►
From YouTube: Grafana Community Call 2021-08-19
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
B
A
Perfect
so
hello,
everyone,
I'm
jio.
So
today
we're
gonna
meet
the
observable
squad,
which
is
us
here,
and
we
have
prepared
a
quick
like
a
short
deck
to
go
through.
We
represent
what
we're
working
on
and
what
are
our
our
responsibilities
and
we're
going
to
do
it
ourselves
and
please
feel
free
to
ask
any
question
you
you
might
have-
or
you
might
want
to
discuss
with
us
today.
A
So
first
thing
meet
the
squad.
This
is
us
quite
a
bunch
of
friendly
folks.
A
A
That
said,
what
do
we
do?
This
is
our
mission,
so
we
make
graphing
the
best
available
and
most
efficient
and
open
observability
platform.
We
beat
workflows
for
effective
troubleshooting
on
top
of
grafana
by
making
it
easy
to
correlate
lmg
signals,
and
we
provide
first-class
monitoring
experience
for
graphing
cloud
storage
back-ends.
A
A
We
we
also
maintain
a
bunch
of
different
data
sources.
I
mean
there
are
a
lot.
I
think
you
can
recognize
most
of,
if
not
all
the
the
logos
in
here
mostly
like
loki
temple
prometheus
but
also
graphite,
influx
jaeger,
elastic
zipping,
and
you
call
you
name
it.
We
try
to
develop
tools
to
to
enable
you
exploring
on
data
your
data
as
much
comfortable
as
possible
and
and
yeah.
That's
our
main
responsibility.
A
So
piotr
and
gabor
are
not
with
us
today,
I'm
gonna
present
for
them.
It's
been
a
while,
since
grafana
is
migrating
from
angular
to
to
react,
angular
actually
version
one,
so
pretty
old
version
creators
are.
I
will,
I,
I
think,
one
of
the
biggest
part
in
graffana,
like
the
at
least
the
the
one
that
users
tends
to
see
a
lot
during
their
workflows
and
migrating
them
to
react,
allows
us
to
move
faster
and
actually
to
be
up
to
date.
With
with
what
grafana.
C
A
And
what
we
want
grafana
to
be
tomorrow,
so
that's
something
that
maybe
might
not
be
recognized
as
much
as
as
other
things,
but
it's
very
important
for
us
to
keep
up
to
date
with
new
technologies
and
and
to
keep
grafana
a
modern
platform.
A
Other
things
we
are
working
on:
data
source,
migration,
backhand,
migration,
sorry
as
of
today
so
grafana
started
back
then,
as
a
front-end.
Only
application,
which
means
everything
was
running
through
the
to
the
front-end
and
query
where
were
being
run
and
being
built
in
the
front
end.
At
some
point,
graphing
introduced
a
back
end
which
allowed
us
to
have
other
thing
features
and
a
different
set
of
things
that
could
only
run
on
the
back
end.
A
A
A
So
we
are
currently
starting
migrating,
prometheus
and
elasticsearch
as
the
first
ones.
That
means
that
there
will
be
no
more
front-end
implementation
for
them.
That
should
be
so
quite
transparent
to
users
and
me,
ivana
and
dimitris
from
the
backend
platform
team
are
working
on
that
actively
now
and
we'll.
We
hope
to
get
something
ready
pretty
soon.
We
we're
starting
with
prometheus
and
we'll
move
forward
with
source
as
well.
D
D
So
one
thing:
what
whatever
on
recently
is
the
temp
service
map,
so
basically
right
now,
temple
can
only
retrieve
retrieve
like
single
traces,
but
we
are
building
together
with
the
temple
team
feature
to
actually
get
some
aggregated
data
over
all
your
services,
and
this
means
that
you
will
be
able
to
show
like
the
service
map
for
all
the
services
that
you
have
in
your
system
and
show
it
in
the
node
visualization
that
you
can
see
here
on
the
in
the
screenshot
and
and
yeah.
D
Basically,
this
could
give
you
like
an
overview
of
your
of
your
infrastructure
and
be
a
starting
point
to
like
explore
like
with
where,
where
you
see
your
problems
and
go
into
like
single
trace
after
that
or
get
some
metrics
for
some
specific
service.
That
was
that
has
error
rate,
for
example,
there's
already
on
on
main
main
branch
behind
the
feature
flag,
so
probably
will
be
ready
for
eight
two,
I
think
hopefully
yeah
conor.
Maybe
you
can
talk
about
a
little
bit
about
your
film.
E
E
C
C
So
previously
you
were
only
able
to
specify
the
the
minimum
step
interval
for
your
queries,
but
now
we've
added
two
more
two
additional
step
modes
as
we
call
them,
namely
a
maximum
step
mode
and
an
exact
step
mode
and
in
the
maximum
step
mode.
You
can
set
an
upper
bound
for
the
interval
between
your
data
points
and,
if
you're,
using
the
exact
mode,
the
step
interval
will
be
exactly
the
value
that
you
specified
as
and
yeah.
C
We
think
this
will
be
good
for
cases
when
let's
say
that
the
default
step
interval
is
15
seconds
and
then
in
this
interval
there
is
some
data
that
is
not
captured.
For
example
spikes.
Then
you
can
set
the
maximum
step
mode
to
set
an
upper
bound,
to
say,
let's
say
10
seconds
and
hopefully
you'll
be
able
to
catch
those
spikes
and
then
and
yeah.
This
is
also
coming
to
explore.
I
think
in
8.2
and
yeah
I
I
hope
this
will
be
a
neat
little
handy
feature
for
you
and
now
I'll
pass
it
over
to
ivana.
B
I
have
actually
a
bunch
of
improvements
that
are
part
of
graphene
8
or
8
1
releases,
and
if
you
are
using
graphene
for
visualizing
your
logs,
I
would
say
that
trying
out
the
newest
version,
girlfriend
eight
one
is,
is
perfect
option
for
you,
because
we
have
added
quite
a
lot
of
improvement,
so
you
can
basically
read
the
things
in
the
slide,
but
I
would
like
to
highlight
maybe
two
of
these
first
one
is
logs
pagination
with
caching,
so
in
graphene
8
we
have
added
this
new
pagination
to
explore,
which
means
that
if
you
are
searching
for
the
logs-
and
let's
say
you
are
searching
for
last
hour,
but
you
are
hitting
the
limits
which
you
can
set
in,
for
example,
elasticsearch
or
walkie,
you
can
use
a
button
to
request
newer,
logs
or
other
logs,
and
if
you
are
basically
going
through
pages,
we
are
caching
these
results.
B
So
if
you
go
to
the
previous
cache
previews
page,
these
results
are
cached
and
you
are
not
running
redundant
queries.
So
that's
a
very
nice
feature,
and
the
second
feature
that
I
would
like
to
highlight
is
the
prettified
json
log
and
it's
perfect
for
those
of
you
who
have
logs
in
the
json
format.
We
have
added
this
new
toggle,
where,
if
you
toggle
it
on
your
jsons
will
be
formatted
nicely
and
they
will
be
like
much
more
readable.
B
So
these
are
the
graphina
8
and
8.1
improvements,
and
on
the
next
slide
I
have
added
a
couple
of
things
that
we
are
currently
working
on.
The
biggest
thing
is
like
histogram,
so
we
got
a
lot
of
feedback
that
the
histogram
that
we
currently
have,
and
that
is
based
on
the
logs
that
we
receive
doesn't
work
that
well
for
our
users.
So
we
are
in
the
design
doc
phase,
where
we
are
trying
to
find
a
way
how
to
make
log
histogram
much
more
usable
and
great
for
for
our
users.
B
Other
things
that
we
are
working
on
is
highlight
of
influx
db,
logs,
a
low
key
result,
solution
for
the
metrics
queries
or
improved
fuzzy,
search
for
label
browser
and
loki.
So
that's
it
from
me.
A
Nice,
thank
you
and
more
so
other
things
we
have
on
our
roadmap.
So,
first
of
all,
you
can
check
the
the
roadmap.
It's
it's
public
on
github
feel
free
to
to
copy
that
url,
we'll
also
post,
that
I
think
in
on
youtube
under
the
description
of
the
video.
So
what
are
we
working
out?
So
what
we
want
to
bring
to
grafana?
We
want
to
have
the
ability
to
link
to
from
logs
to
trace
for
for
extra
for
the
x-ray
of
the
swords
we
want
to
create
dashboards
from
explorer.
A
We
have
this
workflow
now,
for
which
you
can
go
from
dashboards
to
explore
to
explore
your
data,
but
sometimes
we
have
the
feedback
that
people
are
just
exploring
their
data
and
they
find
some
something
useful
for
for
their
needs
or
they
they
they
build
a
query
that
they
find
very
useful,
and
this
will
allow
users
to
to
save
these
explore
state
or
export
visualization
and
queries
to
a
new
dashboards
or
maybe
to
the
library
panels.
That
was
a
recent
introduction
to
ingrifana,
then
yeah
mixed
us
fortune
explorer.
This
is
also.
A
This
has
been
a
very,
very
requested
feature
in
the
past,
and
we
also
want
to
introduce
the
trace
view
for
elasticsearch
for
choices
gathered
through
elastic
apm.
That's
also
very
quite
a
request
feature
and
yeah.
That's
that's
mostly
it.
Thank
you
very
much
and
please
feel
free
to
ask
anything.
You
would
like
to
to
ask
us
or
if
you
have
any
ideas,
sorry,
it
was
last
thing
I
almost
forgot.
We
also
want
to
introduce
continuous
profiling
to
grafana.
This
is
something
we
are
currently
investigating.
A
We
are
exploring
solutions
and
visualizations
types
if
you're
using
right
now
any
continuous
profiling
solution.
Please
let
us
know
either
during
this
call
on
on
youtube
on
in
a
comment
or
feel
free
to
to
join
our
publix
live
to.
Let
us
know
we
would
like
to
know
what
solution
are
you
going
to
using?
How
are
you
are
you
using
continuous
profiling
and
and
what
we
would
like
to
hear
from
you?
That's
what
drive
our
development
and
that
will
be
very
much
appreciated.
A
Yeah,
that's
that's
it.
I'm
gonna
end
the
presentation
now.
F
Thank
you
so
much
that
was.
That
was
a
lot
to
take
in
obviously
you're
working
on
quite
a
lot
these
days.
How
do
you,
how
do
you
determine
like
what
you're
working
on?
Is
it
only
through
github
issues,
or
you
know
how?
How
do
you
decide
your
roadmap?
Basically
so.
A
We
we
did
some
research
when
we
introduced
it
and
we
thought
it
would
have
been
a
good
solution
to
have
it
that
way.
But
of
course
when,
when
we
introduced
something
the
community
always
says
you
know
like
they
are
the
people
using
our
our
software,
so
they
they
know
more
how
they
want
to
use
grafana
than
probably
us.
We
take
a
lot
of
input
from
the
community
and
that's
that
for
sure
drives
our
our
roadmap
a
lot.
A
A
Sometimes
it's
us
doing
stuff
that
you
know
that
think
about
new
things
and
we
try
to
propose
it
and
to
start
sort
of
into
like
input
them
in
our
roadmaps
to
at
least
get
them
in,
in
a
point
of
which
the
community
can
can
actually
provide
us
feedback
so
that
we
can
either
you
know,
continue
developing
them,
or
maybe
it
wasn't
a
good
idea,
so
we
can
drop
them,
but
yeah
like
we
are.
We
are
very
community
driven.
I
would
like
to
say.
F
A
Sure
I
mean
if
anyone
from
the
team
wants
to
jump
in,
please
be
welcome
so
yeah.
The
thing
about
continuous
profiling
is
that,
as
of
now
graphics
is
great
to
look
at
phrases
to
look
at
logs
and
to
look
at
metrics,
but
sometimes
we
feel
our
user
wants
to
wants
to
dig
a
bit
deeper
in
what
is
happening
in
their
software
and
introducing
also
profiling
data
in
grafana
and
would
allow
them
to
really
understand
their
software
in
a
in
a
better
way.
A
So
that's
that's
the
main
reason
we
want
to
introduce
this
feature
in
in
grafana
yeah.
It's
actually
I
mean
it's.
It's
been
requested
also
by
the
community,
and
this
is
this
is
something
we
didn't
come
up
with
fully
so
yeah.
We
think
it's.
It
will
be
a
great
addition
to
graphing
in
general
and
to
the
observability
space.
F
F
So
if
anyone
else
has
any
any
questions,
just
jump
in
and
and
otherwise
I'm
looking
as
well,
if
there's
any
questions
in
the
agenda,
so
if
you
don't
feel
like
being
on
air,
so
to
say
just
post
your
your
question
in
the
agenda,
I
realized
I
haven't
posted
it
here,
so
feel
free
to
post
any
questions
in
the
document
that
I
just
posted
as
well.
F
So
you're,
I
really
I
mean
you're
one
of
the
biggest
teams
in
the
grafana
team
or
I
should
say
squad.
So
we
we
have
squads
in
the
grafana
team
and
you're.
Are
you
that
I
think
you're?
Actually,
the
biggest
one?
Aren't
you
no.
F
F
So
obviously,
there's
a
lot
of
stuff.
I
mean
we
really
believe
in
observability
and
taking
grafana
towards
a
more
you
know,
observing
making
grafana
more
a
platform
for
observability
and
that
kind
of
shows
in
you
know
the
amount
of
people
working
on
this.
So
it's
definitely
something
that
we're
gonna
see
a
lot
of
things
happening,
as
you
have
shown
as
well.
So
really
excited
about
that.
F
Is
there
anything
you
know
on
the
road
map
that
is
kind
of
that
you're
you're
still
evaluating
you
mentioned
a
continuous
profiling,
but
is
there
anything
that
you
know
you're
you're
still
you're
not
on
the
fence?
Is
there
something
that
you
know
the
community
could
weigh
in
on?
If
they
really,
you
know,
want
to
see
this,
but
you're
you're
you're
not
quite
sure
whether
to
pursue
that
yet
is
there
anything
you
know
interesting
and
on
the
horizon,
like
that
other
than
what
you've
already
shown.
D
Yeah,
maybe
I
can
yeah.
I
think,
like
one
interesting
thing
that
we
are
thinking
of
implementing,
but
maybe
will
not
come
to
actually
any
any
any
feature,
but
is
making
explore
like
a
little
bit
more
configurable
in
the
sense
of
how
you
can
basically
use
different
plugins
to
drive
like
your
dashboard
right.
You
can
use
basically
third-party
panels
and
visualizations,
but
we,
but
you
cannot
basically
do
that
in
expo
right.
D
So
we
are
thinking
about
creating
some
sort
of
api
that
will
allow
you
to
define
what
kind
of
visualization
do
you
want
to
have
in
in
explore,
not
in
the
sense
that
you
like
tweak
it,
but
basically
on
global
level.
You
could
say:
okay,
I
have
actually.
You
know
like
better,
better
idea,
how
I
want
to
see
c
logs,
for
example,
and
and
create
like
third
party
visualization
for
that,
so
maybe
you
could
just
switch
it
switch
it
on
in
in
graph
as
a
whole
and
then
in
explore.
D
Basically,
every
every
logs
response
will
get
this
visualized
in
that
in
your
panel
right.
So
we
are
thinking
about
enabling
that
it's
still
in
a
phase
where
it's
basically
just
right.
Just
a
rough
idea
so
yeah
if
anybody
has
any
any
kind
of
like
feedback
or
how
they
could
like
use.
C
D
F
So
that
sounds
a
little
like
you
know,
community
panel
plugins
in
explore.
Is
that
what
you
say?
I
sounds
really
exciting
to
me
at
least
as
a
developer
of
several
panel
plugins.
I
would
love
to
use
them
in
explorer
and
I
imagine
there's
others
who
think
the
same
way.
F
F
Maybe
we
can
call
it
for
today
unless
anyone
has
any
more
to
add
so
this
will
be
available
on
youtube
as
well,
and
you
know
if
you
can
join
here,
you
know,
please
let
your
colleagues
know
your
friends
know.
You
know
we're
gonna
go
through
and
the
entire
grafana
team,
all
the
squads
in
the
team,
so
you're
gonna
get
to
know
people
you
know
far
and
wide
to
really
get
to
know
the
people
that
are
working
on
grafana.
F
So
we're
very
excited
about
that
next
month,
we're
going
to
meet
the
ux
team
and
they're
going
to
introduce
themselves
and
talk
a
little
bit
about
what
they're
doing,
which
is
also
going
to
be
super
exciting,
so
make
sure
you
join
us
yeah
and
thank
you.
The
all
of
you
from
the
observability
squad
has
been
great
having
you.
This
was
really
useful
and
you
know
I
learned
a
lot
learning
about
the
roadmap
sounds
like
there's
a
lot
of
exciting
things
coming.