Publishing details

Changelog

salt-formula-heka (0.2+0~20170518093610.90~1.gbpa50a2c) trusty; urgency=medium

  ** SNAPSHOT build @a50a2c321d43f5144f1e9db6b8963841d6ae45ae **

  [ Ales Komarek ]
  * Round 2
  * Added cache dir
  * Regex encoder fix
  * heka server role, backward compat

  [ Éric Lemoine ]
  * Variable service_name is missing for systemd file
  * preserve_data and ticker_interval are not strings
  * Use "ignore missing" with the j2 include statement
  * Use module_dir instead of module_directory
  * Use the proper module directory
  * Add the extra_fields.lua module
  * Fix the decoder configuration
  * Make Heka send metrics to InfluxDB
  * Make InfluxDB time precision configurable
  * Configure Elasticsearch output through Pillar
  * Use influxdb_time_precision for InfluxDB output
  * Add alarming support
  * Revert "[WIP] Add alarming support"
  * Remove the aggregator output for now
  * Do not place Heka logs in /var/log/upstart
  * Add alarming support
  * Make the aggregator load heka/meta/heka.yml
  * Use filter_by to merge node grains data
  * Make the output/tcp.toml template extendable
  * Add an aggregator.toml output template
  * Add a run_lua_tests.sh script
  * Copy Lua tests from fuel-plugin-lma-collector
  * Fix the afd tests
  * Fix the gse tests
  * Add aggregator config to support metadata
  * Fix state dependency
  * Set influxdb_time_precision in aggregator class
  * Disable the heka service completely
  * Define the highest_severity policy
  * Define the majority_of_members policy
  * Define the availability_of_members policy
  * Generate the gse_policies Lua module
  * Generate gse topology module for each alarm cluster
  * Generate gse filter toml for each cluster alarm
  * Adapt GSE Lua code
  * Remove gse cluster_field parameter
  * Support dimensions in gse metrics
  * Do not rely on pacemaker_local_resource_active
  * Configure outputs in support metadata
  * Fix bug in map.jinja
  * Make remote collector send only afd metrics to influx
  * Add aggregator output to remote collector
  * Fix the gse_policies structure
  * Send only alarm-related data to mine
  * Fix the grains_for_mine function
  * Fix flake8 in heka_alarming.py
  * Make Heka service watch Lua dir
  * Add missing hostname variable in GSE code
  * Simplify message matchers
  * Return an err message when set_member_status fails
  * Fix issue in lma_alarm.lua template
  * Set actual swap_size in collectd decoder
  * Send annotations to InfluxDB
  * Revert "Fix unclosed if statement (#73)"
  * Revert "Stacklight (#71)"
  * Support cinder|neutron|nova_services_percent metrics
  * Take rule value field into account
  * Do not create log file if systemd is used
  * Configure logrotate
  * Add specific logrotate script and crontab
  * Update README.rst
  * Add missing default values
  * Make metric_collector output configurable
  * Revert "Add specific logrotate script and crontab"
  * Change logrotate configuration
  * Use the amqp_ prefix for AMQP variables
  * Remove the majority_of_members policy
  * Revert "For sake of consistency with availability_of_members"
  * Use full action "drop" for logs

  [ Simon Pasquier ]
  * Add HTTP metrics filter to log_collector
  * Add logs counter filter to log_collector
  * Templatize extra_fields.lua file
  * Remove http check input plugin
  * Add generic timezone support to decoders
  * Fix the definition of the remote_collector service
  * Add monitoring of the Heka processes
  * Extend collectd decoder to support vrrp metrics
  * Update collectd decoder to parse ntpd metrics
  * Enable collection of notifications
  * Add a log decoder for Galera
  * Update collectd decoder for nginx metrics
  * Add a log decoder for GlusterFS
  * Fix collectd Lua decoder for system metrics
  * Update collectd decoder for disk metrics
  * Support remote_collector and aggregator in cluster
  * Update collectd decoder for Elasticsearch metrics
  * Add filter to compute resource creation timings
  * Support Oslo messaging v2 notifications
  * Fix message matcher for AFD filters
  * Remove trailing spaces
  * Send log_messages metric as bulk
  * Restore 'block' policy for Elasticsearch output
  * Move convert_to_sec() to lma_utils module
  * Improve inject_bulk_metric()
  * Move hdd_errors filter to the Heka formula
  * Add logger_matcher option to the logs_counter filter
  * Add support for UdpOutput plugin
  * Cast 'http_status' field into an integer
  * Improve the Keystone WSGI log decoder
  * Extend notification decoder to support CADF
  * Send CADF audit events to Elasticsearch
  * Add Lua filter to get metrics from auth messages
  * Fix decoders and filters for hostname-free metrics
  * Remove unused Lua file
  * Rename authentications.lua to audit_authentications.lua
  * Add a filter counting the failed SSH logins
  * Fix filters injecting bulk metrics
  * Deserialize bulk metrics based on the source field
  * Add 'region' field to the InfluxDB tags if present
  * Add support for Syslog output
  * Enhance the collectd decoder for generic metrics
  * Simplify collectd.lua decoder plugin
  * Add Nagios class to the metric_collector metadata
  * Fix InfluxDB line protocol encoder
  * Enforce ownership and permissions
  * Fix collectd decoder for Xenial compute nodes
  * Fix metric name for disk's pending operations
  * Add the ability to disable alarms from the model

  [ Swann Croiset ]
  * Increase maximum open files limit to 102400
  * Cleaning useless default map keys
  * Redefine alerting property
  * Add Nagios output for metric_collector
  * Add Nagios output for the aggregator
  * Configure Hekad poolsize by pillar data
  * Fix the heka grains for the aggregator/remote_collector
  * Use a dimension key for the Nagios host displaying alarm clusters
  * Add Apache access log decoder for StackLight
  * Restart Heka services when Lua lib have changed
  * Fix http_response_time for Apache access decoder
  * Make INFO the default syslog severity level
  * Remove heka self-monitoring
  * Suffix Nagios host with the environment_label dimension
  * Support legacy heka.server without parse error
  * Call once string.format to build host_dimension field
  * Decode process metrics
  * Remove legacy http_check metric decoder
  * Add default sandbox decoder paths
  * Add kubernetes log decoder
  * Deserialize all bulk metrics from log_collector
  * Restart Heka services when removing config files
  * Remove the hostname from AFD when related to cluster
  * Correcly parse Python traceback

  [ swann croiset ]
  * Update map.jinja

  [ Olivier Bourdon ]
  * Add contrail log parsers
  * Add redis log parser
  * Add zookeeper log parser
  * Add cassandra log parser
  * Add ifmap log parser
  * Add metrics for Nova aggregates
  * Fix fatal error in metric_collector execution
  * Add Calico services log files parsers

  [ Ilya Tyaptin ]
  * Add an os_telemetry_collector service
  * Fix issue with endif in heka.yml
  * Fix behavior of heka_alarming.py for None service_data
  * Avoid definition of empty stacklight heka config groups
  * Avoid definition of empty ceilometer_collector config groups
  * Fix issue with missed endif
  * Do not cancel resource injecting after samples failure
  * Validate timestamp and nil fields in resources.lua

  [ Aleš Komárek ]
  * Stacklight (#71)
  * Doc rendering and monitoring generation orchestrate (#77)

  [ Dmirty Kalashnik ]
  * Fix unclosed if statement (#73)
  * Fix unclosed if statement (#73)

  [ Patrick Petit ]
  * Added a policy for members of node clusters
  * Fixed policies
  * For sake of consistency with availability_of_members
  * Final rework of status_of_members policy
  * Adding support for sensu as alerting destination The AFD and GSE metrics are sent by the aggregator to local sensu-client. A sensu-client client should be running on all nodes though (not only the aggregator) for the sensu integrtion to work correctly.

  [ Filip Pytloun ]
  * Unify Makefile, .gitignore and update readme
  * Manage grains using support metadata
  * Try to fix jinja cycle
  * Fix mapping values are not allowed here
  * Fix grains generation when psutil is not installed

  [ Guillaume Thouvenin ]
  * Add local check for Heka processes
  * Improve heka check
  * Add watchdog for Heka services
  * Add check key only if there is at least one check
  * Rename no_alerting field
  * Manage notification handler for Sensu
  * Add Sensu handler if there is a field notification_handler
  * Configure option for sensu_output

  [ teoyaomiqui ]
  * PROD-10359 fixing multiline log parsing for syslog

  [ Martin Polreich ]
  * Fix typo in heka.yml

  [ Tomáš Kukrál ]
  * add missing macro load_grains_file

  [ Petr Michalec ]
  * fix syntax on meta/heka.yml

  [ Autobuilder ]

 -- Autobuilder <email address hidden>  Thu, 18 May 2017 09:36:27 +0000

Available diffs

Builds

Built packages

Package files