1.0.1 (2019-03-20)

• adding unit tests for cloudwatch facade
• Merge pull request #4 from juanrh/improve-coverage-cloudwatch_logger Improve coverage cloudwatch logger
• Make LogManagerFactory mockeable
• Make cloudwatch_logs_common shared lib to use it in other libs
• Merge pull request #1 from xabxx/master [Bug Fix] Resolved false-positive error log messages
• Resolved false-positive error log messages
• Contributors: Abby Xu, Ross Desmond, Ryan Newell, Yuan "Forrest" Yu, hortala

1.1.5 (2020-10-14)

• Limit the rate of cloudwatch log uploading to match API service limits (#61) * Limit the rate of publishing to PutLogEvents by delaying calls inside the CloudWathLogsFacade
• Contributors: Emerson Knapp

1.1.4 (2020-09-17)

• Correct seconds to milliseconds (#59)
• Add AWS logging to batch tests (#58)
• Allow user to automatically delete CloudWatch logs in batch that are > 14 days old (#56) * move docstring and don't break loop * call fileUploadComplete * modify struct and add tests * add tests and change 2 week wording to old * Add lock to discard stale data container * extend stale data protection * remove setter * Reduce lock_guard scope
• add error for invalid log batch (#52) * add error for invalid log batch * modify logic * change to long instead of time * fix syntax * separate to function * change function order * add unit test * Add file manager test to CMakeList * file manager unit test * typo * add test log file manager class * modify filetest class and add unit test * add test publisher * debug * add validate batch function: * organize tuple * for loop to add valid data * pair * new priority queue * back to tuple * add outdated check * modify logic * modify test and pq * modify tests * add readbatch mock * add headers * remove data token * separate to new log_batch test file * add log comparison function * update test * fix typos * update batch test * remove outdated, fix timestamp * add > 24 hour log * remove utility * fix typo * changes based on review * modify macro * modify test * modify log_data * typo * add testfilemanagerstrategy * change logfile to datamanager & more * mock read function * update read and write functions * modify tests based on review * add batch_data validation * remove unnecessary headers * add test fixture * update test fixtures * updates based on review
• Bumping version to match bloom release (#51) Bumping version to 1.1.3
• Fix linting issues found by clang-tidy 6.0 (#50) * clang-tidy fixes * revert explicit constructor declarations to maintain API compatbility * clang-tidy linting issues fixed manually * fix unit tests build break
• Increase package version numbers to 1.1.2 (#44)
• Contributors: Jesse Ikawa, Miaofei Mei, Nick Burek, Ragha Prasad

1.1.1 (2019-09-10)

• Disable error on cast-align warning to support ARMhf builds (#41) * Remove -Wcast-align flag to support ARMhf builds * - bumped versions to 1.1.1 - restored cast-align, but added as a warning - removed unused headers * Removed duplicate Werror
• modify changelist to be compatible with catkin_generate_changelog (#39) Signed-off-by: Miaofei <miaofei@amazon.com>
• Merge pull request #36 from aws-robotics/guard_test_libs Added guards for test libraries
• Added guards for test libraries
• Offline logs feature (#34) * Add file manager and upload complete callback Signed-off-by: Miaofei <miaofei@amazon.com> * Fixed merge conflict Signed-off-by: Miaofei <miaofei@amazon.com> * Add file manager test Still deciding if the write should just go in the FileManagerStrategy. Signed-off-by: Miaofei <miaofei@amazon.com> * Add FileManager to log manager factory Signed-off-by: Miaofei <miaofei@amazon.com> * Add generic file upload logic Signed-off-by: Miaofei <miaofei@amazon.com> * Add reading, discovery and rotation to FileManagerStrategy - Add function getFileToRead that returns the best file to start reading from when sending data from the file to CloudWatch. - Add function discoverStoredFiles which finds all files in the storage directory and keeps them, in case there were offline logs still hanging around from a previous run. - Add rotateActiveFile which creates a new file to write to. This is used when we want to read from the current active file, or when the file size limit has been reached. Signed-off-by: Miaofei <miaofei@amazon.com> * Add file management system. Signed-off-by: Miaofei <miaofei@amazon.com> * Link file management with publisher Signed-off-by: Miaofei <miaofei@amazon.com> * Add test for file management system Signed-off-by: Miaofei <miaofei@amazon.com> * Add reading and writing to FileManagerStrategy - This way the file manager strategy can handle all the core file data stuff, keeping track of the size of the active file and rotating it when neccessary. Signed-off-by: Miaofei <miaofei@amazon.com> * cleanup Signed-off-by: Miaofei <miaofei@amazon.com> * Don't initialize file_manager_strategy yet Signed-off-by: Miaofei <miaofei@amazon.com> * Add file upload statistics Signed-off-by: Miaofei <miaofei@amazon.com> * Move file management files to isolated package Signed-off-by: Miaofei <miaofei@amazon.com> * Moved tests to file_manager folder Signed-off-by: Miaofei <miaofei@amazon.com> * Update offline test location Signed-off-by: Miaofei <miaofei@amazon.com> * Add comments to file management interfaces Signed-off-by: Miaofei <miaofei@amazon.com> * Add more comments to file uploading Signed-off-by: Miaofei <miaofei@amazon.com> * Add futures/promises to file upload tasks Signed-off-by: Miaofei <miaofei@amazon.com> * Add blocking and observed queue interface Signed-off-by: Miaofei <miaofei@amazon.com> * Add comments to observed_queue.h Signed-off-by: Miaofei <miaofei@amazon.com> * Make batch size configurable for file upload manager Signed-off-by: Miaofei <miaofei@amazon.com> * Add subscribe method to file upload manager Signed-off-by: Miaofei <miaofei@amazon.com> * Move queues and status monitor to dataflow folder Signed-off-by: Miaofei <miaofei@amazon.com> * Atomatize status monitor Signed-off-by: Miaofei <miaofei@amazon.com> * Add working dataflow pipeline Signed-off-by: Miaofei <miaofei@amazon.com> * Add template argument for file management factory Signed-off-by: Miaofei <miaofei@amazon.com> * Add priority between queues functionality Signed-off-by: Miaofei <miaofei@amazon.com> * Fix test and rename namespace for dataflow Signed-off-by: Miaofei <miaofei@amazon.com> * Cleanup test Signed-off-by: Miaofei <miaofei@amazon.com> * Organize priority options Signed-off-by: Miaofei <miaofei@amazon.com> * Rename FileUploadManager to FileUploadStreamer Signed-off-by: Miaofei <miaofei@amazon.com> * Add comments Signed-off-by: Miaofei <miaofei@amazon.com> * Convert LogManager to sink and ILogManager Signed-off-by: Miaofei <miaofei@amazon.com> * Amend UploadFunction status type Signed-off-by: Miaofei <miaofei@amazon.com> * Add milliseconds to dequeue Signed-off-by: Miaofei <miaofei@amazon.com> * Add timeout to dequeue() Signed-off-by: Miaofei <miaofei@amazon.com> * Add operator >> overload for source to input stage Signed-off-by: Miaofei <miaofei@amazon.com> * Adding tests for file manager strategy Signed-off-by: Miaofei <miaofei@amazon.com> * Use Tokens with FileManagerStrategy - FileManagerStrategy now only has basic read and write functions. When reading data you're given a token. When done sending this to CloudWatch you can call resolve() on the token which marks that section of the file as complete. After all sections of a file have uploaded the file is deleted. - FileManagerStrategy also now inherits from DataManagerStrategy so that in the future people can build new systems for managing their offline logs and easily hook them into the existing framework. Signed-off-by: Miaofei <miaofei@amazon.com> * Fix operator overload for InputStage Signed-off-by: Miaofei <miaofei@amazon.com> * Remove unnecessary external class declaration Signed-off-by: Miaofei <miaofei@amazon.com> * Add tests for file upload streamer Signed-off-by: Miaofei <miaofei@amazon.com> * Reformat file management Signed-off-by: Miaofei <miaofei@amazon.com> * Reformat log file names and discovery Signed-off-by: Miaofei <miaofei@amazon.com> * Cleanup, add Tests - Rename file_name to file_path to better describe what it represents - Add more tests for FileManagerStrategy Signed-off-by: Miaofei <miaofei@amazon.com> * Delete old files when storage limit is reached - Add a storage_limit option. When this is reached the oldest file on disk will be deleted to clear space. - The storage limit is checked before data is written so that the size of files on disk can never be over the storage limit. Signed-off-by: Miaofei <miaofei@amazon.com> * Task factory (#6) * Don't use atomic memory operations * generic observer definition * init connection monitor * Added publisher interface Refactored log publisher to use interface * Added IPublisher Interface definition and as Task member Added Publisher implementation and templated for data type * Added LogBatcher (renamed LogManager) Added DataStreamer interface * Rename DataStreamer interface to DataBatcher to reduce confusion * Added TaskFactory Added Publisher Interface Added factory to LogBatcher * BasicTask uses a shared pointer for templated data * Added publishing when the DataBatcher reaches a specific size * Use generic FileManager in the TaskFactory Add uploadCompleteStatus as an abstract FileManager method * Added Log Service - worker thread to dequeue tasks - common owner of various actors (file streaming, log batching) - templated to be a generic interface * Added Task cancel method * minor change to push * Fix file permissions * Added dequeue with timeout for LogService * Finished publisher base class implementation Added new states for LogPublisher * Removed shared object from logs * Moved AWS SDK init and shutdown into publisher * uncomment file streamer code - does not compile! Signed-off-by: Miaofei <miaofei@amazon.com> * Fixed build, still todo fix file streamer Signed-off-by: Miaofei <miaofei@amazon.com> * Simple pipeline tests (#9) * initial commit for simple test * Added test for batched size * minor comments * Added ObservableObject class Added simple ObservableObject tests Integrated ObservableObject into base publisher class File Streamer uses ObservableObject registration on publisher owned state * Added basic service interface for generic init, start, and shutdown Signed-off-by: Miaofei <miaofei@amazon.com> * Remove task factory * Task change proposal * Remove the task factory * Working file management tests * Fix tests and minor logic due to merge Signed-off-by: Miaofei <miaofei@amazon.com> * Address comments for task change proposal Signed-off-by: Miaofei <miaofei@amazon.com> * Delete task_factory.h Signed-off-by: Miaofei <miaofei@amazon.com> * Add cloudwatch options * Use explict set/add sink and source functions Signed-off-by: Miaofei <miaofei@amazon.com> * Cleanups (#12) * Added cancel flag * Added thread handling (same as log service) to file upload streamer * Added RunnableService * Added ObservableObject tests Added documentation * Added RunnableService test Added documentation Converted streamer and log service to runnables * Merge Conflict Fixes Added sanity (empty) tests * Fixed pipeline tests * Addressed review comments * fix for merge conflict Signed-off-by: Miaofei <miaofei@amazon.com> * Thorough testing of token system Signed-off-by: Miaofei <miaofei@amazon.com> * Clear file streamer queue on failure to upload * Add locks around dequeue Signed-off-by: Miaofei <miaofei@amazon.com> * Add basic mutex synchronization for ObservedQueue Signed-off-by: Miaofei <miaofei@amazon.com> * Remove uploadStatusComplete from FileManager Remove the uploadStatusComplete function from FileManager as it is not the responsibility of the file manager to determine if data should be written. Instead, a lambda should be used to first check for upload failure then write to the file manager. Signed-off-by: Miaofei <miaofei@amazon.com> * Add construct from backup for TokenStore Signed-off-by: Miaofei <miaofei@amazon.com> * Fix synchronized queue and address comments Signed-off-by: Miaofei <miaofei@amazon.com> * Enable build flags (#16) * Added build flags per team process * Addressed some build fixes found by flags * Fix build issues with new build flags Signed-off-by: Miaofei <miaofei@amazon.com> * Fix publishing (#15) * Removed initialize method (not needed) for service Fixed publishing Reinit AWS SDK each time we configure (needed if gone offline) * Addressed some ToDos Added publisher diagnostics Minor cleanups Added documentation * Fix issue with constant * Propgated no network connection state in publisher * fix pipeline test teardown * Addressed review comments * merge fixes * Added input checking for CloudWatchService Signed-off-by: Miaofei <miaofei@amazon.com> * Don't clear sink on successful upload - Add test and fix bug so that the file upload sink is only cleared when an upload fails. cr https://code.amazon.com/reviews/CR-9559033 Signed-off-by: Miaofei <miaofei@amazon.com> * ROS-2000: [Test] Full pipeline when there is no internet - added input checking for various constructors cr https://code.amazon.com/reviews/CR-9552279 Signed-off-by: Miaofei <miaofei@amazon.com> * ROS-2136: Address migrating core classes to service interface - Define Defaults for File Strategy - Deleted files are deleted on a new thread - Removed code from destructors that may fail - CloudWatchService handles start / shutdown of all services Signed-off-by: Miaofei <miaofei@amazon.com> * ROS-2001: [Test] Full pipeline when there is intermittent internet ROS-2002: [Test] Case when batched data is queued at an untenable rate Signed-off-by: Miaofei <miaofei@amazon.com> * Addressed review comments Signed-off-by: Miaofei <miaofei@amazon.com> * Move dataflow to separate library cr https://code.amazon.com/reviews/CR-9586163 Signed-off-by: Miaofei <miaofei@amazon.com> * Move file management to separate package directory * Modified onPublishStatusChanged in file streamer to remove dependency on cloudwatch cr https://code.amazon.com/reviews/CR-9596692 Signed-off-by: Miaofei <miaofei@amazon.com> * ROS-2147: Move DataBatcher to utils cr https://code.amazon.com/reviews/CR-9640987 Signed-off-by: Miaofei <miaofei@amazon.com> * - addressed review comments - added documentation - moved waiter test utility to separate implementation Signed-off-by: Miaofei <miaofei@amazon.com> * ROS-2166: I can check the state of the CloudWatch publishing service cr https://code.amazon.com/reviews/CR-9763677 Signed-off-by: Miaofei <miaofei@amazon.com> * Add Metric File Manager to Cloudwatch Metrics Common cr https://code.amazon.com/reviews/CR-9607921 Signed-off-by: Miaofei <miaofei@amazon.com> * Improve metric serialization, add tests. Signed-off-by: Miaofei <miaofei@amazon.com> * Add Serialization of StatisticValues Signed-off-by: Miaofei <miaofei@amazon.com> * Add serializing of Dimensions, Value and Values Signed-off-by: Miaofei <miaofei@amazon.com> * Doc and coding style improvements Signed-off-by: Miaofei <miaofei@amazon.com> * Squashed commit of the following: cr https://code.amazon.com/reviews/CR-9769267 commit 41bc857bd30853f80a439bfec0ba389fd4253dc0 Author: Devin Bonnie <dbbonnie@amazon.com> Date: Fri Jun 21 13:52:29 2019 -0700 Various fixes from rebasing commit 0e6149b0733323d80390567c11d65e013318d3f2 Author: Devin Bonnie <dbbonnie@amazon.com> Date: Thu Jun 20 16:39:58 2019 -0700 - addressed review comments - added metrics definition file - removed configure from publisher interface commit 206880df0198d6fba4299f0ebd25fbc23831bc8b Author: Devin Bonnie <dbbonnie@amazon.com> Date: Mon Jun 17 11:43:57 2019 -0700 ROS-2055: Implement DataBatcher for Metrics ROS-2056: Implement MetricService cr https://code.amazon.com/reviews/CR-9769267 commit c2ad314521b17a34c7f481d5ea5c5ca008918ac2 Author: Devin Bonnie <dbbonnie@amazon.com> Date: Fri Jun 14 23:55:23 2019 -0700 ROS-2057: Create immutable metric container commit b2df9419963a67b60b87df9e5aee34d55111d92c Author: Devin Bonnie <dbbonnie@amazon.com> Date: Fri Jun 14 16:50:48 2019 -0700 Moved CloudwatchService to utils commit e07e35e04ed3a4c2d2803a3daf3261dc8f3c2e4b Author: Devin Bonnie <dbbonnie@amazon.com> Date: Fri Jun 14 11:08:40 2019 -0700 ROS-2055: Implement Metric Publisher - moved Publisher to utilities - moved CloudWatchService to utilities - cleaned up headers - fixed namespace issues Signed-off-by: Miaofei <miaofei@amazon.com> * ROS-2226: [Bug] Metrics Facade Class does not properly set network disconnected state cr https://code.amazon.com/reviews/CR-10089409 Signed-off-by: Miaofei <miaofei@amazon.com> * Backup TokenStore to disk - Add TokenStoreOptions so the user can configure the directory the token store is backed up to. - On shutdown save the token store and all active tokens out to disk in JSON format. - On startup load the tokenstore from the file saved on disk. - Tests for shutdown/startup cr https://code.amazon.com/reviews/CR-9736297 Signed-off-by: Miaofei <miaofei@amazon.com> * Improve serialize function, catch invalid JSON - Add a new serialize function instead of overloading << in TokenStore - Catch and continue if we have trouble parsing the TokenStore backup file. Signed-off-by: Miaofei <miaofei@amazon.com> * Improve naming and initialization of variables Signed-off-by: Miaofei <miaofei@amazon.com> * Add better random number generator Signed-off-by: Miaofei <miaofei@amazon.com> * Code style fixes Signed-off-by: Miaofei <miaofei@amazon.com> * ROS-2051: Add FileManagement Pipeline to CW Metrics cr https://code.amazon.com/reviews/CR-10100452 Signed-off-by: Miaofei <miaofei@amazon.com> * Moving options around - Moving TokenStore and FileManagerStrategy options to a separate file so that it can be included and set by the upstream packages. - Renaming the Dataflow options to UploaderOptions - Creating one main CloudwatchOptions in both logs and metrics that has FileManagerOptions and UploaderOptions inside it. Signed-off-by: Miaofei <miaofei@amazon.com> * Change storage limits to kb instead of bytes cr https://code.amazon.com/reviews/CR-10144739 Signed-off-by: Miaofei <miaofei@amazon.com> * File upload streamer integration and unit tested Summary File upload and token cache manages failed and in flight tokens. Files are uploaded when the streamer is notified of an available file and network access. Files that are on the system are after FileStreamer shutdown are uploaded on restart. * Tested with cloudwatch logs * Tested with unit tests cr https://code.amazon.com/reviews/CR-10173529 Signed-off-by: Miaofei <miaofei@amazon.com> * Capitalize W in kDefaultCloudWatchOptions Signed-off-by: Miaofei <miaofei@amazon.com> * Pass options correctly, fixing bugs - Pass options to the FileManager for logs and metrics - Add additional params to handle this option passing. Signed-off-by: Miaofei <miaofei@amazon.com> * Add different file storage options for metrics by default - Metrics files now go in a metrics directory with metric prefix by default, so that they don't get mixed up with offline logs. Signed-off-by: Miaofei <miaofei@amazon.com> * DRY'ify, remove magic numbers, fix tests - Consolidate duplicate path processing code into one area. - Fix magic numbers, move into defines. - Fix tests. Signed-off-by: Miaofei <miaofei@amazon.com> * ROS-2249: [Bug] Log Publisher implementation does not properly handle token init ROS-2250: Restore CloudWatch Logs Facade Unit Test cr https://code.amazon.com/reviews/CR-10253526 Signed-off-by: Miaofei <miaofei@amazon.com> * Added relevant unit tests Minor fixes and cleanup Signed-off-by: Miaofei <miaofei@amazon.com> * - CloudWatchClients are now shared pointers instead of unique - addressed spacing issues - updated CloudWatchLogs facade naming to be consistent with Metrics Signed-off-by: Miaofei <miaofei@amazon.com> * Include <random> in header file cr https://code.amazon.com/reviews/CR-10531916 Signed-off-by: Miaofei <miaofei@amazon.com> * Rename variables and error to match config - Rename the batch size variables to match the config file names. - Update error message so the end user knows what config options are wrong. cr https://code.amazon.com/reviews/CR-10481115 Signed-off-by: Miaofei <miaofei@amazon.com> * Allow batch_trigger_publish_size and batch_max_queue_size to be the same Signed-off-by: Miaofei <miaofei@amazon.com> * Changing back ot publish size must be less than max queue size Signed-off-by: Miaofei <miaofei@amazon.com> * Check batch trigger publish size against kDefaultTriggerSize Signed-off-by: Miaofei <miaofei@amazon.com> * ROS-2231: [Bug] Potential locking issue with DataBatcher child classes - batcher attempt to flush batched data when shutting down - added documentation cr https://code.amazon.com/reviews/CR-10543019 Signed-off-by: Miaofei <miaofei@amazon.com> * Addressed review comments Signed-off-by: Miaofei <miaofei@amazon.com> * Fix up param values - Remove stream_max_queue_size as it's no longer used. - Remove kDefaultUploaderOptions because it's not used as it's always replaced by the default values specified in uploader_options struct. - Pass batch_max_queue_size and batch_trigger_publish_size to the DataBatcher's so they're actually used cr https://code.amazon.com/reviews/CR-10571067 Signed-off-by: Miaofei <miaofei@amazon.com> * ROS-2338: I can configure the amount of streamed data to hold in memory cr https://code.amazon.com/reviews/CR-10578133 Signed-off-by: Miaofei <miaofei@amazon.com> * ROS-2240: Restore existing unit tests - added definitions header to logs cr https://code.amazon.com/reviews/CR-10569452 Signed-off-by: Miaofei <miaofei@amazon.com> * Removed extra definitions file Signed-off-by: Miaofei <miaofei@amazon.com> * ROS-2341: Publisher state refactor cr https://code.amazon.com/reviews/CR-10584550 Signed-off-by: Miaofei <miaofei@amazon.com> * Fixes bug with trying to upload to cloudwatch in batches that aren't chronologically sorted. https://sim.amazon.com/issues/7cbe72f2-28c6-4771-a202-ab0d72587031 cr https://code.amazon.com/reviews/CR-10621402 Signed-off-by: Miaofei <miaofei@amazon.com> * ROS-2346: [Bug] Don't set stats values in metric datums cr https://code.amazon.com/reviews/CR-10623123 Signed-off-by: Miaofei <miaofei@amazon.com> * - doc additions Signed-off-by: Miaofei <miaofei@amazon.com> * - removed other unsupported types via review Signed-off-by: Miaofei <miaofei@amazon.com> * ROS-2263: [Bug] Storage and retry behavior for failed requests Signed-off-by: Miaofei <miaofei@amazon.com> * Addressed review comments Signed-off-by: Miaofei <miaofei@amazon.com> * Added invalid data handling to metrics Signed-off-by: Miaofei <miaofei@amazon.com> * ROS-2368: [Bug] Data is not attempted to be uploaded without an active input cr https://code.amazon.com/reviews/CR-10664962 Signed-off-by: Miaofei <miaofei@amazon.com> * ROS-2369: [Bug] Fix Metrics Serialization Unit Tests cr https://code.amazon.com/reviews/CR-10665643 Signed-off-by: Miaofei <miaofei@amazon.com> * Revert "ROS-2368: [Bug] Data is not attempted to be uploaded without an active input" This reverts commit 67129f977446079a28539833c0d3d7967306f0c2. Signed-off-by: Miaofei <miaofei@amazon.com> * ROS-2368: [Bug] Data is not attempted to be uploaded without an active input cr https://code.amazon.com/reviews/CR-10664962 Signed-off-by: Miaofei <miaofei@amazon.com> * ROS-2380: [Bug] CloudWatch Service Shutdown cr https://code.amazon.com/reviews/CR-10804863 Signed-off-by: Miaofei <miaofei@amazon.com> * Fix bug - logs not being uploaded from disk after reconnecting - If all files on disk were added to the queue the status was set to UNAVAILABLE. Then if they failed to upload the status was never restored. This ensures that if a file fails to upload the status is set back to AVAILABLE so they can attempt to be uploaded again. - Add more DEBUG logs to file management. cr https://code.amazon.com/reviews/CR-10806493 Signed-off-by: Miaofei <miaofei@amazon.com> * Read the newest file in storage instead of the oldest, lock when deleting file - Read the newest file from storage instead of reading the oldest. - When deleting a file to free up storage space, add a lock to ensure we're not reading from that same file. If we are then stop reading from that file. cr https://code.amazon.com/reviews/CR-10886255 Signed-off-by: Miaofei <miaofei@amazon.com> * Add lock to active write file - When checking if the active file should be rotated first lock it to ensure it's not being written to as it's rotated. - Add new log to delete oldest file. Signed-off-by: Miaofei <miaofei@amazon.com> * Add docs for FileManagerStrategy, cleanup unused code - Add documentation to all FileManagerStrategy functions - Remove some un-useful code for the FileManagerStrategy - Function renaming / cleanup to make more sense. Signed-off-by: Miaofei <miaofei@amazon.com> * Remove todo and unused variable Signed-off-by: Miaofei <miaofei@amazon.com> * Remove unneccessary initialization and commented out code Signed-off-by: Miaofei <miaofei@amazon.com> * ROS-2381: [Bug] Items in memory lost on shutdown cr https://code.amazon.com/reviews/CR-10942302 Signed-off-by: Miaofei <miaofei@amazon.com> * ROS-2421: [Bug] Ensure FileManager thrown exceptions are handled cr https://code.amazon.com/reviews/CR-11029944 Signed-off-by: Miaofei <miaofei@amazon.com> * - addressed review comments - changed file upload streamer wait timeout from 1 minute to 5 minutes Signed-off-by: Miaofei <miaofei@amazon.com> * Addressed terse variable names Signed-off-by: Miaofei <miaofei@amazon.com> * increment minor version Signed-off-by: Miaofei <miaofei@amazon.com> * fix compilation errors in unit tests Signed-off-by: Miaofei <miaofei@amazon.com> * fix more compilation errors found in dashing Signed-off-by: Miaofei <miaofei@amazon.com> * fix unit test failures Signed-off-by: Miaofei <miaofei@amazon.com>
• Modify cloudwatch common to depend on gtest, gmock (#19) * Modify cloudwatch common to depend on gtest, gmock Use the macro in aws_common to find test dependencies for ROS1 or ROS2. Signed-off-by: Miaofei <miaofei@amazon.com> * more CMakeLists.txt cleanup Signed-off-by: Miaofei <miaofei@amazon.com> * update travis.yml to be compatible with specifying multiple package names Signed-off-by: Miaofei <miaofei@amazon.com> * update travis.yml test matrix Signed-off-by: Miaofei <miaofei@amazon.com> * update PACKAGE_NAMES Signed-off-by: Miaofei <miaofei@amazon.com>
• Adding old release to change log
• Update package.xml for 1.0.2 release Signed-off-by: Ryan Newell <ryanewel@amazon.com>
• Increase test timeout to fix flakiness (#21) - Increase the test timeout to 5000ms because sometimes SendLogsToCloudWatch isn't being called fast enough and so the tests fail.
• Fixes issue with DescribeLogStreams being called to get the next token - Call DescribeLogStreams only one on startup - Implements a state machine to manage the core run loop instead of checking if the log stream exists every tick.
• Update success logs' severity to DEBUG in cloudwatch_facade.cpp
• Release 1.0.1 (#14) * Release 1.0.1 * 1.0.1
• adding unit tests for cloudwatch facade
• Merge pull request #4 from juanrh/improve-coverage-cloudwatch_logger Improve coverage cloudwatch logger
• Make LogManagerFactory mockeable
• Make cloudwatch_logs_common shared lib to use it in other libs
• Merge pull request #1 from xabxx/master [Bug Fix] Resolved false-positive error log messages
• Resolved false-positive error log messages
• Contributors: AAlon, Abby Xu, Devin Bonnie, M. M, Nick Burek, Ross Desmond, Ryan Newell, Tim Robinson, Yuan "Forrest" Yu, hortala, ryanewel

1.0.0 (2019-03-20)