Hi,
Thanks for your prompt answer.
From your initial response I have to understand that this is not
normal. I was already thinking that when I found nothing in this mailing
list about memory leakage.
Therefore, I have some envrionmental or other situation to deal with. I
just don't get how I could get the same symptoms across all these
different environments.
1. I will prepare some tests with valgrind.
2. I do not at present actively use snmpd. Although snmpd is running, it
is not actively used by anything on our systems today. I am just getting
started with snmpd. What this means is that snmpd is chewing up memory,
even though is has virtually no requests. The most I have done is the
occasional snmpwalk. (manually).
2a) At least one of my copies of snmpd is running with a cong file of:
============snmpd conf ============
rocommunity xxxx
================================
2b) Others (esp. version 5.4) are simply running with the default
snmpd.conf in the distribution.
2c) There may be other programs we are running that may be trying to
read information from some of these systems (we have some versions of
various system monitors running, who may be polling all they can
reach), but this is minimal at best.
3) snmptrapd: This is actively used. It receives traps from Colubris
network devices: heartbeats, user association/disassociation.
Depending on the host, we have somewhere between 20-150 devices sending
traps. heartbeats are set to intervals of 60 seconds on one system, 150
on another.
There are essentially no other traps being received. The actual trap
handling is provided by perl routines. The same perl modules/routines
are used on both versions of Linux and on Solaris.
On Linux, I run the perl code embedded, (if 5.4), on solaris via a small
wrapper that feeds the trap through a socket to a trap receiver (running
the same code as the others).
On one of the Centos systems running 5.4, I turned off the traps (by
redirecting port 162) and two days later no increase in memory usages
was found. This is interesting in itself.
4) For transport, we use udp ports 161 & 162 only.
5) Transient interfaces: On some of these systems we had up to 85 pptpd
connections running. This was a very unstable setup; I don't think that
net-snmp and large numbers of pptpd sessions cohabited very well. For
some time, the snmp traps were going through the pptp links. This is no
longer the case. We had many problems with the systems going into load
averages of 85 for a short while (20-30 minutes or so) on a regular
basis when we had a large number of pptp connections running along
with snmptrapd.
Further information:
net-snmp running on Centos was compiled with gcc:
[EMAIL PROTECTED] install]# gcc -v
Reading specs from /usr/lib/gcc/i386-redhat-linux/3.4.6/specs
Configured with: ../configure --prefix=/usr --mandir=/usr/share/man
--infodir=/usr/share/info --enable-shared --enable-threads=posix
--disable-checking --with-system-zlib --enable-__cxa_atexit
--disable-libunwind-exceptions --enable-java-awt=gtk
--host=i386-redhat-linux
Thread model: posix
gcc version 3.4.6 20060404 (Red Hat 3.4.6-3)
[EMAIL PROTECTED] install]#
net-snmp on solaris was compiled with gcc:
[EMAIL PROTECTED] gcc -v
Reading specs from /usr/sfw/lib/gcc/sparc-sun-solaris2.10/3.4.3/specs
Configured with:
/gates/sfw10/builds/sfw10-gate/usr/src/cmd/gcc/gcc-3.4.3/configure
--prefix=/usr/sfw --with-as=/usr/sfw/bin/gas --with-gnu-as
--with-ld=/usr/ccs/bin/ld --without-gnu-ld --enable-languages=c,c++
--enable-shared
Thread model: posix
gcc version 3.4.3 (csl-sol210-3_4-branch+sol_rpath)
[EMAIL PROTECTED]
net-snmp on RH8 (kernel 2.4.31):
[EMAIL PROTECTED] root]# gcc -v
Reading specs from /usr/lib/gcc-lib/i386-redhat-linux/3.2/specs
Configured with: ../configure --prefix=/usr --mandir=/usr/share/man
--infodir=/usr/share/info --enable-shared --enable-threads=posix
--disable-checking --host=i386-redhat-linux --with-system-zlib
--enable-__cxa_atexit
Thread model: posix
gcc version 3.2 20020903 (Red Hat Linux 8.0 3.2-7)
[EMAIL PROTECTED] root]#
Is there anything particular I can look for while setting up the
valgrind tests? Is there a possibility I am linking with incorrect
systems libraries?
Thomas Anders wrote:
Mike Schmidt wrote:
I have net-snmp 5.4 running on
- RH8 and CentOS 4.4 systems (compiled with no options to configure)
- Solaris 10. (compiled without embedded-perl because of perl issues)
I also have some copies of 5.1.2 (the official CentOS version) installed
as binary from the CentOS distribution.
In all of these installations, both snmpd and snmptrapd suck up memory
until the system runs out of swap space and dies.
We'd need more information to be able to help:
1) Run snmpd and snmptrapd under valgrind (or the like) and tell us the results
2) For snmpd, what objects are you querying and how often? For snmptrapd, what
notifications does it get and how often? Is this all over UDP/IPv4 transport? Is
there anything special about these systems (e.g. transient interfaces like ppp)?
+Thomas
begin:vcard
fn:Mike Schmidt
n:Schmidt;Mike
org:Intello Technologies Inc.
adr;quoted-printable;quoted-printable:;;580 1=C3=A8re Rue;Saint-Jean-sur-Richelieu;Qu=C3=A9;J2X 3B4;Canada
email;internet:[EMAIL PROTECTED]
title:Vice-President
tel;work:450-358-2266 x223
tel;cell:514-835-9497
x-mozilla-html:FALSE
url:http://www.intello.com
version:2.1
end:vcard
-------------------------------------------------------------------------
Take Surveys. Earn Cash. Influence the Future of IT
Join SourceForge.net's Techsay panel and you'll get the chance to share your
opinions on IT & business topics through brief surveys-and earn cash
http://www.techsay.com/default.php?page=join.php&p=sourceforge&CID=DEVDEV
_______________________________________________
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
