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.

Author vinay.sajip
Recipients Frans, eantelman, ggenellina, jpfarias, lowell87, mramahi77, neyro, rcronk, vinay.sajip
Date 2010-03-27.13:46:42
SpamBayes Score 3.4649406e-10
Marked as misclassified No
Message-id <714221.58327.qm@web25805.mail.ukl.yahoo.com>
In-reply-to <1269687501.3.0.0205743205878.issue4749@psf.upfronthosting.co.za>
Content
João Paulo Farias <jpaulofarias@gmail.com> 
added the comment:

> Ok.. I was hoping for a solution that did not involve running a 
> separate process just for logging.

It needn't *just* be for logging - as long as it doesn't spawn any child processes. And the need to run a separate process stems from the fact that there's no reliable cross-platform way (in the Python stdlib) of synchronising access to a resource (such as a file) from multiple processes.

There are other solutions, no doubt, but the separate process will probably be the quickest and easiest to implement.

> Also, I noticed that I get 
> the problem even before spawning any 
> processes.

If you're sure about this, and not running into any of the other gotchas mentioned in the thread, then please post a small script which succinctly demonstrates the problem, and I'll look at it. Numerous people have posted problem scripts in this thread (e.g. Neyro, Robert Cronk, Frans) but the problem has generally turned out to be not related to logging specifically, but more generically handle inheritance, differences/limitations in implementations of subprocess under Windows etc.
History
Date User Action Args
2010-03-27 13:46:45vinay.sajipsetrecipients: + vinay.sajip, ggenellina, mramahi77, lowell87, neyro, rcronk, Frans, eantelman, jpfarias
2010-03-27 13:46:43vinay.sajiplinkissue4749 messages
2010-03-27 13:46:43vinay.sajipcreate