This repository was archived by the owner on May 7, 2024. It is now read-only.
Fix perf issue where main event loop takes 100% of CPU #132
Conversation
This file contains hidden or bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
We have a 2 threads: Thread #1 runs in a loop polling the response queue Thread #2 runs in a loop decoding responses from the sqltoolsservice over stdout and posting them to the response queue Since thread #1 doesn't sleep, it's takes 100% CPU. In addition, running python 2.7 on windows, #2 doesn’t preempt the CPU due to #1 taking all of the CPU cycles, so no response is processed. Fix is simple – thread #1 needs to sleep so thread #2 can get scheduled and get it’s work done.
|
@pensivebrian, |
MrMeemus
approved these changes
Aug 17, 2017
|
Looks like there is pylint issues, perhaps it's length of comment? or white spaces. |
Codecov Report
@@ Coverage Diff @@
## dev #132 +/- ##
==========================================
- Coverage 83.26% 83.19% -0.08%
==========================================
Files 15 15
Lines 1046 1047 +1
==========================================
Hits 871 871
- Misses 175 176 +1
Continue to review full report at Codecov.
|
MrMeemus
added a commit
that referenced
this pull request
Sep 1, 2017
* Use SqlToolsService built on .NET Core 2.0 and a build script updates (#131) * Bump version to 1.0.0a19 * Use .NET Core 2.0 RTM built sqltoolsservice * Add build script to upload to azure blob storage * Upgrade to VS 2017 * Remove 3.3 as supported Python version * Fix perf issue where main event loop takes 100% of CPU (#132) Fix perf issue where main event loop takes 100% of CPU We have a 2 threads: Thread #1 runs in a loop polling the response queue Thread #2 runs in a loop decoding responses from the sqltoolsservice over stdout and posting them to the response queue Since thread #1 doesn't sleep, it's takes 100% CPU. In addition, running python 2.7 on windows, #2 doesn’t preempt the CPU due to #1 taking all of the CPU cycles, so no response is processed. Fix is simple – thread #1 needs to sleep so thread #2 can get scheduled and get it’s work done. * Refine event loop perf fix in main.py Refine event loop perf fix in main.py * Fixing regular expression Previous regex would result in release:a1 and release_version: 12. Modified the regex for part Release to only pick up lower case letters. * Adding missing forward slash on test pypi url * fixing typos/grammar (#138) fixing typos/grammar. * Updating to release version 1.0.0a20.
MrMeemus
added a commit
that referenced
this pull request
Oct 13, 2017
* Use SqlToolsService built on .NET Core 2.0 and a build script updates (#131) * Bump version to 1.0.0a19 * Use .NET Core 2.0 RTM built sqltoolsservice * Add build script to upload to azure blob storage * Upgrade to VS 2017 * Remove 3.3 as supported Python version * Fix perf issue where main event loop takes 100% of CPU (#132) Fix perf issue where main event loop takes 100% of CPU We have a 2 threads: Thread #1 runs in a loop polling the response queue Thread #2 runs in a loop decoding responses from the sqltoolsservice over stdout and posting them to the response queue Since thread #1 doesn't sleep, it's takes 100% CPU. In addition, running python 2.7 on windows, #2 doesn’t preempt the CPU due to #1 taking all of the CPU cycles, so no response is processed. Fix is simple – thread #1 needs to sleep so thread #2 can get scheduled and get it’s work done. * Refine event loop perf fix in main.py Refine event loop perf fix in main.py * Fixing regular expression Previous regex would result in release:a1 and release_version: 12. Modified the regex for part Release to only pick up lower case letters. * Adding missing forward slash on test pypi url * fixing typos/grammar (#138) fixing typos/grammar. * Updating to release version 1.0.0a20. * Create doc for official msft docs page * Updated documentation page with usage_guide * Added link to download adventureworks * Updated with sqlcmd usage * Added in run and cloud shell support * Update/consolidate linux install (#153) * universal linux wheel gen and setup update. * Updating version cfg. * Updating sqltoolsservice container. * Updating spacing for flake8. * Updating team email. (#154)
MrMeemus
added a commit
that referenced
this pull request
Nov 10, 2017
* Merge release 1.0.0a21 (#155) * Use SqlToolsService built on .NET Core 2.0 and a build script updates (#131) * Bump version to 1.0.0a19 * Use .NET Core 2.0 RTM built sqltoolsservice * Add build script to upload to azure blob storage * Upgrade to VS 2017 * Remove 3.3 as supported Python version * Fix perf issue where main event loop takes 100% of CPU (#132) Fix perf issue where main event loop takes 100% of CPU We have a 2 threads: Thread #1 runs in a loop polling the response queue Thread #2 runs in a loop decoding responses from the sqltoolsservice over stdout and posting them to the response queue Since thread #1 doesn't sleep, it's takes 100% CPU. In addition, running python 2.7 on windows, #2 doesn’t preempt the CPU due to #1 taking all of the CPU cycles, so no response is processed. Fix is simple – thread #1 needs to sleep so thread #2 can get scheduled and get it’s work done. * Refine event loop perf fix in main.py Refine event loop perf fix in main.py * Fixing regular expression Previous regex would result in release:a1 and release_version: 12. Modified the regex for part Release to only pick up lower case letters. * Adding missing forward slash on test pypi url * fixing typos/grammar (#138) fixing typos/grammar. * Updating to release version 1.0.0a20. * Create doc for official msft docs page * Updated documentation page with usage_guide * Added link to download adventureworks * Updated with sqlcmd usage * Added in run and cloud shell support * Update/consolidate linux install (#153) * universal linux wheel gen and setup update. * Updating version cfg. * Updating sqltoolsservice container. * Updating spacing for flake8. * Updating team email. (#154) * Updating mssqltoolsservice to be integrated as a package of mssqlscripter. * Updating sqltoolsservice to be loaded from the repro instead of storage account. * Fix index file generation for daily storage account. * Fixing manylinux1 tag. * Updating platform tag for win x64. * Renaming sqltoolsservice win x64 folder. * Adding platform tags for win_amd64, manylinux1_x86_64, manylinux1_i686. * version bumping to 1.0.0a22. * Flake8 format fixes. * Erroring out when build receives invalid flag.
MrMeemus
added a commit
that referenced
this pull request
Nov 17, 2017
* Use SqlToolsService built on .NET Core 2.0 and a build script updates (#131) * Bump version to 1.0.0a19 * Use .NET Core 2.0 RTM built sqltoolsservice * Add build script to upload to azure blob storage * Upgrade to VS 2017 * Remove 3.3 as supported Python version * Fix perf issue where main event loop takes 100% of CPU (#132) Fix perf issue where main event loop takes 100% of CPU We have a 2 threads: Thread #1 runs in a loop polling the response queue Thread #2 runs in a loop decoding responses from the sqltoolsservice over stdout and posting them to the response queue Since thread #1 doesn't sleep, it's takes 100% CPU. In addition, running python 2.7 on windows, #2 doesn’t preempt the CPU due to #1 taking all of the CPU cycles, so no response is processed. Fix is simple – thread #1 needs to sleep so thread #2 can get scheduled and get it’s work done. * Refine event loop perf fix in main.py Refine event loop perf fix in main.py * Fixing regular expression Previous regex would result in release:a1 and release_version: 12. Modified the regex for part Release to only pick up lower case letters. * Adding missing forward slash on test pypi url * fixing typos/grammar (#138) fixing typos/grammar. * Updating to release version 1.0.0a20. * Create doc for official msft docs page * Updated documentation page with usage_guide * Added link to download adventureworks * Updated with sqlcmd usage * Added in run and cloud shell support * Update/consolidate linux install (#153) * universal linux wheel gen and setup update. * Updating version cfg. * Updating sqltoolsservice container. * Updating spacing for flake8. * Updating team email. (#154) * Fixing resource warning for sqltoolsservice (#158) * Fixing resource warning for sqltoolsservice by closing stdout after killing process. * Shortening sleep time during shutdown. * Fixing missing bracket. * Updating doc's after repro rename (#160) * Updating files after repro rename. * Fixing flake8 issues. * Ron/platform wheels support (#161) * Merge release 1.0.0a21 (#155) * Use SqlToolsService built on .NET Core 2.0 and a build script updates (#131) * Bump version to 1.0.0a19 * Use .NET Core 2.0 RTM built sqltoolsservice * Add build script to upload to azure blob storage * Upgrade to VS 2017 * Remove 3.3 as supported Python version * Fix perf issue where main event loop takes 100% of CPU (#132) Fix perf issue where main event loop takes 100% of CPU We have a 2 threads: Thread #1 runs in a loop polling the response queue Thread #2 runs in a loop decoding responses from the sqltoolsservice over stdout and posting them to the response queue Since thread #1 doesn't sleep, it's takes 100% CPU. In addition, running python 2.7 on windows, #2 doesn’t preempt the CPU due to #1 taking all of the CPU cycles, so no response is processed. Fix is simple – thread #1 needs to sleep so thread #2 can get scheduled and get it’s work done. * Refine event loop perf fix in main.py Refine event loop perf fix in main.py * Fixing regular expression Previous regex would result in release:a1 and release_version: 12. Modified the regex for part Release to only pick up lower case letters. * Adding missing forward slash on test pypi url * fixing typos/grammar (#138) fixing typos/grammar. * Updating to release version 1.0.0a20. * Create doc for official msft docs page * Updated documentation page with usage_guide * Added link to download adventureworks * Updated with sqlcmd usage * Added in run and cloud shell support * Update/consolidate linux install (#153) * universal linux wheel gen and setup update. * Updating version cfg. * Updating sqltoolsservice container. * Updating spacing for flake8. * Updating team email. (#154) * Updating mssqltoolsservice to be integrated as a package of mssqlscripter. * Updating sqltoolsservice to be loaded from the repro instead of storage account. * Fix index file generation for daily storage account. * Fixing manylinux1 tag. * Updating platform tag for win x64. * Renaming sqltoolsservice win x64 folder. * Adding platform tags for win_amd64, manylinux1_x86_64, manylinux1_i686. * version bumping to 1.0.0a22. * Flake8 format fixes. * Erroring out when build receives invalid flag. * Fixing tag for win64 * Update libunwind8 install for CentOS * Ron/sqltoolsservice update (#163) * Updating sqltoolsservice with self contained version. * Refreshing sqltoolsservice again. * Making mssql-scripter executable and adding null checks in main.py * Adding clean up step to remove build directory after each build. * Fixing path for build directory. * Removing 'pypi' from upload step. * Flake 8 extra line fix.
Sign up for free
to subscribe to this conversation on GitHub.
Already have an account?
Sign in.
Add this suggestion to a batch that can be applied as a single commit.
This suggestion is invalid because no changes were made to the code.
Suggestions cannot be applied while the pull request is closed.
Suggestions cannot be applied while viewing a subset of changes.
Only one suggestion per line can be applied in a batch.
Add this suggestion to a batch that can be applied as a single commit.
Applying suggestions on deleted lines is not supported.
You must change the existing code in this line in order to create a valid suggestion.
Outdated suggestions cannot be applied.
This suggestion has been applied or marked resolved.
Suggestions cannot be applied from pending reviews.
Suggestions cannot be applied on multi-line comments.
Suggestions cannot be applied while the pull request is queued to merge.
Suggestion cannot be applied right now. Please check back later.
We have a 2 threads:
Thread 1 runs in a loop polling the response queue
Thread 2 runs in a loop decoding responses from the sqltoolsservice over stdout and posting them to the response queue
Since thread 1 doesn't sleep, it's takes 100% CPU. In addition, running python 2.7 on windows, thread 2 doesn’t preempt the CPU due to thread 1 taking all of the CPU cycles, so no response is processed.
Fix is simple – thread 1 needs to sleep so thread 2 can get scheduled and get it’s work done.