On Fri, 2011-02-04 at 09:11 +0100, Stefan Urbanek wrote: > Hi, > > I'm trying to fetch 1m+ pages over HTTP from single site. urlgrabber > seemed like fast solution using "keep alive" connections. However, > after a while my script fails without any stack trace with: "Fatal > Python error: deallocating None"
That's a known pycurl bug, where it doesn't increment the reference to None for a specific code path. I can't find the BZ for it now but this is the rpm changelog: * Tue Aug 17 2010 Jeffrey C. Ollie <[email protected]> - 7.19.0-7 - Add patch developed by David Malcolm to fix segfaults caused by a missing incref _______________________________________________ Yum-devel mailing list [email protected] http://lists.baseurl.org/mailman/listinfo/yum-devel
