
no
pressure
to
integrate
all
fixes
in
dot
free
and
if
something
goes
wrong,
not
all
lost
Algeria's
the
boost
process,
but
I
think
that
the
first
step
would
be
to
just
follow
these
two
weeks
advance
as
as
a
default.
Obviously,
we
could
reconsider
that
if
there
is
something
critical,
but
if
you
just
say
that
two
weeks
is
our
default,
state
I
was
very
strong
consensus
and
the
communique
the
situation.
A
D
Back
to
the
past
few
years,
old
LTS
baselines
elections
work.
We
basically
looked
at
releases
and
in
many
many
times
we
basically
said
well
I
like
to
release
from
two
weeks
ago.
But
if
you
look
at
the
change
look
for
the
release
one
week
ago,
we're
just
fixing
bugs
that
we
want
to
be
LTS
candidates
anyway,
so
we
can
go
one
week
more
recent
and
at
least
I
haven't
checked.
But
that's
how,
in
my
mind,
we
got
to
the
very
aggressive
releases.
D
D
B
Usually
the
conservative
voice
when
it
comes
to
the
discussion
and
I
sort
of
agree
that
there
would
be
different
reasons
why
to
be
more
conservative
when
doing
that
and
I
like
the
point
that
Daniel
make
that
no
exceptions
because
of
features
but
perhaps
exceptional
because
of
the
bug
fixes.
But
at
the
same
time
this
is
the
most
commonly
heard.
Reason
for
people
nominated
something
newer
than
that
right,
usually
when
people
want
us
to
choose
aggressive
OTS
bassline
is
because
of
some
features
API
or
something.
B
A
There
could
be
potential
alternative
if
you
really
need
to
deliver.
For
example,
kudos
to
the
22
in
this
case,
I
believe
it
was
a
good
justification
because
features
deliver.
They
are
really
important,
but,
for
example,
we
could
have
said
that
I
understand
it's
important.
We
agreed
to
go
with
that,
but
let's
say
we
delay
the
OTS
cycle
by
two
weeks
so
that
we
have
more
time
for
integration,
testing
and
feedback
before
shipping
grade
so
yeah.
We
intentionally
shift
everything
by
two
weeks
so
that
there
is
no
collision
is
those
three?
B
B
Don't
think
I
mean
it's
just
another
irregularity
to
them
around
the
timing
of
LTS.
We've
done
that
once
in
a
while,
usually
for
more
serious
reasons,
mm-hmm
I
didn't
feel
that
strongly
here,
cascais
not
here
to
tell
us
that
is
very
important
to
not
to
disappoint
people
under
shred.
You
I
don't
consider
that
death
to
be
all
that
important
that
we
release
every
every
four
weeks.
I.
D
Mean
ultimately,
there's
a
little
big
difference
between
either
approach.
The
difference
is
just
where
do
we
introduce
the
delay
right,
because
both
approaches
of
having
a
later
point,
one
in
the
schedule
and
choosing
an
older
baseline,
mean
that
any
features
will
be
at
least
six
weeks
old
as
LT
s
point
one
is
released.
B
Another
point
there
is,
let's
presume,
that
it
goes
to
discussion
right
so
which
was
one
sorry
two
point
222
as
an
obvious
baseline,
where
it
was
too
late
his
release
and
they
would
choose
to
actually
postpone
everything
by
two
weeks.
Would
it
not
create
a
push
on
the
core
reviewers
to
actually
be
very
conservative
in
the
next
in
the
next
week?
Please.
A
Miss
might
so
I
think
that
it
will
be
natural
force
to
become
more
conservative
when
it
comes
to
the
end
of
the
notch
window.
Newell,
yes,
but
at
the
same
time,
people
will
try
to
deliver
changes
and
be
interested
to
help
them.
So
for
me,
I
would
rather
accept
being
more
conservative
at
whatever
she
has
cut
off
than
before.
B
A
A
D
Mean
I
see
essentially
two
options
here:
one
is
we
really
try
to
enforce
the
minimum
age
role
again,
that
is
documented
and
that
we've
sort
of
given
up
recently
and
when
we
do,
we
will
have
a
better
time,
also
communicating
that
to
contributors
rather
than
you
know.
Recently,
we
always
went
with
the
newest
one
somehow
gets
transformed
to
alright.
This
is
the
exact
weekly
cutoff
for
your
feature
together.
D
F
D
That
will
also
give
us
more
breathing
room
for
the
new
release
automatically,
because
these
super
late
release
just
will
not
exist
at
this
point
and
we
cannot
choose
later.
Obviously,
if
we
go
this
route
and
look
at
older
LTS
releases
only
with
we
are
back
at
this.
This
problem
we
had
like
five
years
ago
six
years
ago
when
the
LTS
releases
were
really
badly
outdated
and
even
the
current
LTS
line
just
felt
wrong.
If
you
were
working
on
Laster.
B
A
B
D
B
Actually
liked
it
a
bit
more
than
making
it
a
14-week
cycle
because
it
would
feel
irregular
or
making
that
one
of
its
special
I
mean
to
me.
This
feels
a
bit
bit
easier.
There
was
one
point:
I
tried
to
make
I
mean
I.
So,
like
you,
you've
pointed
out
that
a
lot
of
people
was
rushing
to
put
their
new
features
into
a
release
that
they
believed.
B
D
D
B
Yeah
I
mean
your
writers,
probably
for
the
first
time
that
there's
so
much
of
a
demand
on
a
feature
side
for
choosing
an
LPS
beta
baseline
and
the
fact
is,
we
didn't
even
have
the
we
didn't
even
have
the
first
release
out.
So
we
really
don't
have
the
entry
experience
whether
it
was
good
or
not
right
perhaps
going
to
blow
up
in
your
face,
and
it
would
be
a
problem
and
we
will
learn
from
that.
But
for
now
we
don't
know
that.
Well,.
D
A
B
A
B
D
C
A
D
Yeah,
it
doesn't
tell
I
mean
right
now:
it
doesn't
help
that
we're
working
in
two
week,
increments,
I'm,
not
sure
I
want
to
go
more
detail
there
right,
because
because
another
problem
is,
or
rather
the
main
problem,
is
that
the
RC
testing
period
and
the
development
period
overlapped.
And
if
we
can
basically
you
say.
D
Ideally,
features
would
arrive
when
they
arrive
and
get
picked
up
in
LTS
in
in
regular
in
the
walls.
So
I'm
not
sure.
We
should
make
this
much
more
explicit
as
project
documentation
and
I
mean
if
someone
also
say
I
have
this
feature
and
I
want
it
to
be
in
the
next
30
as
baseline.
We
can
explain
what
this
one
sentence
on
this
page
of
documentation
means
for
them.
Not
sure
we
want
to
have
you
know.
Lts
driven
development
needs
basically
be
supported
beyond
that.
B
The
thing
that
makes
most
sense
to
me
is
the
suggestion
to
actually
do
the
decision
for
the
next
baseline
two
weeks
early.
It
would
immediately
get
two
more
weeks
of
soaking
before
the
doctor
that
one
is
out
and
people
can
if
they
want
to
voice
their
was
their
preference
to
actually
choose
the
latest,
and
greatest
I
mean
they
can
continue
in
a
trend
of
making
this
aggressive
baseline
choices,
because
it
would
be
two
weeks
later
when
it
when
it
actually
arrives.
D
I
think
that
makes
sense.
My
proposal
here
in
terms
of
implementing
this
would
be
the
use
of
middle
poll
requests
to
Jenkins
I/o
that
modifies
the
LTS
documentation
accordingly,
basically
write
stuff
into
the
table
and
write
a
sentence
that
says
two
to
five
weeks
and
link
it
in
a
discussion
on
the
dev
list
and
let's,
let's
get
some
more
feedback
there
I
think
that's
the
most
pragmatic
approach
and
when
we
once
we
have
consensus
or
nobody
brings
up
any
major
concerns
that
seem
relevant,
we
can.