[Pkg-net-snmp-devel] Bug#510495: memory profiler results

[John Morrissey]

On Sun, Jan 18, 2009 at 02:03:32PM -0500, Dave Johnson wrote:
> It's unclear if ipAddressTable_cache_load is meant to handle IPv6
> addresses and if so, there is clearly brokenness for duplicate IPv6
> link local addresses (which will of course exist when VLANs are
> present)
> 
> _check_entry_for_updates() only does a binary search for the first
> matching ipv4/ipv6 address, ignoring the fact duplicate ip addresses
> appear to be present in the list.  Again, unclear if these duplicates
> should be in the list or not.  someone more familiar with this code
> would have to comment.

Dug around in the upstream bug tracker a little, and found:

[ 1521999 ] ipAddrTable caching code leaks memory
https://sourceforge.net/tracker/index.php?func=detail&aid=1521999&group_id=12694&atid=112694

The patch attached to this upstream bug goes a little too far and
double-frees ipaddress_entry (ipAddressTable_release_rowreq_ctx() frees its
associated data automatically). I've modified it to remove the snmp_log()
noise and double-free; the resulting packaging diff is attached and has not
leaked during the ~half day it's been up here.

The lenny leak must be different, since this fix is already applied upstream
in the lenny snmpd.

john
-- 
John Morrissey          _o            /\         ----  __o
jwm at horde.net        _-< \_          /  \       ----  <  \,
www.horde.net/    __(_)/_(_)________/    \_______(_) /_(_)__
-------------- next part --------------
A non-text attachment was scrubbed...
Name: 510495-etch-snmpd-leakplug.patch
Type: text/x-diff
Size: 1510 bytes
Desc: not available
Url : http://lists.alioth.debian.org/pipermail/pkg-net-snmp-devel/attachments/20090122/4c021623/attachment.patch