►
From YouTube: 2022-07-13 Delivery team weekly EMEA/AMER
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).
A
A
A
Hello,
so
this
is
like
13th
of
July
2022
delivery
group
weekly
for
America
anemia.
This
is
like
what
we
used
to
have
on
Monday,
but
we
move
it
to
to
today.
So
we
have
some
more
people
and
we
took
the
chance
of
moving
the
other
one
since
blame
is
on
holiday,
so
we
have
some
announcement
or
people
being
on
holidays
and
might
have
been
available
today
and
he
already
put
Myra
in
as
a
sink
an
update
for
they
remove
the
Block
in
nature
post,
deploy
migration.
A
So
for
the
removing
the
blocking
nature
of
post-deployed
migration,
epic,
they
started
rollout
of
the
positive
migration
pipeline.
The
announcement
was
already
sent
in
the
in
the
various
channels.
It
was
also
cross
reference
in
the
like
infrastructure,
Lounge
other
channels,
and
there
is
a
video
recorded
to
pick
overview
or
what
what
it's
about
so
I,
don't
know.
I
didn't
watch
it.
Yet
if
you
had
a
chance
to
watch,
it
feel
free
to
go
ahead
later,
I'm
gonna.
A
Do
it
probably
tonight,
as
the
next
steps
we
had
phase
three
so
probably
completing
the
rollout.
We
are
going
to
observe
how
it
behaves
for
this.
We
can
fix
any
problem
that
may
search.
There
are
also
some
pending
issues
that
we
need
to
to
finish
in
order
to
complete
the
O'hare.
A
A
A
And
I'll
take
it
as
a
no
point.
B
is
about
the
reducing
release
management,
oil,
epic
and
we
start
to
to
start
to
measure
something
that
is
happening
in
our
release.
Management
activities.
I
created
two
epics
I,
wanted
to
start
to
count
the
number
of
manual
job,
retries
and
now
they're
epic,
to
start
to
count
to
measure
the
stage
duration.
So
what
metrics?
A
What
are
the
main
action
criteria
is
to
be
able
to
establish
a
Baseline
and
try
to
see
if
you're
able
to
improve
through
this
Baseline
on
state
duration
or
like
number
of
job
retries,
while
going
through
the
various
topics
and
everything
I
also
bumped
into
something
called
git
level,
cerability
stack
that
it
might
be
their
branding
of
opstrates
after
the
acquisition.
It
might
be
something
else
with
extra
addition
and
so
on.
So
for
this,
for
the
sake
of
dog,
fooding
I,
just
added
also
an
issue
there
to
do
a
kind
of
a
gap.
A
Analysis
between
the
two
products,
I
still
don't
have
a
very
good
overview
of
what
git
observity
stack
does
so
I'm,
not
sure
if
it's
something
that
could
make
sense
or
not.
But
if
anyone
has
some
extra
background,
the
ER,
please
feel
free
to
contribute
to
the
issue
where
they
are
another,
any
comments.
A
Alessio
do
you?
Do
you
want
to
assure.
B
Us
sure
so
I
was
reviewing
the
security
linked
and
I
couldn't
find
anything
related
to
get
observability
stack,
because
so
so.
The
first
thing
that
I
did
was
okay.
So
this
is
coming
from
Ops.
What
is
what's
up,
stress,
I,
think
the
acquisition
of
objects.
B
So,
let's
see
what
help
up
stress
is
doing
because
probably
is
going
to
be
something
built
on
top
of
that
and
in
that
regard
it's
it
really
sounds
like
we're
not
comparing
the
same
thing,
because
so
delivery
metrics
was
born
as
a
out
of
the
need
of
not
being
able
to
scrape
metrics
outside
of
a
CI
jobs,
because
there
are
patches
and
sorry
the
solution
for
debt
outside
of
delivery.
Metrics
is
the
push
Gateway
from
Prometheus,
but
then
here
we
had-
and
this
is
what
we
were
using.
B
But
then
we
started
building
metrics
based
on
histograms
and
those
metrics
were
just
ending
up
with
zero
or
one
at
the
end
of
each
Pipeline,
and
this
was
impossible
to
detect
counter
reset,
because
basically
so
the
thing
was
not
working,
and
so
we
ended
up
building
delivery
metrics
as
it's
kind
of
an
API
over
the
regular
Prometheus
metrics.
So
if
you
normally,
you
have
a
service
and
you
have
Prometheus,
and
then
you
can
do
it.
B
You
call
functions
on
it.
What
we
did
here
is
that
we
have
a
service
that
is
running
and
you
can
interact
with
this
over
rest
API
and
they
have
the
same
meaning
of
the
of
the
fact
function.
Call
that
you
can
have
with
the
with
the
Prometheus
Library,
so
it
unless
there
is
something
into
the
gitlab
observability
stack
that
goes
on
the
client
side
and
that
address
the
need
of
State.
I
must
say,
stateless,
not
even
stateless.
It's
not
even
running
so
batch
collecting
metrics
on
matches,
that's
okay!
B
This
is
not
the
same
thing
and
also
and
then
I
conclude.
They
were
also
telling,
in
the
upstairs
page
they're,
describing
this
as
a
something
you
can
put
in
between
Prometheus
and
I.
Don't
know
if
I
don't
remember
what
else.
So
the
idea
is
that
if
you
are
already
publishing
stuff
on
Prometheus,
then
you
can
put
the
thing
in
between
okay.
A
A
I
also
don't
I
found
actually
a
demo
video
that
they
publish
on
YouTube
just
a
couple
of
weeks
ago,
something
that
is
still
not
in
the
product
itself.
So
I'm
gonna
try
to
link
it
now
to
the
agenda
today.
So
we
can.
Everyone
can
go
through
that.
So
I
also
didn't
find
that
much
around
that
I
just
didn't
want
to.
C
B
B
There
is
a
readme
page
in
the
metrics
folder
of
release
tools,
but
may
not
be
up
to
date
with
what
we
did.
It
was
a
description
of
what
we
were
doing
in
the
first
version
of
it.
So
when
we
had
just
one
metric,
so
maybe
it's
worth
to
yeah,
that's
the
one!
Thank
you
we
can
so
let
me
quickly
check
how
out
of
date
it
is
well,
it
kind
of
explains
just
the
histograms,
which
were
the
first
thing
and
then
we
later
added
on
this
also
there's
an
extra
Matrix.
B
Now
that
is
scraping
values
from
oops
from
the
gitlab
API
on
oops.
So
no,
this
is
not
up
to
date
with
what
we
are
doing.
A
Then
I
guess
we
can
move
to
point
three
in
the
agenda.
That
is
the
various
metrics,
so
Risk
Managers
I,
guess
Ahmad
right
now.
Can
you
work
us
through
the
outer
plug
package
dashboard.
D
Yes,
I
hope
my
internet
can
basically
sustain
this,
but
let
me
try,
can
you
see
my
screen.
A
D
Stuff
so
last
week
we
had
a
good
start.
Basically,
we
had
like
Tech
packages
not
promoted
just
one
throughout
the
one
of
the
days.
Then
three,
then
you
know
one
and
this
week
we
have
like
we
started
already
like
yesterday
by
look
promoting
so
much
because
we
had
PCO
and
today
we
have
like
an
incident
I
think
almost
10
hours
now
so
this
week
is
gonna
be
bad
when
we
review
it.
D
The
active,
coordinated
Pipelines
today
also
we
are
getting
a
lot
of
Pipelines,
but
last
week
seems
normal
I.
Think.
D
B
Probably
want
to
move
this
over
the
last
month;
it
it
makes
more
sense.
So
you
have
this.
The
deployment
frequency
there
is
last
week
last
night.
Okay,
so
it
gives
you
a
better
idea
of
what
is
happening.
Yeah.
D
D
This
also
crcd
and
lead
time
last
month
to
be
honest,
I
don't
know
how
to
interpret
these
charts
yet
so.
B
But
I
think
they
are
normal.
So
so
there's
a
funny
thing
happening
here.
That
was
surprising,
so
we
had
a
PCL
last
Monday
and
while
this
is
clearly
visible
on
the
on
the
first
graph,
the
the
one
on
Prometheus
so
because
we
basically
we
are
shifting
everything
right
and
yeah
you're,
just
connection
mine
or
your
connection
is
yeah.
C
B
Thank
you.
So
there
was
a
PCL
at
the
beginning
and
we
also
started
real
I.
Don't
know
what
happened,
but
we
started
really
bad,
because
the
the
the
the
the
day
after
the
PCL
we're
just
Building
packagings
without
promoting
anything
yeah
and
then
we
we
recovered.
But
this
one
I
was
expect.
So
it's
it's
clearly
visible
down
and
the
active
coordinated
pipeline
that
are
just
skyrocketing,
so
we're
not
able
to
deploy
anything,
but
then
there's
no
real
trace
of
this
into
the
mttp,
which
is
the
last
graph.
B
You
were
looking
at
and
I'm
wondering
if
this
is
affected
by
the
family
and
friends
day,
because
usually
over
the
weekend,
we
accumulate
stuff
from
late
Friday
up
until
yeah.
Basically,
someone
maybe
merged
something
over
the
weekend,
but
having
family
and
friends
day
on
Friday
and
the
PCL
on
Monday,
this
kind
of
changed
the
numbers,
because
then
we
we
look
so
usually
on
Monday
in
this
case
on
Tuesday,
because
Monday
was
sorry,
Monday
was
PCL.
We
have
what's
the
name
in
the
graph
Elite
time.
B
I
think
we
have
a
highly
time,
which
is
very
it's
yeah,
this
one.
So,
as
you
can
see
everything
we
start
the
week,
we
it's
up
yeah,
it's
at
this
point
in
time,
it's
very
high,
but
this
didn't
really
happen
here,
even
though
the
the
spread
is
is
longer
because
we
had
Friday
family
and
friends
day
and
PCL
on
Monday,
the
the
value
were
still
kind
of
aligned,
which
is
funny.
A
Are
you
referring
to
the
last
family
of
friends
day
but
I
think
it
was
Monday
Friday
was?
It
was
a
normal
day.
B
Yeah
the
24th,
because
it's
Saint
John,
which
is
a
national
holiday
here
in
Florence
yeah.
So
that's
the
thing
but
yeah.
So
that's
funny,
because
we
we
never
look
at
those
numbers
together
and
I
was
expecting
also
having
a
PCL
higher
lead
time,
which
it
didn't
happen
and
then
a
question
for
release
managers
which
may
not
remember
what
happened
is
why
on
Tuesday,
we
were
not
able
to
deploy
for
basically
half
a
day
or
even
a
full
day.
B
I,
don't
remember
because
we
see
the
the
package
talk
going
up
together
were
not
promoted.
So
maybe
this
is
a
as
a
suggestion.
Try
to
annotate
things
when
weird
things
happen,
try
to
put
our
notes
on
this
graph,
because
otherwise
we
will
not
remember
one
week
after
because
the
this
is.
This
is
unusual
right,
so
it
should
have.
We
usually
go
between
two
one,
two
three
wasted
packages,
but
we
had
six
at
the
beginning,
so
it's
kind
of
it.
It
looks
like
something
like
today
right,
so
we
are
not
deploying,
and
so
we
are.
B
We
are
accumulating
the
number
of
packages
that
we
are
not
going
to
deploy.
So
if
we
don't
deploy
by
the
end
of
the
day,
we
will
end
up
with
six
seven
Target
package
and
six
seven
not
promoted.
C
Sorry
I
was
just
going
to
say
about
that,
lead
time,
increasing
on
Mondays
and
sometimes
not
increasing
I
wonder
if
it
would
be
good
to
track
the
number
of
Mrs
merged
in
in
the
git
lab
repo,
then
we
could
actually
see.
Maybe
the
Mrs
merged
went
down
because
lots
of
people
took
took
Friday
off
for
the
long
weekend
or
something
like
that.
B
But
this
is
not
affected
by
I
mean
number
of
merge
requests,
yes,
but
the
the
value
itself
is
not
affected
by
people
taking
the
day
off
because
it
comes
from
merge
time.
So
if
they
are
off,
they
are
not
merging.
So
it
will
not
affect
the
lead
time,
because
lead
time
is
from
merge
to
deployed
into
production.
B
Which
is
probably
why
it
wasn't
really
that
high,
because
people
were
not
working
and-
and
so
it
just
delayed
everything
but
I
mean
I-
probably
will
figure
this
out
over
time.
Once
we
get
more
familiar
with
the
shape
of
those
graphs
and
trying
to
figure
out
how
they
behave
and
change
together
with
how
we
work.
C
Yeah
I
meant
it
would
show
us
if
people
are
like
work,
merging,
more
or
merging
less.
It
might
not
directly
affect
the
value,
but
it
will
give
us
that
information.
B
Yeah
I
don't
know
if
it's
one
of
the
other
graphs
is
showing
something
like
this
I
I
don't
know
in
in
science
we
have
the
value,
maybe
they're,
just
because
we
we
see
the
the
list
of
merge
requests
down
in
the
tables.
We
never
look
at
them,
but
there
is
a
there's,
a
list
of
mergipus
and
it's
a
good
point.
Reuben.
C
About
the
poster
post
deploy
pipeline
so
once
these
three
issues
are
complete,
the
okr
is
done
for
this.