Slide 1
Slide 1 text
metrics,
monitoring,
logging
mathias meyer, @roidrage
http://paperplanes.de
Slide 2
Slide 2 text
No content
Slide 3
Slide 3 text
No content
Slide 4
Slide 4 text
No content
Slide 5
Slide 5 text
No content
Slide 6
Slide 6 text
problem?
Slide 7
Slide 7 text
no one noticed
no one got alerted
no automatic recovery
Slide 8
Slide 8 text
it happened to me
it happened to you
Slide 9
Slide 9 text
devops shmevops
Slide 10
Slide 10 text
your code, your responsibility
Slide 11
Slide 11 text
what is your application doing right now?
Slide 12
Slide 12 text
do you know when it fails?
Slide 13
Slide 13 text
failure means customers lose trust
Slide 14
Slide 14 text
failure means customers go elsewhere
Slide 15
Slide 15 text
failure means you lose money
Slide 16
Slide 16 text
application = providing value
Slide 17
Slide 17 text
monitoring
metrics
logging
Slide 18
Slide 18 text
monitoring
Slide 19
Slide 19 text
is the application available?
Slide 20
Slide 20 text
pingdom
pagerduty
nagios
icinga
sensu
sheriff
Slide 21
Slide 21 text
pingdom
Slide 22
Slide 22 text
http://pingdom.com
Slide 23
Slide 23 text
tcp/ip
http(s)
ping
Slide 24
Slide 24 text
nagios
Slide 25
Slide 25 text
nagios can check everything
Slide 26
Slide 26 text
it's still terrible
Slide 27
Slide 27 text
http://www.nagios.org/
Slide 28
Slide 28 text
#monitoringsucks
Slide 29
Slide 29 text
sensu
http://www.sonian.com/cloud-monitoring-sensu/
Slide 30
Slide 30 text
sheriff
https://github.com/dawanda/sheriff
Slide 31
Slide 31 text
monit
runit
bluepill
god
upstart
Slide 32
Slide 32 text
is this service currently providing value?
Slide 33
Slide 33 text
is this service consuming too many resources?
Slide 34
Slide 34 text
monit
Slide 35
Slide 35 text
check process unicorn
with pidfile /var/run/unicorn/unicorn.pid
start program = "/etc/init.d/unicorn start"
stop program = "/etc/init.d/unicorn stop"
if mem is greater than 300.0 MB for 1 cycles then restart
if cpu is greater than 50% for 2 cycles then alert
if cpu is greater than 80% for 3 cycles then restart
group unicorn
http://mmonit.com/monit/
Slide 36
Slide 36 text
bluepill
Slide 37
Slide 37 text
Bluepill.application("unicorn") do |app|
app.working_dir = "/var/www/app/current"
app.process("unicorn") do |process|
process.start_command = "/etc/init.d/unicorn start"
process.stop_command = "kill -QUIT {{PID}}"
process.restart_command = "kill -USR2 {{PID}}"
process.stdout = process.stderr = "/var/www/app/current/log/unicorn.log"
process.pid_file = "/var/run/unicorn/unicorn.pid"
process.checks :mem_usage, :every => 10.seconds, :below => 300.megabytes, :times => [3, 5]
process.start_grace_time = 10.seconds
process.start_grace_time = 10.seconds
process.restart_grace_time = 10.seconds
process.checks :flapping, :times => 2, :within => 30.seconds, :retry_in => 7.seconds
process.monitor_children do |cp|
cp.checks :mem_usage, :every => 10, :below => 400.megabytes, :times => [3, 5]
process.checks :cpu_usage, :every => 10.seconds, :below => 50, :times => 5
cp.stop_command = "kill -QUIT {{PID}}"
end
end
end
https://github.com/arya/bluepill
Slide 38
Slide 38 text
runit
Slide 39
Slide 39 text
#!/bin/sh
cd /var/www/app/current
./bin/unicorn_rails -c config/unicorn.rb -e production
http://smarden.org/runit/
Slide 40
Slide 40 text
metrics
Slide 41
Slide 41 text
No content
Slide 42
Slide 42 text
measurements
historical data
graphs
Slide 43
Slide 43 text
how many customers are on my site?
Slide 44
Slide 44 text
how many customers were on my site yesterday?
Slide 45
Slide 45 text
how slow is paypal's api?
Slide 46
Slide 46 text
how slow was paypal's api yesterday?
Slide 47
Slide 47 text
how much memory is available on my servers?
Slide 48
Slide 48 text
how much has memory usage grown over four weeks?
Slide 49
Slide 49 text
number of open database connections
number of redis commands
number of 500 errors
rate of HTTP requests
number of HTTP connections
median response time
Slide 50
Slide 50 text
number of failed resque jobs
number of twitter followers
99th percentile github api response time
95th percentile mysql query time
deployments
Slide 51
Slide 51 text
cpu usage
incoming network traffic
load average
disk usage
iops
Slide 52
Slide 52 text
munin
ganglia
graphite
scout
server density
librato metrics
Slide 53
Slide 53 text
munin
Slide 54
Slide 54 text
http://munin-monitoring.org/
Slide 55
Slide 55 text
ganglia
Slide 56
Slide 56 text
http://ganglia.info/
Slide 57
Slide 57 text
#monitoringsucks
Slide 58
Slide 58 text
#rrdtoolsucks
Slide 59
Slide 59 text
No content
Slide 60
Slide 60 text
access to single data points matters
Slide 61
Slide 61 text
graphite
Slide 62
Slide 62 text
modern graphing
not using rrdtool
extensible
http://graphite.wikidot.com/
Slide 63
Slide 63 text
graphite dashboards
Slide 64
Slide 64 text
https://github.com/ripienaar/gdash
Slide 65
Slide 65 text
https://github.com/paperlesspost/graphiti
Slide 66
Slide 66 text
https://github.com/obfuscurity/tasseo
Slide 67
Slide 67 text
cube & cubism
Slide 68
Slide 68 text
http://square.github.com/cube/
Slide 69
Slide 69 text
commercial tools
Slide 70
Slide 70 text
newrelic
http://newrelic.com
Slide 71
Slide 71 text
scout
http://scoutapp.com
Slide 72
Slide 72 text
server density
http://serverdensity.com
Slide 73
Slide 73 text
boundary
Slide 74
Slide 74 text
http://boundary.com
Slide 75
Slide 75 text
librato metrics
Slide 76
Slide 76 text
metrics as a service
resolutions to the second
real-time updates
Slide 77
Slide 77 text
http://metrics.librato.com
Slide 78
Slide 78 text
No content
Slide 79
Slide 79 text
No content
Slide 80
Slide 80 text
No content
Slide 81
Slide 81 text
collectd (honorary mention)
http://collectd.org
Slide 82
Slide 82 text
riemann (honorary mention)
Slide 83
Slide 83 text
http://aphyr.github.com/riemann/
Slide 84
Slide 84 text
track everything that moves
Slide 85
Slide 85 text
No content
Slide 86
Slide 86 text
adding metrics should be easy
Slide 87
Slide 87 text
statsd
https://github.com/etsy/statsd
Slide 88
Slide 88 text
metriks
https://github.com/eric/metriks
Slide 89
Slide 89 text
counters
meters
timers
Slide 90
Slide 90 text
Metriks.meter("travis.github.requests").mark
Slide 91
Slide 91 text
Metriks.counter("travis.repositories").increment
Slide 92
Slide 92 text
librato metrics
log stream
graphite
proc title
Slide 93
Slide 93 text
percentiles > averages
Slide 94
Slide 94 text
dashboards
Slide 95
Slide 95 text
combine graphs
Slide 96
Slide 96 text
put them up in your office
Slide 97
Slide 97 text
visibility is important
Slide 98
Slide 98 text
logging
Slide 99
Slide 99 text
the papertrail
Slide 100
Slide 100 text
#syslogsucks
Slide 101
Slide 101 text
collect logs from everywhere
Slide 102
Slide 102 text
index, aggregate, analyze
Slide 103
Slide 103 text
grep, awk, sort
Slide 104
Slide 104 text
No content
Slide 105
Slide 105 text
centralized logging
Slide 106
Slide 106 text
syslog://
Slide 107
Slide 107 text
logstash
http://logstash.net/
Slide 108
Slide 108 text
log inputs
process
outputs
Slide 109
Slide 109 text
graylog
Slide 110
Slide 110 text
http://graylog2.org/
Slide 111
Slide 111 text
loggly
Slide 112
Slide 112 text
http://loggly.com
Slide 113
Slide 113 text
papertrail
Slide 114
Slide 114 text
https://papertrailapp.com/
Slide 115
Slide 115 text
integrates with librato metrics
Slide 116
Slide 116 text
No content
Slide 117
Slide 117 text
bits and pieces
Slide 118
Slide 118 text
travis metrics
Slide 119
Slide 119 text
No content
Slide 120
Slide 120 text
https://github.com/eric/metriks_log_webhook
Slide 121
Slide 121 text
lograge
Slide 122
Slide 122 text
sane rails logging
Slide 123
Slide 123 text
No content
Slide 124
Slide 124 text
https://github.com/mattmatt/lograge
Slide 125
Slide 125 text
#monitoringsucksless
Slide 126
Slide 126 text
own your monitoring
Slide 127
Slide 127 text
own your metrics
Slide 128
Slide 128 text
own your logging
Slide 129
Slide 129 text
none of them is optional
Slide 130
Slide 130 text
go forth and correlate
Slide 131
Slide 131 text
http://www.paperplanes.de/2011/1/5/the_virtues_of_monitoring.html
http://about.travis-ci.org/blog/2012-04-02-metrics-monitoring-infrastructure-oh-my/
http://pivotallabs.com/talks/139-metrics-metrics-everywhere
http://bitmonkey.net/post/18854033582/introducing-metriks
http://code.flickr.com/blog/2008/10/27/counting-timing/
Slide 132
Slide 132 text
we're not hiring ❤