Possible bug in svn, svn dav, neon, apache etc.
From: Antti Siukola <siukola.antti_at_gmail.com>
Date: Thu, 28 Apr 2011 09:11:13 +0300
We've encountered a strange behaviour with our quite loaded (avg 1500 http
The problem is that occasionally couple of httpd processes start eating huge
Workarounds exist but stracing the bad processes reveal quite clear bug
mmap(NULL, 1048576, PROT_READ|PROT_WRITE, MAP_PRIVATE|MAP_ANONYMOUS, -1, 0)
This repeats forever (returned mem addr changes every time of course) until
We can see this only in our production environment which makes debugging a
One notable thing might be that the storage is on FC and IO scheduler is
Br, Antti Siukola
This is an archived mail posted to the Subversion Users mailing list.