[issue25569] Memory leak in SSLSocket.getpeercert()

[Roundup Robot]

Roundup Robot added the comment:

New changeset 3b9fb8ebf44f by Benjamin Peterson in branch '2.7':
fix build with older openssl (#25569)
https://hg.python.org/cpython/rev/3b9fb8ebf44f

New changeset f13a75544b6f by Benjamin Peterson in branch '3.4':
fix build with older openssl (#25569)
https://hg.python.org/cpython/rev/f13a75544b6f

New changeset 83ea7e75605a by Benjamin Peterson in branch '3.5':
merge 3.4 (#25569)
https://hg.python.org/cpython/rev/83ea7e75605a

New changeset 8ac09e46ca45 by Benjamin Peterson in branch 'default':
merge 3.5 (#25569)
https://hg.python.org/cpython/rev/8ac09e46ca45

--

___
Python tracker 

___
___
Python-bugs-list mailing list
Unsubscribe: 
https://mail.python.org/mailman/options/python-bugs-list/archive%40mail-archive.com

[issue25569] Memory leak in SSLSocket.getpeercert()

[Alex Gaynor]

Alex Gaynor added the comment:

fixed

--
status: open -> closed

___
Python tracker 

___
___
Python-bugs-list mailing list
Unsubscribe: 
https://mail.python.org/mailman/options/python-bugs-list/archive%40mail-archive.com

[issue25569] Memory leak in SSLSocket.getpeercert()

[Martin Panter]

Martin Panter added the comment:

Seems to have broken the SSL module. Looks like you typoed “dps” as “dsp” in 
some conditional code.

http://buildbot.python.org/all/builders/AMD64%20Snow%20Leop%203.4/builds/1326/steps/compile/logs/stdio

building '_ssl' extension
gcc -fno-strict-aliasing -g -O0 -Wall -Wstrict-prototypes 
-Werror=declaration-after-statement -I./Include -I. -IInclude 
-I/Users/buildbot/buildarea/3.4.murray-snowleopard/build/Include 
-I/Users/buildbot/buildarea/3.4.murray-snowleopard/build -c 
/Users/buildbot/buildarea/3.4.murray-snowleopard/build/Modules/_ssl.c -o 
build/temp.macosx-10.6-x86_64-3.4-pydebug/Users/buildbot/buildarea/3.4.murray-snowleopard/build/Modules/_ssl.o
/Users/buildbot/buildarea/3.4.murray-snowleopard/build/Modules/_ssl.c: In 
function ‘_get_crl_dp’:
/Users/buildbot/buildarea/3.4.murray-snowleopard/build/Modules/_ssl.c:1084: 
error: ‘dsp’ undeclared (first use in this function)
/Users/buildbot/buildarea/3.4.murray-snowleopard/build/Modules/_ssl.c:1084: 
error: (Each undeclared identifier is reported only once
/Users/buildbot/buildarea/3.4.murray-snowleopard/build/Modules/_ssl.c:1084: 
error: for each function it appears in.)
/Users/buildbot/buildarea/3.4.murray-snowleopard/build/Modules/_ssl.c: In 
function ‘_setup_ssl_threads’:
/Users/buildbot/buildarea/3.4.murray-snowleopard/build/Modules/_ssl.c:3828: 
warning: comparison is always false due to limited range of data type

--
nosy: +martin.panter
status: closed -> open
versions: +Python 3.4

___
Python tracker 

___
___
Python-bugs-list mailing list
Unsubscribe: 
https://mail.python.org/mailman/options/python-bugs-list/archive%40mail-archive.com

[issue25569] Memory leak in SSLSocket.getpeercert()

[Martin Panter]

Changes by Martin Panter :


--
nosy: +benjamin.peterson

___
Python tracker 

___
___
Python-bugs-list mailing list
Unsubscribe: 
https://mail.python.org/mailman/options/python-bugs-list/archive%40mail-archive.com

[issue25569] Memory leak in SSLSocket.getpeercert()

[Roundup Robot]

Roundup Robot added the comment:

New changeset 10c3646b2d59 by Benjamin Peterson in branch '2.7':
fix memory leak in _get_crl_dp (closes #25569)
https://hg.python.org/cpython/rev/10c3646b2d59

New changeset aabe273b20ab by Benjamin Peterson in branch '3.4':
fix memory leak in _get_crl_dp (closes #25569)
https://hg.python.org/cpython/rev/aabe273b20ab

New changeset 07a298572d93 by Benjamin Peterson in branch '3.5':
merge 3.5 (#25569)
https://hg.python.org/cpython/rev/07a298572d93

New changeset fb55b1ab43fc by Benjamin Peterson in branch 'default':
merge 3.5 (#25569)
https://hg.python.org/cpython/rev/fb55b1ab43fc

--
nosy: +python-dev
resolution:  -> fixed
stage:  -> resolved
status: open -> closed

___
Python tracker 

___
___
Python-bugs-list mailing list
Unsubscribe: 
https://mail.python.org/mailman/options/python-bugs-list/archive%40mail-archive.com

[issue25569] Memory leak in SSLSocket.getpeercert()

[Stéphane Wirtel]

Stéphane Wirtel added the comment:

Hi alex,

Thank you for your feeback, I will propose an other patch with your remarks.

Stephane

--

___
Python tracker 

___
___
Python-bugs-list mailing list
Unsubscribe: 
https://mail.python.org/mailman/options/python-bugs-list/archive%40mail-archive.com

[issue25569] Memory leak in SSLSocket.getpeercert()

[Stéphane Wirtel]

Stéphane Wirtel added the comment:

sorry for the delay.

here is a new version of my patch, if I can have feedback about this patch, I 
will really appreciate.

thank you

--
Added file: http://bugs.python.org/file40976/issue25569-2.patch

___
Python tracker 

___
___
Python-bugs-list mailing list
Unsubscribe: 
https://mail.python.org/mailman/options/python-bugs-list/archive%40mail-archive.com

[issue25569] Memory leak in SSLSocket.getpeercert()

[Alex Gaynor]

Alex Gaynor added the comment:

Tests pass and the original script runs without a leak using this patch. It 
could probably be shorter if we converted from local returns to `goto fail` or 
something, but I don't really have an opinion.

--

___
Python tracker 

___
___
Python-bugs-list mailing list
Unsubscribe: 
https://mail.python.org/mailman/options/python-bugs-list/archive%40mail-archive.com

[issue25569] Memory leak in SSLSocket.getpeercert()

[Alex Gaynor]

New submission from Alex Gaynor:

Run the following code:

import socket
import ssl
import sys


def main():
ctx = ssl.create_default_context()
s = socket.create_connection(('www.bing.com', 443))
s = ctx.wrap_socket(s, server_hostname='www.bing.com')
while True:
s.getpeercert()

sys.stderr.write('.')
sys.stderr.flush()


if __name__ == '__main__':
main()



and observe memory usage, you'll see it increases by a MB/s or so.

--
messages: 254210
nosy: alex
priority: normal
severity: normal
status: open
title: Memory leak in SSLSocket.getpeercert()
versions: Python 2.7, Python 3.5, Python 3.6

___
Python tracker 

___
___
Python-bugs-list mailing list
Unsubscribe: 
https://mail.python.org/mailman/options/python-bugs-list/archive%40mail-archive.com

[issue25569] Memory leak in SSLSocket.getpeercert()

[Alex Gaynor]

Changes by Alex Gaynor :


--
nosy: +christian.heimes, dstufft, giampaolo.rodola, janssen, pitrou

___
Python tracker 

___
___
Python-bugs-list mailing list
Unsubscribe: 
https://mail.python.org/mailman/options/python-bugs-list/archive%40mail-archive.com

[issue25569] Memory leak in SSLSocket.getpeercert()

[Stéphane Wirtel]

Stéphane Wirtel added the comment:

Maybe you are right, I didn't see that this function returns a pointer in 
function of the parameter. sorry.


Could you try with this patch, really sorry but my internet connection is 
really slow and I have a problem with the _ssl module on my laptop (problem 
with the compilation and the dependencies because I am on OSX).

Thank you

--
keywords: +patch
Added file: http://bugs.python.org/file40963/issue25569.patch

___
Python tracker 

___
___
Python-bugs-list mailing list
Unsubscribe: 
https://mail.python.org/mailman/options/python-bugs-list/archive%40mail-archive.com

[issue25569] Memory leak in SSLSocket.getpeercert()

[Alex Gaynor]

Alex Gaynor added the comment:

A probably source of the leak is here: 
https://github.com/python/cpython/blob/master/Modules/_ssl.c#L1073-L1076 `dps` 
is never freed. (This is with OpenSSL 0.9.8zg)

--

___
Python tracker 

___
___
Python-bugs-list mailing list
Unsubscribe: 
https://mail.python.org/mailman/options/python-bugs-list/archive%40mail-archive.com

[issue25569] Memory leak in SSLSocket.getpeercert()

[Stéphane Wirtel]

Stéphane Wirtel added the comment:

Hi Alex,

not sure, but I think the right function to free the memory is 
AUTHORITY_INFO_ACCESS_free(dps)

I will provide a patch for a review asap

--
nosy: +matrixise

___
Python tracker 

___
___
Python-bugs-list mailing list
Unsubscribe: 
https://mail.python.org/mailman/options/python-bugs-list/archive%40mail-archive.com

[issue25569] Memory leak in SSLSocket.getpeercert()

[Alex Gaynor]

Alex Gaynor added the comment:

I think you want sk_DIST_POINT_free actually.

--

___
Python tracker 

___
___
Python-bugs-list mailing list
Unsubscribe: 
https://mail.python.org/mailman/options/python-bugs-list/archive%40mail-archive.com