Skip to content

Instantly share code, notes, and snippets.

@maximilien
Last active May 19, 2017 16:51
Show Gist options
  • Save maximilien/09597f41f4c50019737c11e91c955d2d to your computer and use it in GitHub Desktop.
Save maximilien/09597f41f4c50019737c11e91c955d2d to your computer and use it in GitHub Desktop.
CF-Abacus $git log --since=05/25/2016 --pretty=oneline
More Sampling Options (#474)
Dataflow reducer takes care of reducing data (#471)
Temporarily disable bloom filter. (#452)
Fixed shifting bug in aggregator. (#456)
Accumulator should not persist first start (#435)
Merge pull request #418 from KRuelY/resourceI
resourceInstance adjusts to reporting time.
Merge pull request #416 from KRuelY/reportResourceInstance
Merge pull request #415 from KRuelY/tw
Limit the number of db connections used to list error docs
resource instance usage report update.
Timewindow - adjustWindows and dataflow - db.
Show 0 charge in report.
Merge pull request #409 from rajkiranrbala/batchthrottle
Merge pull request #408 from rajkiranrbala/duplicatedetectionid
No breaker on throttled functions
Use endtime/startime in duplicate detection documents
Merge pull request #407 from rajkiranrbala/collectorerrordb
Resource level aggregated_usage modification (#406)
Log errors in their respective months
Report's accountUsage use the correct endpoint.
Use default partitioning fn for mongoclient tests
Ignore 409 while logging input (#404)
Disable test causing dataflow problems with Mongo
Do not clean mongodb if other DB is used
Reduce the docs under CI test in mongoclient
Add cooldown logic in mongoclient tests
Enable sink dataflow tests
Merge pull request #394 from KRuelY/noDedupe
removed eureka's dependency on SECURED env var (#388)
Dataflow with no dedupe throws 409
Calculation details (#391)
Merge pull request #389 from KRuelY/timeWindowAdjustment
Timewindow's cellfn access the object.
Remove previous_quantity from the report
Include previous quantity in windows that change during aggregation
Post to output DB when sink returns 409
Merge pull request #384 from rajkiranrbala/UsageFieldsAcrossPipeline
Exclude resource_id and plan_id from org report
Update integration tests
Include resource_id and plan_id as a part of the document
Default accumulate function fix.
Support zero delay batch config with BATCH_DELAY=0
Back-propagate once per batch instead of once per sample
Allow concurrent posts of reduction output docs
Configurable retry post to dataflow sink
Configurable retries in the retry module
Configurable caching duration for the plans
Base logic for anomalous usage detection
Merge pull request #375 from KRuelY/runtimePlans
Fixed missing previous_running_instances in metering config.
Solved processed time overwrite issue #371 (#372)
Document database configuration
Introduce minimum hystrix delay
Improve memory management in usage analyzer service
Skeleton of optional analyzer service module
Duplicate Detection while extending the Pipeline #364 (#365)
Fix request tests for osx network stack
Merge pull request #363 from rajkiranrbala/CIssue
Use separate keys to address sink partitions
Tests for concurrency
Merge pull request #362 from rajkiranrbala/LocalSink
Resolves local sink url correctly
Do not cache undefined/unexisting responses
Use the correct req url behind a proxy
Add weight matrices to matrix module
Return correct flags in case of conflict in couchclient
Fix conflict test for couchclient
Merge pull request #357 from rajkiranrbala/FixReporting
Read accumulated usage from accumulator db
Remove duplicate copy of resource instance usage
Report only events that are older than a limit
Merge pull request #354 from rajkiranrbala/ResourceInstanceReference
Move resource instance references to consumer aggregated docs
Add space id back to consumer usage aggregation doc ids
Don't include space id in consumer usage aggregation doc ids
Return app version in an HTTP response header
Fix perf test start sequence and reenable it
Add a _rev field to new inserted docs as well
Delete rev field instead of setting it undefined
Remove id field from stored docs
Optional downsampling of accumulated and aggregated usage history
Support downsampling of reduction history
Add a function to sample down seq ids
Shrunk the consumer ids for saving space. (#346)
Propagate error messages from error fields
Modify collector and dataflow.
Support false values in .apprc configs
Add provider, resource, plan ids to collected usage keys
Add a utility to help rewire tested modules
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment