You are correct; you got me there. I only looked at the running output. Which already had a bunch of initialization leaks which I ignored because that was not my concern - my target was the ever increasing leak over time that top reported. But valgrind did not report anything overtime, it must be because I did not stop snmpd.
I will re-run by stopping snmpd and report the output to you. FYI - I'm now running with if-mib code mostly removed because I suspect the if-mib; the if-mib is not removable from net-snmp to the best of my knowledge. Please note that on linux host (not embedded hardware) "top" does not report much of a memory leak (its less than 0.1% overnight which is not as alarming as a 50% memory consumption while starting at 7% initially). Kavita --- On <b>Mon, 9/27/10, Bart Van Assche <i><bvanass...@acm.org></i></b> wrote: <blockquote style="border-left: 2px solid rgb(16, 16, 255); margin-left: 5px; padding-left: 5px;"> From: Bart Van Assche <bvanass...@acm.org> Subject: Re: Memory leak in 5.5 To: "kavita raghunathan" <kraghuna2...@yahoo.com> Cc: net-snmp-users@lists.sourceforge.net Date: Monday, September 27, 2010, 11:12 AM <div id="yiv1629996917"><font size="2">On Mon, Sep 27, 2010 at 5:45 PM, kavita raghunathan <span dir="ltr"><<a rel="nofollow" ymailto="mailto:kraghuna2...@yahoo.com"; target="_blank" href="/mc/compose?to=kraghuna2...@yahoo.com">kraghuna2...@yahoo.com</a>></span> wrote: </font><div class="yiv1629996917gmail_quote"><blockquote class="yiv1629996917gmail_quote" style="border-left: 1px solid rgb(204, 204, 204); margin: 0pt 0pt 0pt 0.8ex; padding-left: 1ex;"> <table border="0" cellpadding="0" cellspacing="0"><tbody><tr><td style="font-family: inherit; font-style: inherit; font-variant: inherit; font-weight: inherit; line-height: inherit; font-size-adjust: inherit; font-stretch: inherit;" valign="top"> <font size="2">Valgrind found no leaks. I left it over night. But top still reports that snmpd is leaking memory. I did do what he suggested. What I'm trying to say is, is there any known memory leaks with net-snmp 5.5 ? Has anyone else reported a problem ? Google does not show anyone reporting a problem, but maybe you all know of one. Should I be trying dmalloc ? </font></td></tr></tbody></table></blockquote></div><font size="2"> Hello Kavita, Are you aware that Valgrind only prints a leak report when snmpd stops, not while it is running ? After snmpd has been running for a whole night under Valgrind, you should stop snmpd and capture the Valgrind output in a file. Analysis of that output will tell whether or not there is a leak. Note: Valgrind will report many "leaks" on snmpd. Not everything that Valgrind reports is a bug though - Valgrind also complains when memory is allocated at startup but not freed at exit. That is not the same as a memory leak. Careful analysis of the Valgrind output is necessary before a conclusion can be drawn. </font><font size="2">Regarding known bugs in 5.5: please have a look at the Net-SNMP 5.6.pre3 ChangeLog. All bugs that have been fixed since version 3-0A1 are documented in that file. Bart. </font> </div></blockquote> ------------------------------------------------------------------------------ Start uncovering the many advantages of virtual appliances and start using them to simplify application deployment and accelerate your shift to cloud computing. http://p.sf.net/sfu/novell-sfdev2dev _______________________________________________ Net-snmp-users mailing list Net-snmp-users@lists.sourceforge.net Please see the following page to unsubscribe or change other options: https://lists.sourceforge.net/lists/listinfo/net-snmp-users
