Log how long each extension module takes to import #1171
Merged
Conversation
This file contains 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
Based on some detailed debugging in 2i2c-org/infrastructure#2047 to figure out why a Jupyter Server *process* somtimes takes more than 30s to start, the primary culprit was server extensions that took multiple seconds to just even import. Thanks to some ad-hoc patching (2i2c-org/infrastructure#2047 (comment)), I was able to figure out which were the slow extensions. This PR emits extension import time as log messages, so this information is *much* more visible. I also explicitly chose info instead of debug, primarily because I believe it is *very* important to surface this performance information to users, so they can go bug the appropriate extension. Otherwise, it just feels like 'jupyter server is slow!'. This is compounded by the fact that while notebook server doesn't import *disabled* extensions, jupyter_server does seem to - so it's hard to isolate this.
for more information, see https://pre-commit.ci
This was referenced Jan 12, 2023
Codecov ReportBase: 79.93% // Head: 79.56% // Decreases project coverage by
Additional details and impacted files@@ Coverage Diff @@
## main #1171 +/- ##
==========================================
- Coverage 79.93% 79.56% -0.37%
==========================================
Files 68 68
Lines 8124 8130 +6
Branches 1601 1602 +1
==========================================
- Hits 6494 6469 -25
- Misses 1205 1222 +17
- Partials 425 439 +14
Help us with your feedback. Take ten seconds to tell us how you rate us. Have a feature suggestion? Share it here. ☔ View full report at Codecov. |
|
Lint failure seems totally unrelated? |
davidbrochart
approved these changes
Jan 12, 2023
Thanks @yuvipanda, yes it is unrelated. |
kevin-bates
reviewed
Jan 12, 2023
|
Thanks a lot for getting this out quickly, @davidbrochart and @blink1073 :) |
yuvipanda
added a commit
to yuvipanda/jupyter_server
that referenced
this pull request
Jan 14, 2023
Without this, the log messages printed in jupyter-server#1171 don't actually make it out. Whoops! LoggingConfigurable inherits from HasTraits, and sets up Logging appropriately so we can have a log passed through.
yuvipanda
added a commit
to yuvipanda/jupyter_server
that referenced
this pull request
Jan 14, 2023
Without this, the log messages printed in jupyter-server#1171 don't actually make it out. Whoops! LoggingConfigurable inherits from HasTraits, and sets up Logging appropriately so we can have a log passed through.
This was referenced Jan 14, 2023
kevin-bates
pushed a commit
that referenced
this pull request
Jan 14, 2023
Without this, the log messages printed in #1171 don't actually make it out. Whoops! LoggingConfigurable inherits from HasTraits, and sets up Logging appropriately so we can have a log passed through.
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
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.
Based on some detailed debugging in
2i2c-org/infrastructure#2047 to figure out why a Jupyter Server process sometimes takes more than 30s
to start, the primary culprit was server extensions that took multiple seconds to just even import. Thanks to some ad-hoc patching (2i2c-org/infrastructure#2047 (comment)), I was able to figure out which were the slow extensions. This PR emits extension import time as log messages, so this information is much more visible.
I also explicitly chose info instead of debug, primarily because I believe it is very important to surface this performance information to users, so they can go bug the appropriate extension. Otherwise, it just feels like 'jupyter server is slow!'. This is compounded by the fact that while notebook server doesn't import disabled extensions, jupyter_server does seem to - so it's hard to isolate this.