Mercurial > p > roundup > code
view website/www/index.txt @ 6565:2c2dbfc332ba
Try to handle multiple connections better.
The session database is a hot spot. When multiple requests (e.g. 20)
come in at the same time session database contention can get great.
The original code didn't retry session database access when the open
failed. This resulted in errors at the client.
The second pass delayed 0.01 seconds and retried. It was better but we
still had multiple second stalls. I think the first request got in,
everybody else backed up and then retried at the same time. Again they
stepped on each other. With logging I would see many counters go all
the way to low single digits or to -1 indicating falure.
This pass uses randomint to generate delays from 0-.125 seconds in 5ms
increments. This performs better in testing. I rarely saw a counter
less than 13 (2 failed retries). Current logging starts after 6
failures and counts down until success or failure.
| author | John Rouillard <rouilj@ieee.org> |
|---|---|
| date | Thu, 16 Dec 2021 20:02:00 -0500 |
| parents | cbc18a8bc61f |
| children | aa52c5e114b2 |
line wrap: on
line source
Roundup Issue Tracker ===================== .. raw:: html <div class="release_info note">Download: <a href="https://pypi.org/project/roundup/">latest</a></div> Roundup is a simple-to-use and -install issue-tracking system with command-line, web and e-mail interfaces. It is based on the winning design from Ka-Ping Yee in the Software Carpentry "Track" design competition. The current stable version of Roundup is 2.1.0. It is a bug fix and minor feature release for the major 2.0.0 release which added: * Python 2 and Python 3 support * a new REST interface * updates to jinja2 templates including security improvements Fixes and features in the 2.1.0 release include: * Installation uses setuptools and not distutils. * Mysql backend now uses an index to make sure that key values are not duplicated when two roundup processes run in parallel. * Postgres back end now uses a server side cursor, so large queries won't consume huge amounts of memory. * Security fixes for jQuery, markdown handling, * Valid class names are documented and enforced. All class names now match ``[A-z][A-z0-9_]+[A-z_]``. * Fixes/improvements to jinja2 templates * Fixes for python3 compatibility. * Fix sorting of multilinks in templating code. * Password reset documented in user guide. More info on the 58 changes can be found in the `change note`_. For more information on Roundup see the :doc:`design overview <docs/design>`, and all the other :doc:`documentation <docs>`. Roundup has been deployed for: * bug tracking and TODO list management (the classic installation) * customer help desk support (with a wizard for the phone answerers, linking to networking, system and development issue trackers) * issue management for IETF working groups * sales lead tracking * conference paper submission and double-blind referee management * weblogging (well, almost :) ...and so on. It's been designed with :doc:`flexibility <docs/customizing>` in mind - it's not just another bug tracker. Roundup ships with a *demo tracker* to play with - after you've unpacked the source, just run "python demo.py" and load up the URL it prints out! Roundup was originally released as version 0.1.1 in late August, 2001. The first `change note`_ written said: Needed a bug tracking system. Looked around. Tried to install many Perl-based systems, to no avail. Got tired of waiting for Roundup to be released. Had just finished major product project, so needed something different for a while. Roundup here I come... .. _`download`: https://pypi.org/project/roundup/ .. _`change note`: https://sourceforge.net/p/roundup/code/ci/tip/tree/CHANGES.txt .. _`its own set of docs`: https://www.roundup-tracker.org/dev-docs/docs.html