Merge branch 'master' into reorganize_code
Conflicts: monitoring/heka/telemetry_channel_60days.lua
This commit is contained in:
Mark Reid
Коммит
a08644cc93
|
|
@ -3,6 +3,7 @@
|
|||
*.bz2
|
||||
*.swp
|
||||
*.swo
|
||||
*.out
|
||||
.idea/
|
||||
data/
|
||||
histogram_cache/
|
||||
|
|
|
|||
|
|
@ -0,0 +1,28 @@
|
|||
{
|
||||
"incoming_bucket": "telemetry-incoming-v1",
|
||||
"incoming_queue": "telemetry-incoming-v1",
|
||||
"incoming_batch_size": 8,
|
||||
"publish_bucket": "telemetry-published-v1",
|
||||
"image": "ami-76831f46",
|
||||
"skip_conversion": false,
|
||||
"loop": true,
|
||||
"instance_type": "c1.xlarge",
|
||||
"ssl_key_name": "mreid",
|
||||
"ssl_key_path": "~/.ssh/aws/mreid.pem",
|
||||
"ssl_retries": 10,
|
||||
"ssl_user": "ubuntu",
|
||||
"security_groups": ["telemetry"],
|
||||
"region": "us-west-2",
|
||||
"placement": "us-west-2c",
|
||||
"name": "telemetry-process-incoming-v1-1",
|
||||
"default_tags": {
|
||||
"Owner": "mreid",
|
||||
"Application": "telemetry-server"
|
||||
},
|
||||
"ephemeral_map": {
|
||||
"/dev/xvdb": "ephemeral0",
|
||||
"/dev/xvdc": "ephemeral1",
|
||||
"/dev/xvdd": "ephemeral2",
|
||||
"/dev/xvde": "ephemeral3"
|
||||
}
|
||||
}
|
||||
|
|
@ -3,8 +3,8 @@ Telemetry MapReduce
|
|||
|
||||
We provide a basic [MapReduce][1] framework to process Telemetry Data.
|
||||
|
||||
The base code is in [job.py](job.py), and there are a few examples of job
|
||||
scripts (and filters) in the `examples/` directory.
|
||||
The base code is in [job.py](../job.py), and there are a few examples of job
|
||||
scripts (and filters) in the [examples/](../examples) directory.
|
||||
|
||||
Given the storage layout, it is easy to parallelize the processing of data
|
||||
between all available CPU cores within a machine, but also across multiple
|
||||
|
|
@ -15,8 +15,8 @@ which avoids having to parse the json string for simple count-type tasks.
|
|||
|
||||
You can specify a `filter` that leverages the [storage layout][2] to quickly and
|
||||
easily limit the data set you want to work with. It uses the same type of
|
||||
[Telemetry Schema](telemetry_schema.py) document to determine which files are
|
||||
included. There are examples of filters in the `examples/` dir as well.
|
||||
[Telemetry Schema](../telemetry_schema.py) document to determine which files are
|
||||
included. There are examples of filters in the [examples/](../examples) dir as well.
|
||||
|
||||
[1]: http://en.wikipedia.org/wiki/MapReduce "MapReduce"
|
||||
[2]: StorageLayout.md "On-disk Storage Layout"
|
||||
|
|
|
|||
|
|
@ -5,7 +5,7 @@ The basic idea is to partition the data by a number of useful dimensions, then
|
|||
use the dimensions to form a directory hierarchy. Finally, the actual
|
||||
submissions will be stored in compressed files that may be read and processed
|
||||
in parallel. The files in a directory will be split into manageable sized
|
||||
pieces. Each line in the file will be of the form <uuid><tab><json>. See
|
||||
pieces. Each line in the file will be of the form `<uuid><tab><json>`. See
|
||||
[StorageFormat](StorageFormat.md) for more details about the contents of the
|
||||
files.
|
||||
|
||||
|
|
@ -92,7 +92,7 @@ This is the approach that will be used.
|
|||
|
||||
### `telemetry_schema.json`
|
||||
|
||||
The schema is defined in [telemetry_schema.json](telemetry_schema.json) and
|
||||
The schema is defined in [telemetry_schema.json](../telemetry_schema.json) and
|
||||
contains an array of `dimensions` that are used to determine what is allowed
|
||||
at each level of the storage hierarchy. Currently supported values are:
|
||||
- String value `*`: allow any value
|
||||
|
|
@ -108,7 +108,7 @@ that the "long tail" of dimension values does not cause a huge number of small
|
|||
files to be created.
|
||||
|
||||
Code for handling a schema is found in the `TelemetrySchema` class
|
||||
in [telemetry_schema.py](telemetry_schema.py)
|
||||
in [telemetry_schema.py](../telemetry_schema.py)
|
||||
|
||||
Considered, but unused approaches
|
||||
---------------------------------
|
||||
|
|
|
|||
|
|
@ -0,0 +1,6 @@
|
|||
ip_address = "127.0.0.1:5565"
|
||||
[signer]
|
||||
name = "telemetry"
|
||||
hmac_hash = "md5"
|
||||
hmac_key = "TODO change on deploy"
|
||||
version = 0
|
||||
|
|
@ -200,10 +200,9 @@ class Job:
|
|||
conn = S3Connection(self._aws_key, self._aws_secret_key)
|
||||
bucket = conn.get_bucket(self._bucket_name)
|
||||
for r in remote_files:
|
||||
key = bucket.lookup(r)
|
||||
size = key.size
|
||||
dims = self._input_filter.get_dimensions(".", r)
|
||||
remote = {"type": "remote", "name": r, "size": size, "dimensions": dims}
|
||||
size = r.size
|
||||
dims = self._input_filter.get_dimensions(".", r.name)
|
||||
remote = {"type": "remote", "name": r.name, "size": size, "dimensions": dims}
|
||||
#print "putting", remote, "into partition", min_idx
|
||||
partitions[min_idx].append(remote)
|
||||
sums[min_idx] += size
|
||||
|
|
@ -240,10 +239,14 @@ class Job:
|
|||
conn = S3Connection(self._aws_key, self._aws_secret_key)
|
||||
bucket = conn.get_bucket(self._bucket_name)
|
||||
|
||||
count = 0
|
||||
# TODO: potential optimization - if our input filter is reasonably
|
||||
# restrictive an/or our list of keys is very long, it may be
|
||||
# a win to use the "prefix" and "delimiter" params.
|
||||
for f in bucket.list():
|
||||
count += 1
|
||||
if count % 1000 == 0:
|
||||
print "Listed", count, "so far"
|
||||
dims = self._input_filter.get_dimensions(".", f.name)
|
||||
#print f.name, "->", ",".join(dims)
|
||||
include = True
|
||||
|
|
@ -252,9 +255,9 @@ class Job:
|
|||
include = False
|
||||
break
|
||||
if include:
|
||||
out_files.append(f.name)
|
||||
out_files.append(f)
|
||||
conn.close()
|
||||
print "Done!"
|
||||
print "Done! Fetched", len(out_files), "files"
|
||||
return out_files
|
||||
|
||||
def filter_includes(self, level, value):
|
||||
|
|
|
|||
|
|
@ -2,21 +2,42 @@
|
|||
maxprocs = 4
|
||||
base_dir = "."
|
||||
max_timer_inject = 100
|
||||
max_process_duration = 500000
|
||||
|
||||
[TelemetryInput]
|
||||
[ProtobufDecoder]
|
||||
pool_size = 1
|
||||
encoding_name = "PROTOCOL_BUFFER"
|
||||
|
||||
[TCP:5565]
|
||||
type = "TcpInput"
|
||||
parser_type = "message.proto"
|
||||
decoder = "ProtobufDecoder"
|
||||
address = ":5565"
|
||||
[TCP:5565.signer.telemetry_0]
|
||||
hmac_key = "TODO change on deploy" # TODO update on deploy
|
||||
|
||||
[TelemetrySandboxManager]
|
||||
type = "SandboxManagerFilter"
|
||||
message_signer = "telemetry"
|
||||
message_matcher = "Type == 'heka.control.sandbox'"
|
||||
working_directory = "sandbox"
|
||||
max_filters = 100
|
||||
|
||||
[TelemetryServerInput]
|
||||
type = "LogfileInput"
|
||||
logfile = "/var/log/telemetry/telemetry-server.log"
|
||||
decoder = "TelemetryDecoder"
|
||||
seek_journal_name = "telemetry.journal"
|
||||
decoder = "TelemetryServerDecoder"
|
||||
seek_journal_name = "telemetry-server.journal"
|
||||
use_seek_journal = true
|
||||
|
||||
[TelemetryDecoder]
|
||||
[TelemetryServerDecoder]
|
||||
type = "PayloadJsonDecoder"
|
||||
pool_size = 1
|
||||
timestamp_layout = "2006-01-02T15:04:05.999Z"
|
||||
|
||||
#{"url":"/submit/telemetry/60dfb8c5-9966-4fd8-9b82-3a5ec2ac3d6a/saved-session/Firefox/23.0.1/release/20130814063812","duration_ms":0.547324,"code":200,"size":4819,"level":"info","message":"OK","timestamp":"2013-09-10T20:43:17.217Z"}
|
||||
|
||||
[TelemetryDecoder.json_map]
|
||||
[TelemetryServerDecoder.json_map]
|
||||
Timestamp = "$.timestamp"
|
||||
Severity = "$.level"
|
||||
url = "$.url"
|
||||
|
|
@ -25,13 +46,13 @@ code = "$.code"
|
|||
size = "$.size"
|
||||
message = "$.message"
|
||||
|
||||
[TelemetryDecoder.severity_map]
|
||||
[TelemetryServerDecoder.severity_map]
|
||||
debug = 1
|
||||
warning = 2
|
||||
info = 3
|
||||
|
||||
[TelemetryDecoder.message_fields]
|
||||
Type = "TelemetryLog"
|
||||
[TelemetryServerDecoder.message_fields]
|
||||
Type = "TelemetryServerLog"
|
||||
Payload = ""
|
||||
url = "%url%"
|
||||
duration|ms = "%duration%"
|
||||
|
|
@ -41,7 +62,7 @@ message = "%message%"
|
|||
|
||||
[TelemetryFilter]
|
||||
type = "SandboxFilter"
|
||||
message_matcher = "Type == 'TelemetryLog'"
|
||||
message_matcher = "Type == 'TelemetryServerLog'"
|
||||
ticker_interval = 60
|
||||
script_type = "lua"
|
||||
filename = "telemetry.lua"
|
||||
|
|
@ -52,7 +73,7 @@ output_limit = 64000
|
|||
|
||||
[TelemetryChannel]
|
||||
type = "SandboxFilter"
|
||||
message_matcher = "Type == 'TelemetryLog'"
|
||||
message_matcher = "Type == 'TelemetryServerLog'"
|
||||
ticker_interval = 60
|
||||
script_type = "lua"
|
||||
filename = "telemetry_channel.lua"
|
||||
|
|
@ -61,19 +82,33 @@ memory_limit = 8000000
|
|||
instruction_limit = 1000
|
||||
output_limit = 64000
|
||||
|
||||
[TelemetryChannel.config]
|
||||
rows = 1440
|
||||
sec_per_row = 60
|
||||
|
||||
[TelemetryChannel60Days]
|
||||
type = "SandboxFilter"
|
||||
message_matcher = "Type == 'TelemetryLog'"
|
||||
message_matcher = "Type == 'TelemetryServerLog'"
|
||||
ticker_interval = 60
|
||||
script_type = "lua"
|
||||
filename = "telemetry_channel_60days.lua"
|
||||
filename = "telemetry_channel.lua"
|
||||
preserve_data = true
|
||||
memory_limit = 8000000
|
||||
instruction_limit = 1000
|
||||
output_limit = 64000
|
||||
|
||||
[TelemetryChannel60Days.config]
|
||||
rows = 1440
|
||||
sec_per_row = 3600
|
||||
|
||||
[Dashboard]
|
||||
type = "DashboardOutput"
|
||||
address = "0.0.0.0:4352"
|
||||
ticker_interval = 5
|
||||
address = ":4352"
|
||||
ticker_interval = 10
|
||||
working_directory = "dashboard"
|
||||
|
||||
[AMQPOutput]
|
||||
url = "amqp://guest:guest@10.250.68.186"
|
||||
exchange = "testout"
|
||||
exchangeType = "fanout"
|
||||
message_matcher = "Type == 'heka.sandbox-output' && Fields[payload_type] == 'cbufd'"
|
||||
|
|
|
|||
|
|
@ -5,7 +5,7 @@
|
|||
local rows = 1440
|
||||
local sec_per_row = 60
|
||||
|
||||
request = circular_buffer.new(rows, 4, sec_per_row)
|
||||
request = circular_buffer.new(rows, 4, sec_per_row, true)
|
||||
local SUCCESS = request:set_header(1, "Success" , "count")
|
||||
local FAILURE = request:set_header(2, "Failure" , "count")
|
||||
local AVG_REQUEST_SIZE = request:set_header(3, "Request Size", "B", "avg")
|
||||
|
|
@ -45,6 +45,7 @@ function timer_event(ns)
|
|||
-- request:add(ns, 1, 0)
|
||||
-- sums:add(ns, 1, 0)
|
||||
|
||||
output(request)
|
||||
inject_message("cbuf", "Request Statistics")
|
||||
local title = "Request Statistics"
|
||||
inject_message(request:format("cbuf"), title)
|
||||
inject_message(request:format("cbufd"), title)
|
||||
end
|
||||
|
|
|
|||
|
|
@ -2,15 +2,23 @@
|
|||
-- License, v. 2.0. If a copy of the MPL was not distributed with this
|
||||
-- file, You can obtain one at http://mozilla.org/MPL/2.0/.
|
||||
|
||||
local rows = 1440
|
||||
local sec_per_row = 60
|
||||
|
||||
all = circular_buffer.new(rows, 2, sec_per_row)
|
||||
local REQUESTS = all:set_header(1, "Requests")
|
||||
local TOTAL_SIZE = all:set_header(2, "Total Size", "KiB")
|
||||
local rows = read_config("rows")
|
||||
local sec_per_row = read_config("sec_per_row")
|
||||
local REQUESTS = 1
|
||||
local TOTAL_SIZE = 2
|
||||
|
||||
channels = {}
|
||||
|
||||
local function add_channel(channel)
|
||||
channels[channel] = circular_buffer.new(rows, 2, sec_per_row, true)
|
||||
local c = channels[channel]
|
||||
c:set_header(REQUESTS, "Requests")
|
||||
c:set_header(TOTAL_SIZE, "Total Size", "KiB")
|
||||
return c
|
||||
end
|
||||
|
||||
all = add_channel("ALL")
|
||||
|
||||
function process_message ()
|
||||
local ts = read_message("Timestamp")
|
||||
local rs = tonumber(read_message("Fields[size]"))
|
||||
|
|
@ -33,10 +41,7 @@ function process_message ()
|
|||
|
||||
local c = channels[channel]
|
||||
if not c then
|
||||
channels[channel] = circular_buffer.new(rows, 2, sec_per_row)
|
||||
c = channels[channel]
|
||||
c:set_header(1, "Requests")
|
||||
c:set_header(2, "Total Size", "KiB")
|
||||
c = add_channel(channel)
|
||||
end
|
||||
c:add(ts, REQUESTS, 1)
|
||||
c:add(ts, TOTAL_SIZE, rs)
|
||||
|
|
@ -45,10 +50,8 @@ function process_message ()
|
|||
end
|
||||
|
||||
function timer_event(ns)
|
||||
output(all)
|
||||
inject_message("cbuf", "ALL")
|
||||
for k, v in pairs(channels) do
|
||||
output(v)
|
||||
inject_message("cbuf", k)
|
||||
inject_message(v:format("cbuf"), k)
|
||||
inject_message(v:format("cbufd"), k)
|
||||
end
|
||||
end
|
||||
|
|
|
|||
|
|
@ -1,54 +0,0 @@
|
|||
-- This Source Code Form is subject to the terms of the Mozilla Public
|
||||
-- License, v. 2.0. If a copy of the MPL was not distributed with this
|
||||
-- file, You can obtain one at http://mozilla.org/MPL/2.0/.
|
||||
|
||||
local rows = 1440
|
||||
local sec_per_row = 60 * 60
|
||||
|
||||
all = circular_buffer.new(rows, 2, sec_per_row)
|
||||
local REQUESTS = all:set_header(1, "Requests")
|
||||
local TOTAL_SIZE = all:set_header(2, "Total Size", "KiB")
|
||||
|
||||
channels = {}
|
||||
|
||||
function process_message ()
|
||||
local ts = read_message("Timestamp")
|
||||
local rs = tonumber(read_message("Fields[size]"))
|
||||
local url = read_message("Fields[url]")
|
||||
|
||||
local cnt = all:add(ts, REQUESTS, 1)
|
||||
if not cnt then return 0 end -- outside the buffer
|
||||
if rs then
|
||||
rs = rs / 1024
|
||||
else
|
||||
rs = 0
|
||||
end
|
||||
all:add(ts, TOTAL_SIZE, rs)
|
||||
|
||||
local channel = url:match("^/submit/telemetry/[^/]+/[^/]+/[^/]+/[^/]+/([^/]+)")
|
||||
if not channel then return 0 end
|
||||
if channel ~= "release" and channel ~= "beta" and channel ~= "aurora" and channel ~= "nightly" then
|
||||
channel = "other"
|
||||
end
|
||||
|
||||
local c = channels[channel]
|
||||
if not c then
|
||||
channels[channel] = circular_buffer.new(rows, 2, sec_per_row)
|
||||
c = channels[channel]
|
||||
c:set_header(1, "Requests")
|
||||
c:set_header(2, "Total Size", "KiB")
|
||||
end
|
||||
c:add(ts, REQUESTS, 1)
|
||||
c:add(ts, TOTAL_SIZE, rs)
|
||||
|
||||
return 0
|
||||
end
|
||||
|
||||
function timer_event(ns)
|
||||
output(all)
|
||||
inject_message("cbuf", "ALL")
|
||||
for k, v in pairs(channels) do
|
||||
output(v)
|
||||
inject_message("cbuf", k)
|
||||
end
|
||||
end
|
||||
|
|
@ -25,6 +25,7 @@ from telemetry.convert import Converter, BadPayloadError
|
|||
from telemetry.revision_cache import RevisionCache
|
||||
from telemetry.persist import StorageLayout
|
||||
import boto.sqs
|
||||
import traceback
|
||||
|
||||
S3FUNNEL_PATH = "/usr/local/bin/s3funnel"
|
||||
def fetch_s3_files(incoming_files, fetch_cwd, bucket, aws_key, aws_secret_key):
|
||||
|
|
@ -164,6 +165,10 @@ class ReadRawStep(PipeStep):
|
|||
print self.label, "ERROR: Found corrupted data for record", record_count, "in", raw_file, "path:", path, "Error:", err
|
||||
self.bad_records += 1
|
||||
continue
|
||||
if len(data) == 0:
|
||||
print self.label, "ERROR: Found empty data for record", record_count, "in", raw_file, "path:", path
|
||||
self.bad_records += 1
|
||||
continue
|
||||
|
||||
# Incoming timestamps are in milliseconds, so convert to POSIX first
|
||||
# (ie. seconds)
|
||||
|
|
@ -232,6 +237,7 @@ class ReadRawStep(PipeStep):
|
|||
if err_message != "Missing in payload: info.revision":
|
||||
# TODO: recognize other common failure modes and handle them gracefully.
|
||||
self.write_bad_record(key, dims, data, err_message, "Conversion Error:")
|
||||
traceback.print_exc()
|
||||
|
||||
if self.print_stats:
|
||||
this_update = datetime.now()
|
||||
|
|
@ -248,6 +254,7 @@ class ReadRawStep(PipeStep):
|
|||
except Exception, e:
|
||||
# Corrupted data, let's skip this record.
|
||||
print self.label, "- Error reading raw data from ", raw_file, e
|
||||
traceback.print_exc()
|
||||
|
||||
def write_bad_record(self, key, dims, data, error, message=None):
|
||||
self.bad_records += 1
|
||||
|
|
|
|||
|
|
@ -5,7 +5,7 @@
|
|||
"ssl_retries": 10,
|
||||
"base_dir": "/mnt/telemetry",
|
||||
"instance_type": "t1.micro",
|
||||
"image": "ami-bf1d8a8f",
|
||||
"image": "ami-ace67f9c",
|
||||
"security_groups": ["telemetry"],
|
||||
"region": "us-west-2",
|
||||
"placement": "us-west-2c",
|
||||
|
|
|
|||
|
|
@ -47,7 +47,12 @@ def create_instance(config, aws_key=None, aws_secret_key=None):
|
|||
print "Creating a new instance of type", itype
|
||||
# Known images:
|
||||
# ami-bf1d8a8f == Ubuntu 13.04
|
||||
# ami-76831f46 == telemetry-base (based on Ubuntu 13.04 with dependencies already installed)
|
||||
# ami-ace67f9c == Ubuntu 13.10
|
||||
# ami-76831f46 == telemetry-base - based on Ubuntu 13.04 with dependencies
|
||||
# already installed
|
||||
# ami-260c9516 == telemetry-server - Based on Ubuntu 13.10, everything is
|
||||
# ready to go, server will be auto-started on boot. Does
|
||||
# NOT auto-start process-incoming.
|
||||
|
||||
# See if ephemerals have been specified
|
||||
mapping = None
|
||||
|
|
|
|||
|
|
@ -46,12 +46,31 @@ class TelemetryServerLauncher(Launcher):
|
|||
def start_suid_script(self, c_file, username):
|
||||
sudo("echo 'setuid {1}' > {0}".format(c_file, username))
|
||||
sudo("echo 'setgid {1}' >> {0}".format(c_file, username))
|
||||
# Set the ulimit for # open files in the upstart scripts (since the
|
||||
# ones set in limits.conf don't seem to apply here)
|
||||
sudo("echo 'limit nofile 10000 40000' >> " + c_file)
|
||||
sudo("echo 'script' >> " + c_file)
|
||||
|
||||
def end_suid_script(self, c_file):
|
||||
sudo("echo 'end script' >> {0}".format(c_file))
|
||||
sudo("echo 'respawn' >> {0}".format(c_file))
|
||||
|
||||
def create_logrotate_config(self, lr_file, target_log, create=True):
|
||||
sudo("echo '%s {' > %s" % (target_log, lr_file))
|
||||
sudo("echo ' su {1} {1}' >> {0}".format(lr_file, self.ssl_user))
|
||||
sudo("echo ' rotate 5' >> {0}".format(lr_file))
|
||||
sudo("echo ' daily' >> {0}".format(lr_file))
|
||||
sudo("echo ' compress' >> {0}".format(lr_file))
|
||||
sudo("echo ' missingok' >> {0}".format(lr_file))
|
||||
if create:
|
||||
sudo("echo ' create 640 {1} {1}' >> {0}".format(lr_file, self.ssl_user))
|
||||
else:
|
||||
sudo("echo ' copytruncate' >> {0}".format(lr_file))
|
||||
sudo("echo '}' >> " + lr_file)
|
||||
with settings(warn_only=True):
|
||||
# This will warn if there's no file there.
|
||||
sudo("logrotate -f {0}".format(lr_file))
|
||||
|
||||
def post_install(self, instance):
|
||||
# Install some more:
|
||||
self.install_nodejs_bin()
|
||||
|
|
@ -82,19 +101,11 @@ class TelemetryServerLauncher(Launcher):
|
|||
run("echo 'Soft limit:'; ulimit -S -n")
|
||||
run("echo 'Hard limit:'; ulimit -H -n")
|
||||
|
||||
# Setup logrotate for the stats log
|
||||
lr_file = "/etc/logrotate.d/telemetry"
|
||||
sudo("echo '/var/log/telemetry/telemetry-server.log {' > " + lr_file)
|
||||
sudo("echo ' su {1} {1}' >> {0}".format(lr_file, self.ssl_user))
|
||||
sudo("echo ' rotate 10' >> {0}".format(lr_file))
|
||||
sudo("echo ' daily' >> {0}".format(lr_file))
|
||||
sudo("echo ' compress' >> {0}".format(lr_file))
|
||||
sudo("echo ' missingok' >> {0}".format(lr_file))
|
||||
sudo("echo ' create 640 {1} {1}' >> {0}".format(lr_file, self.ssl_user))
|
||||
sudo("echo '}' >> " + lr_file)
|
||||
with settings(warn_only=True):
|
||||
# This will warn if there's no file there.
|
||||
sudo("logrotate -f /etc/logrotate.d/telemetry")
|
||||
# Setup logrotate for the stats log and process-incoming log
|
||||
self.create_logrotate_config("/etc/logrotate.d/telemetry",
|
||||
"/var/log/telemetry/telemetry-server.log")
|
||||
self.create_logrotate_config("/etc/logrotate.d/telemetry-incoming",
|
||||
"/var/log/telemetry/telemetry-incoming.out", False)
|
||||
|
||||
# Create startup scripts:
|
||||
code_base = "/home/" + self.ssl_user + "/telemetry-server"
|
||||
|
|
@ -103,12 +114,25 @@ class TelemetryServerLauncher(Launcher):
|
|||
sudo("echo ' cd {1}/server' >> {0}".format(c_file, code_base))
|
||||
sudo("echo ' /usr/local/bin/node ./server.js ./server_config.json >> /var/log/telemetry/telemetry-server.out' >> {0}".format(c_file))
|
||||
self.end_suid_script(c_file)
|
||||
sudo("echo 'start on runlevel [2345]' >> {0}".format(c_file))
|
||||
sudo("echo 'stop on runlevel [016]' >> {0}".format(c_file))
|
||||
|
||||
c_file = "/etc/init/telemetry-export.conf"
|
||||
base_export_command = "/usr/bin/python -u ./export.py -d {0}/data -p '^telemetry.log.*[.]finished$' -k '{1}' -s '{2}' -r '{3}' -b '{4}' -q '{5}' --remove-file".format(base_dir, self.aws_key, self.aws_secret_key, self.config["region"], self.config.get("incoming_bucket", "telemetry-incoming"), self.config.get("incoming_queue", "telemetry-incoming"))
|
||||
|
||||
self.start_suid_script(c_file, self.ssl_user)
|
||||
sudo("echo ' cd {1}' >> {0}".format(c_file, code_base))
|
||||
sudo("echo \" /usr/bin/python -u ./export.py -d {1}/data -p '^telemetry.log.*[.]finished$' -k '{2}' -s '{3}' -r '{4}' -b '{5}' -q '{6}' --remove-files --loop >> /var/log/telemetry/telemetry-export.out\" >> {0}".format(c_file, base_dir, self.aws_key, self.aws_secret_key, self.config["region"], self.config.get("incoming_bucket", "telemetry-incoming"), self.config.get("incoming_queue", "telemetry-incoming")))
|
||||
sudo("echo \" {1} --loop >> /var/log/telemetry/telemetry-export.out\" >> {0}".format(c_file, base_export_command))
|
||||
self.end_suid_script(c_file)
|
||||
# after we receive "stop", run once more in non-looping mode to make
|
||||
# sure we exported everything.
|
||||
sudo("echo 'post-stop script' >> {0}".format(c_file))
|
||||
sudo("echo ' cd {1}' >> {0}".format(c_file, code_base))
|
||||
sudo("echo \" {1} >> /var/log/telemetry/telemetry-export.out\" >> {0}".format(c_file, base_export_command))
|
||||
sudo("echo 'end script' >> {0}".format(c_file))
|
||||
# Start/stop this in lock step with telemetry-server
|
||||
sudo("echo 'start on started telemetry-server' >> {0}".format(c_file))
|
||||
sudo("echo 'stop on stopped telemetry-server' >> {0}".format(c_file))
|
||||
|
||||
# Install a specific aws_incoming.json to use
|
||||
process_incoming_config = self.config.get("process_incoming_config", "aws_incoming.json")
|
||||
|
|
@ -120,6 +144,9 @@ class TelemetryServerLauncher(Launcher):
|
|||
# Use unbuffered output (-u) so we can see things in the log
|
||||
# immediately.
|
||||
sudo("echo \" /usr/bin/python -u process_incoming_queue.py -k '{1}' -s '{2}' ./aws_incoming.json >> /var/log/telemetry/telemetry-incoming.out\" >> {0}".format(c_file, self.aws_key, self.aws_secret_key))
|
||||
# NOTE: Don't automatically start/stop this service, since we only want
|
||||
# to start it on "primary" nodes, and we only want to stop it in
|
||||
# safe parts of the process-incoming code.
|
||||
self.end_suid_script(c_file)
|
||||
|
||||
c_file = "/etc/init/telemetry-heka.conf"
|
||||
|
|
@ -127,27 +154,25 @@ class TelemetryServerLauncher(Launcher):
|
|||
sudo("echo ' cd {1}/heka' >> {0}".format(c_file, code_base))
|
||||
sudo("echo \" /usr/bin/hekad -config heka.toml >> /var/log/telemetry/telemetry-heka.out\" >> {0}".format(c_file))
|
||||
self.end_suid_script(c_file)
|
||||
sudo("echo 'kill signal INT' >> {0}".format(c_file))
|
||||
# Start/stop this in lock step with telemetry-server
|
||||
sudo("echo 'start on started telemetry-server' >> {0}".format(c_file))
|
||||
sudo("echo 'stop on stopped telemetry-server' >> {0}".format(c_file))
|
||||
|
||||
def run(self, instance):
|
||||
# Start up HTTP server
|
||||
sudo("start telemetry-server")
|
||||
print "Telemetry server started"
|
||||
|
||||
# Start up exporter
|
||||
sudo("start telemetry-export")
|
||||
print "Telemetry export started"
|
||||
# Note: This also starts up telemetry-export and telemetry-heka due to dependencies.
|
||||
|
||||
# Start up 'process incoming' only on the primary node
|
||||
# TODO: pass in user-data to set this.
|
||||
if self.config.get("primary_server", False):
|
||||
sudo("start telemetry-incoming")
|
||||
print "Telemetry incoming started"
|
||||
else:
|
||||
print "Not starting telemetry-incoming since this is not a primary server"
|
||||
|
||||
# Start up heka
|
||||
sudo("start telemetry-heka")
|
||||
print "Heka daemon started"
|
||||
|
||||
def main():
|
||||
try:
|
||||
launcher = TelemetryServerLauncher()
|
||||
|
|
|
|||
|
|
@ -0,0 +1,10 @@
|
|||
{
|
||||
"motd": " ==== Spot Telemetry Server. Accepting Submissions since 2013. ====",
|
||||
"max_data_lenth": 204800,
|
||||
"max_path_lenth": 10240,
|
||||
"port": 8080,
|
||||
"log_path": "/mnt/telemetry/data",
|
||||
"max_log_age_ms": 120000,
|
||||
"max_log_size": 50000000,
|
||||
"stats_log_file": "/var/log/telemetry/telemetry-server.log"
|
||||
}
|
||||
|
|
@ -254,7 +254,7 @@ def main(argv=None):
|
|||
uploadables = exporter.find_uploadables(args.data_dir)
|
||||
if len(uploadables) == 0:
|
||||
print "No files to export yet. Sleeping for a while..."
|
||||
time.sleep(60)
|
||||
time.sleep(10)
|
||||
continue
|
||||
|
||||
print "Processing", len(uploadables), "uploadables:"
|
||||
|
|
|
|||
|
|
@ -61,7 +61,7 @@ def unpack(filename, raw=False, verbose=False):
|
|||
data = fin.read(len_data)
|
||||
error = None
|
||||
if not raw:
|
||||
if ord(data[0]) == 0x1f and ord(data[1]) == 0x8b:
|
||||
if len(data) > 1 and ord(data[0]) == 0x1f and ord(data[1]) == 0x8b:
|
||||
# Data is gzipped, uncompress it:
|
||||
try:
|
||||
reader = StringIO.StringIO(data)
|
||||
|
|
|
|||
Загрузка…
Ссылка в новой задаче