►
From YouTube: Grafana OnCall Community Call - 2022-06-22
Description
Join our next community call: https://github.com/grafana/oncall/discussions/categories/community-calls
0:00 Meet the OnCall team!
6:15 Helm Chart, Quick overview of latest updates.
15:30 Grafana OnCall developer setup. How to start working with a code?
35:00 Open discussion, Q&A.
A
And
we
are
live
so
welcome
everyone.
This
is
our
first
community
call
for
raphael
call
product
rafa,
now
open
source
incident
resolution
tool,
which
we
recently
announced,
and
we
have
a
small
agenda
for
today.
A
A
Athletes
will
have
a
session
about
how
to
set
up
development
in
the
environment,
how
to
configure
everything
locally
and
make
changes
and
build
features
fixed
box,
whatever
you
want
with
the
source
code
and
at
the
end
q,
a
session
so
majority
of
engineers
who
are
working
on
graphene
on
call.
They
they're
here
they're
ready
to
answer
your
questions
and
we'll
be
happy
to
discuss
and
listen
from
you,
etc,
etc.
A
So
yeah
my
name
is
matvee.
I
didn't
introduce
myself,
I'm
leading
the
project,
and
here
with
me
I
see
ilder
could
take
a
word.
B
Hi
folks,
my
name
is
hildar
and
I
was
with
the
project
from
the
very
beginning,
I
was
in
charge
of
back-end
infrastructure
a
little
bit
of
front-end
and
if
you
have
any
questions
on
any
decisions
that
has
been
made
been
made,
you
can
reach
to
me
I'll,
be
happy
to
answer
your
questions.
C
C
D
F
Yes,
hi
guys
yeah.
My
name
is
julia.
I'm
also
from
an
engineer
here
on
outcome.
Co,
team
yeah
super
excited
about
our
first
discussion
session
and
in
general,
our
open
source
version,
so
yeah
feel
free
to
ask
us
anything.
Next
is
michael.
G
Hey
I'm
michael,
I
work
mostly
on
devops
and
backhand
for
fun
on
call
and
yeah.
I'm
super
happy
to
be
here
as
well.
Who's
next
matias,.
H
I
Hi
everyone,
my
name,
is
ann
kinsey.
I
work
as
software
engineer
on
the
graphene
on
call
and
I'm
working
on
that
for
two
years
already,
I'm
responsible
for
all
kind
of
bacon
development
and
for
that
reform
provider
like
we
have
traffic
providers.
Somebody
didn't
know
that
so
glad
that
there
are
not
only
a
graphological
developers
in
the
school.
So
next
one
is
julia.
Artuchina.
J
K
A
A
E
B
B
We
provided
the
basic
home
chart
and
the
default
configuration
is
internet
mostly
for
proof
of
concept,
and
if
you
want
to
use
this
application
in
production,
we
really
recommend
you
to
move
all
the
stateful
services
to
some
pass
solutions
to
cloud
providers
etc
and
not
run
stateful
services
as
part
of
this
release.
But
for
today's
demo
I
will
be
running
our
default.
B
Configuration
basically
to
start
using
our
helm
chart
you
need
to
add
the
grafana
helm,
charts
repo
to
your
machine
and
after
that
you
just
need
to
run
the
helm
style
command
with
a
few
flex,
flack
weight
it's
needed
to
for
some
helm,
internal
things
like
hooks,
then
you
need
to
set
the
base
url
and
it
will
be
just
your
dns
name
also.
If
you
want
to
have
grafana
as
part
of
this
release,
you
should
tweak
some
microphone
configuration
as
well
with
the
dns
name.
B
Then
you
use
the
name
and
the
chart
name
in
refund
repo,
after
that,
after
running
helm,
install
in
a
few
minutes,
you'll
have
the
output
with
all
the
instructions
how
to
connect.
The
genus
record
took
how
to
create
the
dns
record,
how
to
access
the
grafana
that
was
installed
as
part
of
the
release
and
how
to
connect
grafana
on
call
backhand
that
we
just
deployed
to
the
grafana.
B
Hopefully
not
that
risky,
because
I've
already
run
that
command,
because
it
usually
takes
some
time
until
all
the
pots
are
run.
All
the
migrations
are
done
so
right
before
the
call
I've
run
the
command
helm,
install
that
I
was
showing
to
you
with
the
defined
dns
name,
ugari
dot
me
just
my
guinness
name
that
I
have
well
after
a
few
moments.
He
output
it
with
the
following
instruction.
B
It
tells
you
to
create
a
dns
record
that
points
to
the
specific
ap
address
and
we're
prompted
to
run
the
following
command.
Let
me
do
that,
so
I'm
running
keeps
yelling
grass
getting
grass
etc,
and
I
have
the
ip
address
which
I
have
to
copy
and
paste
into
the
dns
settings
of
my
dns
provider.
Let
me
actually
show
where
is
that
happening
yeah?
As
you
see,
I
have
the
dns
settings
for
my
business
name.
There
me
and
I've
created
the
record
a
record
with
a
host
at
and
as
a
value.
B
I
put
the
ip
address
that
I
was.
I
got
with
that
command.
That's
pretty
much
it!
Maybe
if
you're
doing
that
first
time,
you'll
need
to
have
a
couple
of
moments
until
this
dinosaur
is
propagated,
let's
go
back
into
the
terminal
and
the
next
part
we
need
to
get
the
password
for
the
fresh,
install
grafana
by
the
way
check
our
values.yaml
file.
B
B
B
B
B
Here
we
are,
we
can
start
using
our
interface.
For
example,
we
can
create
our
first
integration.
If
we
click
on
the
new
integration,
we
can
connect
even
the
existing
grafana,
but
it's
fresh,
so
you
probably
want
to
connect
some
other
monitoring
system.
Let's
use
the
most
generic
one,
formatted
webhook.
B
You
use
this
link
to
send
the
webhooks
to
your
new,
fresh
grafana
on
call.
Let
me
just
send
some
dem
alerts
just
using
this
button.
If
I
go
to
the
alert
groups,
we'll
see
that
they're
here
and
we
can
acknowledge,
resolve
and
continue
setting
up
refund
on
call
setting
up
new
integrations
new
escalation
chains
set
up
on
call
schedules,
connect
your
slack
or
other
messengers,
and
so
on.
B
Yes,
that
was
a
quick
demo
of
our
helm,
chart
and
production
installation
fourteen
monthly.
I'm
done
thank.
A
A
H
H
So
one
of
the
first
things
we
need
to
do
is
to
run
the
stateful
service,
the
stateful
services.
We
need
to
run
a
phone
call
in
context,
so
we
have
our
mycq
grafana
and
all
the
other
environment
services.
We
need
to
run
our
project
and
for
that
we
have
a
docker
compose
file,
which
is
the
developer
suffix
one,
and
we
need
to
get
everything
up.
H
H
Yeah
things
in
going
live
could
be
free,
okay,
we
have
our
services
running
there
and
this
could
take
a
bit,
but
everything
is
already
created
and
set
up,
and
the
idea
is
to
set
up
the
python
environment
in
which
we
are
going
to
run.
H
We
do
that.
We
don't
need
to
tweak
anything
here
at
first,
it's
just
a
few
environment
variables
that
you
may
need
to
change
if
you
need
to
work
on
some
specific
things,
but
the
default
values
should
be
okay,
so
we
have
that
and
to
make
sure
things
were
we
need
to
export
these
environment
variables
in
our
environment.
H
So
we
run
this
export
command,
which
will
export
the
environments
in
our
environment,
and
the
next
step
will
be
to
set
up
our
vehicle
and
to
work
with
the
python
dependencies
we
need
for
that.
We
need
to
confirm:
we
have
the
right
version
of
python.
In
our
case,
we
require
python
3.9,
at
least
so
we're
okay
in
that
front,
and
we
can
set
up
our
environment
where
we
are
going
to
install
our
dependencies.
H
Activate
environment
so
something
to
be
aware
and
take
into
account,
you
always
need
to
export
the
environment
variables
and
work
inside
your
virtual
level
because
things
could
fail.
Otherwise,
I'm
sure
I
will
forget
this
later,
but
I
mention
it
out.
So
I
try
to
remember
later,
and
so
we
have
our
problem
next
step
will
be
to
install
the
dependencies
for
that
we
have
the
requirement
files
inside
the
engine
directory.
H
Next,
we
will
be
setting
up
our
database
by
running
the
django
migrate
command,
which
will
populate
the
tables
in
our
database,
which
should
be
already
running
because
we
do.
We
did
the
docker
compose
that
something
I'm
not
going
through
here
is
installing
docker
docker
compose
or
nodejs.
I'm
assuming
you
have
your
environment
set
up
with
those
tools
because
they
are
not
part
of
this
demo.
H
E
H
Something
here
that
you
need
to
follow
in
instructions
is
what
the
port
in
which
we
will
want
the
service
to
be
running
and
something
else
that
is
not
in
the
instructions,
but
I
will
be
updating
later
today
is
like.
If
you
want
this
run
server
to
be
accessible
from
the
docker
container.
Where
you
are
running
the
grafana
instance,
you
will
need
to
expose
the
id
to
be
accessible
from
outside
by
running
run
server
with
the
oip.
H
H
Then
we
can
start
salary
by
running
the
jungle
command,
star
salary
that
will
start
solid
workers.
You
may
also
need
to
run
salary
bit,
but
that's
not
usually
the
case.
So
unless
you
need
it,
you
can
leave
it
that
for
later,
or
only
when
you
need
it.
So
we
have
tango,
running
selling,
rewards
running
and
we
have
it
started
also
the
docker
compose
starting
the
graphina
instance.
H
H
H
H
G
H
And
now,
if
we
refresh
here,
we
should
see
that
we
have
the
setup
available
and
we
have
this
now
set
fixed
values.
This
is
us
just
show
either
in
his
deployment.
We
need
to
set
up
these
urls
for
the
on-call
vacant,
the
graffana
url,
and
this
invitation
token-
and
in
this
case,
since
I'm
running
in
a
docker
container,
I
will
use
the
internal
name
or
our
own
call
instance,
and
I
need
to
generate
an
imitation
token.
H
The
token
I
will
start
server
and
then
we
have
the
token
the
on-call
service,
url,
android
url.
This
should
work
and
we
will
be
connecting
the
frontend
plugin
with
the
backend
service.
There
we
go
at
this
point.
You
have
a
running
phone
call
plugin
in
your
graphana
instance,
and
you
can
create
integrations,
send
alerts,
etc.
H
So
that
should
be
the
first
part
of
this
demo.
If
anyone
has
any
questions
or
comments
from
the
team
or
from
anyone
else,
this
could
be
a
good
time
to
ask.
Otherwise
I
could
show
a
little
demo
on
how
to
fix
a
small
bike,
so
you
get
a
look
of
how
the
code
is
organized
and
how
things
work
when
proposing
a
fix
for
something
sounds
good.
H
H
H
H
I
have
my
editor
with
the
on-call
project
open,
and
you
will
see
that
the
django
project
is
under
the
engine
directory
there.
You
will
find
the
apps
that
make
the
project
work,
particularly
I'm
going
to
the
public
api
app
right
now
and
just
to
find
the
url
we
are
looking
for.
This
is
the
users
api,
so
we
are
going
to
take
a
look
at
the
user
view
where
we
will
notice
that
the
query
set
that
will
return
for
the
user's
api
is
filtering
by
role
only
admin
and
editors.
H
H
H
G
H
H
H
H
H
C
H
I've
been
also
testing
by
test
with
the
parallel
running
plug-in,
which
makes
tess
run
a
bit
faster,
but
I
need
to
do
some
polishing
to
implement
it
here,
but
here
we
have
all
the
linkedin
stuff
passing
and
all
the
tests
should
pass
and
you
can
double
check
that
with
the
original
code.
The
test
reality
will
fail,
because
that's
what
you
expect
whenever
of
course,
one
time
error:
okay,
that's
a
new
one!
Something
failed
in
the
database,
not
sure
why,
but
that's
another
thing:
it's
not
related
to
our
stuff.
H
A
H
A
E
A
C
B
Well,
at
the
moment
we
started
the
project.
We
started
it
as
a
slack
application
and
the
best
sdk
was
written
in
python
for
django
and
also.
B
B
C
B
A
I
right,
oh
document
graphene
grafana,
do
not
see
on
call.
A
I
can
answer
no
it's
available
for
any
graphon
instance.
Graphene
call
is
distributed
as
separate
servers.
You
need
to
launch
so
run
like
docker
and
graphene.
Grafana
go
to
settings
plugins
find
on
call
launch
server
as
described
in
our
readme,
and
you
feel
free
to
use
it.
Even
if
you
don't
use
enterprise.
B
A
A
You
know
stupid.
Probably
we
could
do
it
better.
Somehow,
thanks
everyone,
let's
wrap
up
the
recording
of
the
meeting
will
be
in
youtube
and
if
you
watch
this
recording
in
youtube,
please
go
to
our
github
to
discussions
and
sign
up
for
next
community
call
we'll
have
them
every
month.
Thanks
everyone,
bye-bye.