Message 195256 - Python tracker

Message195256

Author	pitrou
Recipients	christian.heimes, neologix, pitrou, sbt, vstinner
Date	2013-08-15.13:44:51
SpamBayes Score	-1.0
Marked as misclassified	Yes
Message-id	<1376574283.2755.10.camel@fsol>
In-reply-to	<520CD915.9090003@cheimes.de>

Content
> When the OpenSSL's CPRNG is already initialized before 3) than all child > processes created by 3) will have almost the same PRNG state. According > to http://bugs.ruby-lang.org/issues/4579 the PRNG will return the same > value when PID numbers are recycled. Thanks. Here is some discussion of the reseeding strategy: http://marc.info/?l=openssl-dev&m=130432419329454&w=2 More precisely, instead of reseeding in the child, you can simply perturb the PRNG with a constant in the parent, to make sure the same PRNG state doesn't get re-used.

Content

> When the OpenSSL's CPRNG is already initialized before 3) than all child
> processes created by 3) will have almost the same PRNG state. According
> to http://bugs.ruby-lang.org/issues/4579 the PRNG will return the same
> value when PID numbers are recycled.

Thanks. Here is some discussion of the reseeding strategy:
http://marc.info/?l=openssl-dev&m=130432419329454&w=2

More precisely, instead of reseeding in the child, you can simply
perturb the PRNG with a constant in the parent, to make sure the same
PRNG state doesn't get re-used.

History
Date	User	Action	Args
2013-08-15 13:44:51	pitrou	set	recipients: + pitrou, vstinner, christian.heimes, neologix, sbt
2013-08-15 13:44:51	pitrou	link	issue18747 messages
2013-08-15 13:44:51	pitrou	create