Hello. My apologies to Niels if he has already replied to this thread.
The NULL-pointer dereferencing patch was applied today to 5.0 patches, 5.1. patches, and current branches.
The snmptrapd memory leaks that you provided fixes for was managed with different patches, [Thanks to Niels!!], but nonetheless, you should see marked improvement with the next release.
If you are impatient, please download net-snmp-5.1.2.pre2.tar.gz, and apply your p2_netsnmp.patch (the one which patches snmp_api.c). If you can re-run your memory leak tests, we believe you'll be happy-er with the results.
Bosko Milekic wrote:
Hi,
A week or so ago I sent you guys a diff for partial OID matching as well as a bug-fix for snmptrapd which fixes a NULL pointer dereference on memory allocation failure. The responses seemed to indicate that the partial OID matching (or a modification thereof) would get committed in time for your next release. I don't know what the concensus was regarding the second bug-fix patch because I am not subscribed to this list and your mail archives on sourceforge seem to only archive up-to the previous month, so I could not follow the discussion.
I'm resending you the NULL-pointer dereference bug fix, but I'm also sending you now an additional bug-fix which I believe fixes a couple of pretty serious memory leaks in snmptrapd. I'll elaborate a little on the memory leaks.
We run snmptrapd here and it receives a rather large number of snmp traps. I've noticed that as traps are received and handled, the size (including resident size) of snmptrapd grows without bound, to the point where this morning (after ~24hrs of trap processing), the size of snmptrapd was 105M. If I stop sending traps to the daemon, its size remains the same.
I've built and linked snmptrapd with debugging symbols and with the Hans Boehm GC and after some fiddling around spotted at least the two memory leaks I think I plugged with the p3 patch included in this Email. There may be a few more leaks, but I have not determined with certainty where they are (yet).
Please include me in your CC lines. I'd love to subscribe but I'm not sure I can handle the mail load right now. :-)
Regards,
------------------------------------------------------- This SF.Net email is sponsored by The 2004 JavaOne(SM) Conference Learn from the experts at JavaOne(SM), Sun's Worldwide Java Developer Conference, June 28 - July 1 at the Moscone Center in San Francisco, CA REGISTER AND SAVE! http://java.sun.com/javaone/sf Priority Code NWMGYKND _______________________________________________ Net-snmp-coders mailing list [EMAIL PROTECTED] https://lists.sourceforge.net/lists/listinfo/net-snmp-coders
