►
Description
In this briefing, Eduardo Silva, Open Source Developer at Treasure Data will give us an overview of Fluentd and demonstrate how to use it with OpenShift & Kubernetes. Fluentd was recently added as a Cloud Native Computing Foundation (CNCF) member project.
Background:
Regardless of your environment, logging can be complex. System services and specific application logs need to be consumed different ways and the data retrieved likely comes in a variety of different formats, which presents an interest challenge. In the Cloud Native era, we see this complexity increase when deployment happens at scale. At this point having a non-generic logging tool is not enough to solve the problem. Instead a custom solution capable to integrate, understand and connect the dots between different end-points is highly recommended; that’s why Fluentd was created.
Fluentd allows you to implement an unified logging layer in any type of environment. It was designed with flexibility in mind, with a pluggable architecture of more than 600 extensions provided by the community, and can collect, parse, filter and deliver logs from any source to most of the well known destinations like local databases or cloud services. Find out more at http://www.fluentd.org/
Guest Speaker: Eduardo Silva, Open Source Developer at Treasure Data
A
A
Today,
to
have
the
folks
from
Treasurer
data
on
a
call
in
wardo
Silva
is
going
to
give
us
that
presentation
on
Susie
it's
a
bit
of
an
overview
of
Sandy
and
logging
fluently.
I
thought
that
would
be
such
a
good
baseline
for
everybody
to
do
this
on
one
of
the
projects
under
the
cloud
native
foundations,
umbrella
now
and
they're,
trying
to
make
sure
that
we
gave
people
time
to
everything
from
kubernetes
to
Prometheus
facility
and
the
groceries
most
of
the
projects
that
are
under
the
students,
es
umbrella,
history
and
different
reasons,
as
well.
A
So
without
further
ado
I'm
going
to
let
Eduardo
it
reduce
yourself
and
talk
about
something
you
can
ask
questions
in
the
chat,
we're
going
to
try
them
ahead
and
get
through
this
presentation.
So
if
you
have
like
a
burning
question
that
maybe
really
continues
to
make
sure
you
let
me
down,
but
otherwise
I'm
going
to
save
your
questions
for
after
seven
tation
and
then
they'll
be
in
forbidden
Q&S,
so
Eduardo
take
it
away
and
thank
you
very
much.
C
Okay,
any
Thank
You
Diane
for
deputation
to
participate
from
this
session.
My
name
is
Eduardo
Silva
I'm
from
the
fluency
team
at
pressure
data,
and
they
were
very
close
with
Co
developers
of
boondi.
So
we
always
try
to
see
how
we
can
improve
login
across
different
area.
Few
years
ago
was
mostly
on
on
standalone
services.
Then
we
start
the
integrated
how
to
fix
log
in
with
a
containers,
and
then
we
join
it
in
the
cognitive
space
with
a
work
with
orchestrators
flexibilities
and
other
systems.
C
So
when
we
talk
about
swindie,
we
need
to
understand
in
the
concept
of
logging.
In
the
experience
we
can
see
that
most
of
people
I
do
not
understand
very
well
about
what
logging
is
and
why
it's
important.
So
that's
why
I
want
to
introduce
a
little
bit
about
it.
The
first
thing
is
that
when
we
have
applications,
our
main
goal
is
always
to
analyze
how
these
applications
are
they.
But
in
order
to
accomplish
that,
we
need
to
some
mechanism
that
allow
us
to
collect
some
information
from
these
applications.
C
Otherwise,
analysis
will
be
quite
difficult,
so
the
normal
way
to
perform
some
analyses
is
that
from
the
beginning,
the
application
we
try
to
generate
some
log
files
from
any
kind
of
log
information
here
in
the
file
system
or
to
the
network.
And
then
we
try
to
centralize
that
information
in
some
kind
of
common
storage.
So
then
we
can
do
analysis.
But
in
order
to
get
to
that
point,
it's
quite
hard.
So
we
can
make
an
application
to
write.
Log
is
quite
easy,
but
centralized
the
log
is
a
bit
complex.
So
how
will
you
publish
that?
C
So
when
we
mentioned
about
a
community
or
logging
logging
pipeline,
we're
referring
to
this
scope
from
the
input
to
the
output,
and
we
have
many
phases
in
the
middle-
and
this
is
not
so
easy
to
handle,
because
there
are
many
problems.
For
example,
if
you
think
about
the
different
inputs
that
we
can
have,
you
can
say
that
maybe
the
syslog
messages
are
quite
different
from
Apache
web
servers,
a
log
messages,
so
the
input
side
needs
to
be
quite
a
flexible.
C
So
at
the
beginning,
a
few
years
ago,
this
is
was
a
quite
complex,
because
if
you
have
many
kind
of
inputs,
you
used
to
create
your
own
scripts
to
part
the
data
filter,
the
data
and
try
to
put
that
information
back
in
some
a
database
or
cloud
service.
But
the
problem
is
that
for
any
kind
of
new
input
sources
that
you
have,
you
try
to
create
a
new
script
or
a
new
program
or
attitude
cron
job
and
it's
complex.
That
is
a
solution
that
does
not
scale
in
Queen
architectures.
C
So
a
when
we
talk
about
login
and
then
we
start
talking
about
maker
services.
The
things
becomes
more
complex
because,
as
you
know,
everything
now
it's
mostly
running
in
containers.
Applications
run
in
containers,
so
it
contains,
has
its
own
dog
has
application
logs,
and
we
take
this
to
the
cluster
level,
it's
more
complex,
so
they
we
cannot
have
a
small
scripts
guided
in
data
if
we
have
more
than
dealing
of
continuous
line
to
the
wiki.
C
So
the
thing
is
how
we
can
solve
this
problem
so,
and
we
can
say
now
that,
after
a
few
years
of
work
with
different
companies
with
different
eight
users
of
adopters
who
containers
and
Club
solutions
we
can
set
up.
Windy
is
a
reliable
solution
that
fix
the
whole
logging
problem.
Wendy
was
created
initially
by
pressure
data
and
we
are
in
the
primary
sponsor
for
project,
but
they
are
also
other
companies
contributing
to
it,
and
we
donated
the
project
to
the
cognitive
computing
foundation
all
last
year
on
November.
So
right
now
is
tissue.
C
Data
is
in
primary
sponsor,
but
the
fridge
right
now
is
hosted
with
the
CNCs,
so
we
will
mention
about
the
login
pipeline,
so
we're
trying
to
see
how
to
fix
this,
and
we
can
say
that
the
whole
login
solution
now
is
crew
to
indie.
So
fluently
allows
you
to
take
the
whole
logs
from
different
inputs
and
centralize
that
information
back
on
any
kind
of
storage
solution
so
and
things
become
easier,
and
despite
that,
if
lundi
is
a
really
good
solution
for
the
problem,
also,
we
need
to
understand
how
things
works
internally.
C
Otherwise
you
cannot
scale
very
well
the
solution.
If
you
want
to
fix
a
problem,
you
need
to
understand
what
are
the
real
pain
so
about
20.
We
can
say
that
effluent
is
quite
big
right
now
and
despite
a
our
company
may
be
all
companies
we
maintain
around
20
plugins.
We
can
say
that
we
have
more
than
600
plugins
available,
most
more
than
500
made
by
the
community,
so
in
that
aside,
I
really
huge
wind
for
the
whole
system.
That
needs
a
reliable
solution.
Wendy
also
has
a
pluggable
detector.
C
That
means
that,
if
you
want
to
create
your
own
filters,
you
know
parses,
you
can
create
it
as
a
as
a
simple
plugin.
It
has
built-in
reliability
so
as
much
as
possible.
It
would
try
to
deliver,
did
collect
the
data,
deliver
the
data
and
try
to
done
do
not
lose
the
data,
and
we
have
native
integration,
of
course,
with
dogged
and
kubernetes
and
poety
is
reading
a
mix
of
Ruby
and
C
language.
The
most
critical
parts,
mostly
for
performance
and
data
civilization,
are
made
in
C
and
ecosystem.
It's
quite
B
right
now,
so
with.
C
Indeed,
you
can
deal
with
mostly
of
solution
on
most
of
bad
games.
In,
for
example,
we
can
deal
with
elasticsearch,
you
can
integrate
with
Splunk
locally
a
Google
cloud
services,
AWS
and
many
others,
so
they
think
is
that
in
yellow
over
here,
if
we
get
a
new
cloud
service
available,
a
new
back-end
there's,
always
someone
riding
up
flagging
for
flu
in
D
to
centralize
their
logs,
proble
and
cindy
has
a
modular
architecture.
That
means
a
when
I
was
talking
about
the
login
type
and
at
the
beginning.
C
C
Sometimes
if
you
are
collecting,
for
example,
one
terabyte
of
data
or
I,
don't
know
mm
mm
blocks,
maybe
you
don't
want
to
process
all
of
them,
so
you
want
to
discuss
some
of
them,
so
the
filter
plugins
allows
you
to
filter
data,
maybe
to
discard
some
specific
records
to
match
some
ones
or
also
to
modify
the
data.
For
example.
C
Maybe
if
you
are
collecting
data
from
syslog
messages,
you
would
like
to
obtain
the
hostname
or
any
specific
detail
for
the
machine
with
these
syslog
services
running
so
a
filter
plugin
allows
you
to
enrich
your
logs
or
discard
data
or
match
specific
logs
then,
when,
before
you
want
to
send
this
end,
the
data
that
is
already
filled
to
send
it
to
some
detonation
to
some
output.
You
also
would
like
to
buffer
this
data.
You
know,
there's
some
timing
associated
there's
a
flash
time
associated.
C
So
when
you
password
the
data
means
that
you're
going
to
start
temporary
this
data,
either
in
the
memory
or
in
the
file
system.
Most
of
the
really
enterprise
deployment,
a
complexities
in
the
file
system.
You
don't
want
to
have
it
all
everything
in
memory,
because
if,
for
some
reason
the
process
crash
or
the
container
cannot
continue
working,
you
could
be,
you
can
lose
data.
So
most
of
the
big
deployments
begin.
The
best
customers
try
to
use
buffer
system
with
the
file
system
and
every
record
that
goes
to
the
buffered.
C
C
This
is
myself,
look
and
then
I
click
and
create
some
rules,
so
the
router
then
can
decide
where
to
send
this
information
based
on
that
tag,
for
example,
elasticsearch
or
maybe
to
Google
cloud
platform
or
Amazon
s3,
and
then
we
have
the
record.
The
record,
of
course,
is
the
data
which
has
a
structure,
and
then
we
have
the
output
plugins
to
open.
C
So
internal
from
the
use
kind
of
JSON
structure
is
no
Jason,
because
it's
a
binary
version
which
in
message
back
format,
but
it
has
many
a
different
kind
of
sections.
So
the
data
becomes
very
easy
to
read.
So
you
can
understand
that
a
message
has
different
metadata
and
slowing
the
and
sort
of
login
it's
working.
In
a
lot
of
a
space.
We
started
with
a
integrate
magnetic
logging
driver
for
Joker.
It's
fully
integrated
with
kubernetes
right
now,
at
least
epsilonv
is
like
the
main
aggregator
for
elastic
search
and
Google
cloud
platform
for
Google
next
deployment.
C
You
can
deploy,
of
course,
and
all
looking
agents,
but
so
indeed,
it's
by
default
for
the
elastic
search
and
Google
platform
and
also
the
Google
cloud
platform
is
a
fact.
Driver
team,
so
aspect
driver
service,
use,
windy
as
the
main
agent
and,
of
course,
open
chip.
A
use
windy
after
main
aggregator,
which
is
great
also
funda,
has
been
integrated,
with
structure
login
on
across
different
sections
and
different
components.
So
a
right
now
I
would
like
to
talk
a
little
bit
about
how
fluently
integrates
with
Google
native
a
for-cause
familiar
with
open
chief.
C
You
know
that
open
chief
manager,
a
user
needs
to
manage
containers
and
orchestrate
everything.
So
the
way
that
we
install
su
in
the
inside
of
a
kubernetes
cluster
is
at
well.
As
you
know,
we
have
the
concept
of,
for
example,
the
API
server
and
we
have
the
node
and
each
node
has
different
ports
and
force
has
different
containers.
C
So
20
its
employee,
as
a
daemon
set
and
a
demon
set,
is
just
a
board
that
runs
on
every
node
from
in
your
cluster
because,
as
you
know,
in
every
node
that
you
have
in
your
system,
every
now
has
different
port
and
every
port
is
generating
different
log
services.
So
the
goal
is
that
you
have
fluent
it
deploys
a
demon
set
and
this
demon
set.
So
this
windy
as
a
amounting
bowling
with
the
whole
logs
and
start
collecting
the
logs.
C
C
But
also
it
does
a
mobile
solution
with
the
API
servers,
because
also
you
would
like
to
know
what
kind
of
metadata
and
what
kind
of,
for
example,
labels
and
annotations
are
associated
with
the
Todd,
where
these
containers
are
running
so,
and
this
is
possible
because
a
rehab
team
wrote
absolute
looking
only
this
metadata
filter.
So
when
70
it
is
in
solid
in
communities
in
your
own
cluster.
It
has
this
redhead
plugin.
C
C
So
we
can
say
that
20
is
quite
flexible
as
an
architecture,
but
also
the
plug
in
the
filter
is
making
a
good
job,
of
course,
that
we're
going
to
read
each
log
files,
we're
not
going
to
talk
to
the
API
server
every
time.
So
we
have
a
local
cache
for
every
information
that
we're
getting
and
also
when
you
have
to
indeed
trying
to
deliver
the
data
to
our
storage.
C
It
doesn't
matter
if
you
stretch
in
services
running
into
the
classes
outside
sometime,
we
can
face
some
issues,
some
network
outage
or
some
connection
problem
or
so
on.
Maybe
in
some
DNS
problem
so,
but
if
you
remember-
and
you
recall
the
image
one
of
the
initial
graphics-
you
can
see
that
we
have
a
buffer.
So
when
we're
going
to
send
the
data
lundi
is
always
reading
the
buffer.
Of
course
you
want
to
have
your
buffer
in
the
file
system,
but
if
something
happens,
for
example,
it
cannot
deliver
the
data.
C
You
can
configure
flu
indeed
to
perform
a
next
time
of
a
price
or
try
to
take
a
different
approach
for
this
kind
of
situation.
As
you
know,
on
every
classes
you
need
to
prepare
for
any
kind
of
failure
that
you
can
face,
because
you
don't
want
to
lose
data,
and
always
you
want
to
have
a
reliable
system,
a
process,
crashes,
application
crashes,
but
the
things
is
how
you
hold
what
you
can
do
to
deal
with
that
so
fooling
D
in
the
ons
case.
C
If
you
cannot
talk
to
elasticsearch
because
of
any
external
problem,
it
will
reply
and
Alexis
it
by
default,
and
we
can
you
can
say
you
can
specify
your
own
interval
for
that.
You
can
say
you
can
specify
an
exponential
interval
and
the
commentation.
There
are
more
options
in
to
deal
with
that
specific
thing.
Also,
you
can
say,
for
example,
if
you
cannot
talk
to
elastic
search,
you
can
say
you
can
try
to
implement
your
own,
a
fallback
mode
or
try
to
do
some
load
balancing.
C
So
you
know
in
that
way
you
cannot
just
put
the
whole
lot
on
just
one
service
and
when
so
nd
a
is
deployed
in
a
for
example,
in
a
ascend
team
on
set,
you
know
that
a
demon
said
for
kubernetes
is
always
a
gem
on
file
which
specifies
how
fluent
in
needs
to
be
deployed.
So
here
in
the
slides
which
end
you
will
have
access
later,
you
can
see
a
specific
and
open-source
project
repository,
which
has
like
a
template
for
their
fluent
Deaconess
demo
set
a
communities
already
come
to
sweet
windy
in
the
source
code.
C
So
you
have
your
own.
A
sorry
in
you
have
your
own
demo
set
by
default,
but
also
we
try
to
maintain
and
not
agnostic.
A
from
the
Goonies
demo
set
demo
files
with
different
rules
to
play
with
different
backends,
for
example,
which
closely
with
for
Amazon
s3
for
elastic
search.
So
so
we
have
maintained
more
configurations
that
we
can
find
in
the
the
bunny
like
when
it
is
a
source
code
and
now
a
who's
using
windy.
C
I
will
say
that,
basically,
if
everybody
who's
on
this
on
this
session,
if
you
are
using
open,
chief
you're
using
AF,
will
be
in
the
backend.
Also
one
of
the
biggest
users
is
Microsoft.
As
you
know,
Microsoft
has
a
system
which
is
called
operations,
managed
management,
suite
OMS
and
they
try
to
monitor
each
node
or
each
service
that
is
running
in
the
customers
platforms,
including
P.
C
It's
like
the
default
agent
to
collect
data
from
every
service
on
every
application
that
is
running
on
those
nodes,
so
the
OMS,
a
logging
agent,
it's
build
is
fluent
D,
plus
some
specific
plugin
solution,
and
we
can
find
this
model
in
different
ways.
For
example,
Google
is
using
Lindy
as
a
main
agent
Microsoft
and
work
right
now
is
open
to.
C
For
example,
if
you
are
writing
your
own
application
go
length,
we
have
a
goal
and
a
package
for
fluency,
so
you
can
make
your
own
application
talk,
deeply
doing
the
inset
of
let
fluidity
to
consume
the
looks
from
the
file
system,
so
it
can
talk
to
a
network,
but
also
I
want
to
say
with
this
that
Mundi
has
many
components,
because
we
just
not
try
to
consume
log
files.
We
try
to
fix
the
problem
of
a
login
pipeline
in
a
complete
way.
C
C
I,
take
cares
about
to
go
check
the
logs
from
some
point
and
send
this
log
to
an
aggregator
and
every
Gator
is
is
methanol
that
a
full
solution
that
has
really
strong
buffering
cat
abilities
so,
for
example,
a
cool
indie
can
work
as
both
as
an
aggregator
and
as
a
forwarded,
also,
if
you
just
just
windy,
you're
using
both
modes
at
the
same
time,
but
also
there
is
some
problems.
So
if
you
have
many
a
nodes,
this
hasn't
cost,
as
you
know,
handle
logging
and
parse
log
is
not
cheap.
C
This
has
a
cost
personal
string
or
plain
regular
expression.
This
is
quite
expensive.
You
have
computing
time
and,
of
course,
you
can
reflect
that
at
the
end
in
your
build.
If
your
cluster
is
quite
big
and
you're
running
on
AWS
or
Google
cloud
platform,
you
can
say
that
that
pay
some
money
to
to
sustain
that.
So
it
depends
on
how
you
configure
things.
It
depends
how
much
resources
you're
going
to
consume,
but
what
happens
when
you
have
many
nodes
and
these
nodes
start
and
start
to
grow.
C
So
if
fluent
in
records,
for
example,
is
a
minimum
of
40
megabytes
to
run
the
average
is
200
megabytes
a
Negroni,
this
cluster
per
node,
if
you
deploy
a
quite
hand
with
500,
will
be
quite
expensive.
So
also
from
a
few
indie
team
perspective,
we
are
trying
to
see
how
to
help
our
end
users
to
reduce
the
cost
of
their
own
deployments.
So
one
solution
that
we
come
up
is
to
separate
in
the
log
forwarders
from
the
log
aggregator.
C
That
means,
for
example,
20
works
of
both,
but
what
would
happen
if
we
create
a
separate
and
lightweight
load
for
whether
to
make
things
cheaper,
so
I
went
into
this
very
quickly
to
the
plate,
which
is
called
Coulomb
bit,
which
is
like
a
child
break
of
windy
reading
from
scratch
to
try
to
solve
in
this
problem
in
fluid,
it's
part
of
a
flow
in
the
ecosystem,
and
so
indeed
is
also
they
also
under
the
CNCs
so
flowing
bead.
It's
a
solution,
completely
redundancy.
C
It
tries
to
be
very
aligned
to
about
how
their
architect
of
flu
in
the
east,
so
it's
super
plug
in
it,
has
built-in
reliability
and
it's
fully
event-driven
and,
of
course,
that
means
that
if
that's
a
lot
of
a
synchronous,
I/o
operations
over
the
network,
so
what's
windy
as
a
forward.
The
good
thing
is:
is
that
so,
and
it
can
also
support
many
kind
of
inputs
can
filter
the
data
and
also
support
different
destinations.
C
It
has
a
built
in
person
support
that
means
that
you
can
take,
for
example,
unstructured
text
messages
and
give
it
a
structure
and
at
the
same
minimum
it
requires
no
more
than
500
kilobytes.
Of
course,
if
you
are
doing
a
lot
of
parts
in
a
lot
of
things,
this
requirement
will
increase,
but
it's
a
huge
win.
If
you
have
a
big
deployment,
you
have
a
lot
of
nodes,
so
it's
times
better.
C
So
the
approach
that
we
are
trying
to
test
right
now,
affluent
babies-
and
we
have
already
some
early
adopters-
is
trying
to
put
cooling
bit
inside
a
the
most
critical
nodes,
the
mid
performance
and
try
to
keep
their
memory
usage
with
quite
low
and
make
fluent
big
sake.
Ok
to
flu.
Indeed,
so,
on
this
case,
doing
peep
it's
working
as
a
forward
and
flu.
Indeed,
just
an
aggregator
and
fluently
it's
taking
care
to
store
the
logs
in
a
reliable
way
in
their
own
back-end.
C
So
with
this
model,
we
can
have
a
cheap
forwarding
of
logs.
So
all
of
these
is
mostly
about
strategy.
There's
no
effective
solution
for
each
use
case,
but
we
can
find
that
a
different
people
has
different
problems
and
different
problems
need
to
be
addressed
in
different
ways:
stop
fooling
bit
and
it's
to
solve
the
problem,
high
memory,
consumption
and
try
to
reduce
the
cost
in
the
cluster
in
deployment
and
when
they
come
over
one
of
the
complex
things
about
when
dealing
with
cloud
native
features
in
fat,
which
is
called
back
pressure.
C
That
means
a
back
pressure
is
mostly
it
comes
in
about
when
you
get
a
lot
of,
for
example,
in
on
this
specific
context,
when
you
get
a
lot
of
data
from
your
input,
but
you
cannot
switch
the
data
out
at
the
same
rate.
Of
course,
you're
going
to
be
get
some
back
pressure,
because
you
cannot
send
the
data
out
as
much
as
fast
as
it,
so
you
would
like
to
so
what
that
means.
C
If
we
take
this
example,
the
weather
we
go
out
and,
of
course,
we're
going
you're
going
to
get
some
back
pressure
because
you
cannot
deal
with
that.
The
output
can
not
deal
with
that
kind
of
data
ingestion
and
be
happens
mostly
with
different
kind
of
packing
databases,
and
mostly
there,
the
cloud
service
that
you
are
going
to,
where
you're
going
to
send
your
logs
and
mode
as
it's
quite
a
time
so
slower
that
the
time
that
you
can
send
your
own
logs
so
dealing
with
back
pressure,
it's
most
is
quite
easy.
C
If
you
domain
implemented
it
in
the
right
way,
so
fluently
and
flow
embed
implement
the
back
pressure
in
solution
in
which
it
will
not
ingest
more
data
until
this
data
can
be
ingested.
Of
course,
he
has
some
be,
has
some
pros
and
cons,
but
since
in
our
class,
from
our
context,
a
sub
soft
from
our
kubernetes
context
point
of
view.
This
is
quite
simple
because
since
we
are
just
reading
log
files,
the
log
files
becomes
our
buffers
okay.
C
So
we
are
not
going
to
consume
more
log
files,
so
we're
not
going
to
load
more
data
from
the
log
files
and
we
can
flush
the
data,
and
once
you
can
flush
the
data,
then
we
can
ask
to
ingest
mo
data
from
the
log
files
again
to
the
backend.
So,
with
this
mechanism
we
can
solve
a
bag
for
evaporation
and
difference
in
different
system
from
different
users
are
different,
and
sometimes
you
cannot
have
just
a
default
configuration
for
everybody.
C
So
a
for
pool
embed
it
by
hand
to
back
pressured
Hospital
in
security
things
it
also
support.
It
has
its
own
filter
together
when
it
is
metadata-
and
we
are
now
in
this
version
suit
at
11
who
in
bid
will
turn
two
years
in
now
into
light,
and
then
expiration
suit
at
12
will
support
a
timestamp
with
nanoseconds
a
set
of
fractional
seconds
in
and
a
second
unit.
C
So
at
the
moment
we
don't
have
a
full
documentation
how
to
write
plugins
offering
bid.
But
if
you
look
at
the
examples
that
we
have
is
quite
straightforward
and
if
you
want
to
deploy
full
and
bit
as
a
kubernetes
team
on
set,
you
have
to
test
how
it
can
behave,
a
compare
with
windy.
We
have
a
full
repository
and
a
full
token
image
that
has
the
same
pretty
much
the
same
configuration
that
the
flew
in
dig
in
team
on
set
and
is
by
default.
It
also
found
it
can
talk
directly
to
elastic
search.
C
C
We
agree
to
elastic
search
and
if
you
want
more
information
about
wounded
and
the
break
is
flim
bid,
that
IO
is
fully
open
source
Apache
License,
it's
under
the
CM,
safe
as
I
said,
and
also
the
hopeful
impede
community,
is
under
our
main
slack
channel
and
right
now,
I
would
like
to
present
to
an
Regatta
who's.
Our
product
money
for
flowing
the
enterprise
who's
going
to
give
a
really
small
presentation
about
what
are
the
enterprise
features
are
coming
for.
Funding.
B
My
name
is
honor
ago,
as
Eduardo
said,
and
I
also
work
at
treasurer
data
and
I
am
the
product
manager
for
fluency
enterprise,
so
just
kind
of
going
as
an
overview
about
the
messages
that
Eduardo
said
is
hey.
You
really
need
logging
at
a
unified
layer.
We
need
something:
that's
reliable,
something
modular,
that's
vendor
agnostic
in
the
backend,
that's
flexible
and
and
that's
all
great
and
fluency
really
addresses
a
lot
of
those
issues.
B
But
what
we've
noticed
from
both
treasurer
data
and
my
background
from
being
at
Microsoft,
is
there's
a
lot
of
large
scale
deployments
that
require
some
some
additional
features
and
security.
They
require
support
and
deployment.
They
need
some
best
configuration
practices
and
SLA
is
associated
with
those
backends,
and
so
that's
where
a
who
in
the
enterprise
comes
in-
and
you
can
see
here
from
the
picture-
it's
it's
really.
B
It's
built
on
that
same
open
core
flinty
platform,
but
with
some
additional
output
plugins
for
enterprise
for
any
backends
likes
plumb,
and
you
have
things
like
our
own
treasure
data
service
and
then
we're
really
making
sure
that
we
are
adding
features
that
are
really
rich
for
enterprise
and
security.
Space,
if
you
want
to
click
Next,
Award,
Oh,.
B
Perfect,
so
pretty
blanket
slide
here,
but
just
to
go
over.
It
is
intense
security,
fluid
enterprise,
there's
a
really
powerful
buffering
mechanism
and
we've
added
encryption
to
that,
so
that
one
source
comes
in
from
say,
a
firewall.
Not
everyone
can
read
that
data
today,
including
T.
It's
just
a
message:
pack
format
stored.
So,
if
you
can,
you
know,
take
a
look
at
the
buffer
file.
You
have
access
to
it.
You
can
see
all
the
data
that's
flowing
through
the
certified
enterprise
plugin.
B
These
are
both
source
and
inputs,
for
example,
for
Splunk
we
do
a
big
benchmark
with
our
fluid
enterprise
bits.
We
make
sure
that
we
can
run
at
you
know
a
thousand
ten
thousand
hundred
thousand
messages
per
second
give.
You
performance
count
configuration
around
that
as
well
as
some
of
the
cpu
layered
with
it
I'm
world-class
support.
So
this
is
the
makers
of
fluent
dr
with
us.
We
have
folks
who
have
written
message
pack,
we
have.
You
know
ed
Wardle,
who
is
craig
fluent
bit.
A
Perfect
timing,
thanks
very
much
guys
for
their
doing
this.
If
there's
one
question
that
someone
just
athle
is
asking:
if
there
are
pieces
for
correlating
entries
from
various
sources
to
identify
events,
don't
cause
analysis,
centers
and
those
our
enterprise
features
or
are
typically
provided
by
third-party
projects
or
products
right.
B
I
would
put
that
as
more
of
a
analytic,
back-end
Mundi,
it's
great
at
unifying
the
log
streams
for
correlating
entries.
So
you
can
take
your
events
from
your
web
server,
your
application
server,
your
stack
traces,
throw
it
all
into
one
analytic
back
end
and
hopefully
that
one
analytic
back-end
can
do
some
of
the
things
you
said
around
the
bed
store
and
through
cost
analysis
into
in
terms
of
event,
storms.
A
C
Actually,
a
use
in
like
two
years
ago
used
to
be
mostly
the
mailing
list.
They
made
a
list.
I
would
say
that
has
a
lot
of
traffic
and
with
most
of
the
flinty
teams,
trying
to
respond
most
as
possible
a
but
also
we
have
seen
in
the
last
year
huge
grow
on
our
slack
channel.
What
so
inflectional
we
have
like
600
members,
but
mostly
a
handle
directly
so
right
now
in
the
position
where
you
can
go
either
to
slag.
C
If
you
have
more
complicated
things
and
you
want
that
more
people
look
at
your
problem
and
the
mailing
list
also
say
good
resources
and
you
can
find
it
on
the
flow
indeed
at
work
side
in
this
fourth
page
in
this
community
or
in
the
community
section,
there's
the
mailing
list,
which
is
a
google
group
and
more
information
about
different
or
channels
that
could
be
treated.
But
most
of
the
big
issues
are
how
to
bide
mail
lists,
and
second
s,
slack.
A
A
Events
like
this,
but
the
flat
channel
is
where
people
are
really
connecting
these
days
and
that's
probably
been
true
for
us
for
common
ownership,
Commons
is
slacked
up
or,
and
it's
pretty
pretty
active
these
days
so
I
think
that's
probably
a
good
place
to
pause
and
stop
on
and
I'll
put
the
recording
of
this
up
on
YouTube
in
a
day
or
so,
and
if
you
could
send
me
the
PDF
version
of
this
slide.
Deck
I
will
post
that
as
well.
We
want
an
on
a
blog
post
on
blog
that
OpenShift
comm.
A
What
I
think
might
be
interesting,
too,
is
to
do
a
follow
on,
and
maybe
demo
excuse
me
using
fluent
the
enterprise
and
actually
do
some
live
demo.
Showing
it
off
to
I
did
great
things.
We
did
the
follow
on
in
the
film
month
or
so
post
Red
Hat
from
that
post,
the
next
openshift
Commons
Gathering,
which
is
May
1st
in
Boston
or
co-locating
that
or
any
of
the
indeed
treasure
data
folks
registered
yet
set
up
seen
your
name
on
the
list,
but
about
300
attendees
are
ready.
A
All
right,
everyone
on
the
call
there's
a
bunch
of
you
there
being
very
quiet
lurking
if
there
any
other
questions
and
it's
not
going
going
gone.
Thank
you
again
for
a
wonderful
review
of
fluency
I
hadn't
heard
about
the
fluent
bit
part,
so
that
actually
was
really
cool
for
me
and
I
hope
everybody
else
enjoyed
it
and
reach
out
for
long
contribute
to
fluency
and
ask
questions
and
honors
and
Eduardo,
which
is
awesome.
So
thank
you
very
much
that
thanks.