►
From YouTube: Plan Error Budget Investigation Basics
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
This
is
a
quick
introduction
to
the
plan
error
budget
dashboards,
not
a
deep
dive
by
any
means,
but
hopefully
a
couple
of
tips
and
tricks
you
can
use
to
identify
where
you're
spending
error
budget-
the
dashboard
I
have
open
here
is
the
Stage
Group
dashboard
for
product
planning.
A
A
If
you
have
a
one-to-one
with
me.
This
is
the
dashboard
that's
linked
from
our
one-to-one
template
and
on
this
dashboard.
The
time
selector
actually
does
work
but
I'm
going
to
leave
it
at
28
days
for
a
minute.
You
can
see
it's
the
same
number,
so
there
are
a
few
things
going
on
here.
That
would
be
of
interest
to
me.
First
of
all,
something
appears
to
have
been
fixed.
That
was
causing
an
error
previously
prior
to
about
the
10th.
A
So
first
of
all,
I
probably
would
want
to
know
what
the
impact
of
this
is
on
the
overall
error
budget.
So
I
might
just
start
by
Framing
that
out
to
see
if
there's
any
change,
it's
99.88
currently
and
it's
still
99.88.
So
it's
most
likely
that
optx
is
the
significant
contributor
here
and
so
like
the
next
thing
might
be
to
shorten
the
time
window
here.
So
we
can
see.
Maybe
these
two
large
drops
here
so
I
could
just
do
this
and
take
a
closer
look
at
this
time
period.
A
A
Also
and
focus
on
these
Point
drops
so
even
I
would
maybe
go
in
and
just
really
Zone
in
on
this
one
just
to
see
so,
you
can
now
expand
service
level
indicators
to
try
and
identify
like
what's
actually
causing
this
optex
drop,
we're
looking
at
a
period,
that's
kind
of
three
o'clock
on
the
third.
A
You
can
solo
this
chart,
so
you
can
see
that
the
web
app
text
is
affected.
Also,
the
API
art
decks
seems
to
be
affected
as
well.
So
the
first
question
that
might
come
to
mind
for
me
would
be:
is
this
something
Global?
Is
this
something
that
other
teams
are
experiencing,
or
perhaps
the
whole
application,
and
in
order
to
figure
that
out?
There
are
some
links
up
here:
I'm
going
to
select
web
detail
I.
Suppose
you
could
look
at
API
detail
as
well.
A
So
it's
not
that
you're
absolved
from
any
further
investigation,
but
it
does
give
you
some
context
about
the
impact
of
the
problem.
It
could
be
still.
You
know,
a
change
made
by
your
group
or
a
small
number
of
groups
affected
that
that
drives
the
whole
system
availability
down
below
SLO,
but
at
this
point,
I'd,
probably
look
in
The,
Incident,
Management
Channel
and
try
to
figure
out.
A
Is
there
something
coinciding
with
this?
Perhaps
pressure
on
a
downstream
service
database
issue?
Redis.
A
You
know
something
like
that
coincides
with
this
and
that
might
give
you
more
information
before
you
go
switching
off
some
changes
that
have
been
made,
or
you
know,
having
some
false
attribution
to
some
change.
The
team
has
made
if
you
feel
that
it
is
likely
to
be
attributable
to
your
group,
or
you
just
would
like
more
information.
There
are
further
charts
below
that
will
be
useful,
like
SLI
detail
in
this
case,
I
think
we
identified
that
it
was
the
rails
requests
service.
A
So
there
are
further
charts
here
that
you
might
look
into
that.
Might
give
you
more
information,
this
one's
pretty
interesting,
mostly
because
it
shows
you
the
various
different
endpoints
that
are
spending
your
error
budget.
You
can
see
the
drops
here.
You
can
also
like
if
you
hover
over,
you
can
see
which
endpoints
are
the
least
performant
relative
to
their
target,
and
you
can
of
course,
solo
things
here
to
see
see
them
in
isolation.
A
So
you
may
do
that,
but
I
think
at
this
point,
if
I
really
thought
that
it
was
something
to
do
with
my
team
I
would
go
into
Cabana
I'm
not
going
to
go
into
that
now,
because
I
think
it's
it's
a
lot
more
detailed.
But
if
you
expand
the
budget
spend
attribution
section
you
can
see
here
within
this
time
period
and
you
can
shorten
or
lengthen
the
time
period
to
the
period
that
you're
interested
in
you
can
see.
A
You
know
which
services
are
spending
the
most
error
budget
and
in
this
case
web
appdx
is
the
top
one.
There
are
links
here
to
Cabana
if
you'd
like
to
investigate
further.
A
This
is
the
one
for
slow,
Optics
requests,
so
that's
the
one
I
would
go
into
about
as
I
said,
I'm
not
going
to
go
into
in
this
video.
So
hopefully
this
this
was
helpful
and
gave
some
initial
tips
on
how
to
use
grafana
and
how
to
use
their
budget
dashboards,
to
diagnose
a
problem.