➜
This issue tracker has been migrated to GitHub,
and is currently read-only.
For more information,
see the GitHub FAQs in the Python's Developer Guide.
Created on 2010-07-29 09:02 by pitrou, last changed 2022-04-11 14:57 by admin. This issue is now closed.
| Messages (4) | |||
|---|---|---|---|
| msg111901 - (view) | Author: Antoine Pitrou (pitrou) *  |
Date: 2010-07-29 09:02 | |
See http://mail.python.org/pipermail/python-checkins/2010-July/095648.html test_smtpd leaked [25, 25, 25] references, sum=75 |
|||
| msg111912 - (view) | Author: Richard Jones (richard) * |
Date: 2010-07-29 11:43 | |
That's odd. I didn't run the refcount tests because I was only adding Python code. I'll look into compiling a debug build and running the tests locally with a view to tracking down the problem. |
|||
| msg111914 - (view) | Author: Antoine Pitrou (pitrou) * |
Date: 2010-07-29 11:53 | |
> That's odd. I didn't run the refcount tests because I was only adding > Python code. That's not odd. A reference leak can indicate that some objects hang in memory permanently rather than being cleaned up after the test run. Looking at the code you added, DummyServer.messages will slowly accumulate more and more Python objects as the tests get run several times. (it's not necessarily the only cause of the refleak, though) |
|||
| msg111919 - (view) | Author: Georg Brandl (georg.brandl) * |
Date: 2010-07-29 13:20 | |
The culprit was indeed DummyServer.messages. Fixed in r83222. |
|||
| History | |||
|---|---|---|---|
| Date | User | Action | Args |
| 2022-04-11 14:57:04 | admin | set | github: 53658 |
| 2010-07-29 13:20:00 | georg.brandl | set | status: open -> closed nosy: + georg.brandl messages: + msg111919 resolution: fixed |
| 2010-07-29 11:53:38 | pitrou | set | messages: + msg111914 |
| 2010-07-29 11:43:13 | richard | set | messages: + msg111912 |
| 2010-07-29 09:02:30 | pitrou | create | |