On Sun, Dec 16, 2001 at 06:08:02PM +0100, Olivier Bornet wrote: tnx a lot, i've try the patch , and after a little change it work fine :)
regards and tnx nextime > Hello, > > > On this machine i've apt-get installed snmpd from a woody source: > > > > 05:38:54 danex:~# snmpget <hostname> <community> ifOutOctets.2 > > interfaces.ifTable.ifEntry.ifOutOctets.2 = 4294967295 > > Try cat /proc/net/dev : you will see the counter is more high than this. > > I was having the same problem. This is a bug in snmpd and values bigger > than 2^32-1. To resolve my problem, I have get and installed snmpd from > the sources : http://www.net-snmp.org/, and also applied a patch from a > net-snmp developer. (this was for version 4.2.3) > > I have attached the original message of the developper of net-snmp > giving me the patch. Please note you need to edit a little the patch, > because the lines break are not correct. > > Hope this help. > > Olivier > -- > Olivier Bornet | français : http://puck.ch/f > Swiss Ice Hockey Results | english : http://puck.ch/e > http://puck.ch/ | deutsch : http://puck.ch/g > [EMAIL PROTECTED] | italiano : http://puck.ch/i > Get my PGP-key at http://puck.ch/pgp or at http://wwwkeys.pgp.net > X-Auth-No: > From: Niels Baggesen <[EMAIL PROTECTED]> > To: Olivier Bornet <[EMAIL PROTECTED]>, > [email protected] > Subject: Re: Problem with counter overflow > In-Reply-To: <[EMAIL PROTECTED]> > Organization: State and University Library > Errors-To: [EMAIL PROTECTED] > X-BeenThere: [email protected] > X-Mailman-Version: 2.0.5 > Precedence: bulk > X-Original-Date: Wed, 12 Dec 2001 19:30:50 +0100 > Date: Wed, 12 Dec 2001 19:30:50 +0100 > > On Wed, Dec 12, 2001 at 09:16:22AM +0100, Olivier Bornet wrote: > > Hello, > > > > > > interfaces.ifTable.ifEntry.ifInOctets.2 = Counter32: 1611864348 > > > > interfaces.ifTable.ifEntry.ifOutOctets.2 = Counter32: 4294967295 > > > > > > > > So, the In is OK, but the Out seem to limit to 2^32-1. As this is now > > > > since > > > > about a week... > > > > > > Hmmm, I thought we had gotten these ruled out. What does /proc/net/dev > > > look like? > > > > So, the /proc/net/dev contain (among other things) : > > eth0:1613699263 9019257 ... 4540832067 7894842 ... > > > > The values are good in /proc/net/dev > > > > > What kernel version is this? > > > > This is a 2.4.4 on SPARC. > > Ahhh, I should have spotted that you answered your question yourself: > counters are unsigned 32-bit, and > > $ bc > bc 1.06 > Copyright 1991-1994, 1997, 1998, 2000 Free Software Foundation, Inc. > This is free software with ABSOLUTELY NO WARRANTY. > For details type `warranty'. > obase=16 > 4540832067 > 10EA79943 > > so this kernel keeps the counter in a 64-bit, which means that when we > read it with a %lu format it is returned as 0xFFFFFFFF. We have to read > it into an unsigned long long and then discard the excess ourselves > (until we get the ifXTable in place) > > Maybe you could try this patch: > > --- agent/mibgroup/mibII/interfaces.c 2001/11/28 22:30:33 > 1.75.2.16 > +++ agent/mibgroup/mibII/interfaces.c 2001/12/12 18:29:37 > @@ -1266,11 +1266,11 @@ > struct ifreq ifrq; > struct ifnet **ifnetaddr_ptr; > FILE *devin; > - unsigned long rec_pkt, rec_oct, rec_err, snd_pkt, snd_oct, snd_err, > coll; > + unsigned long long rec_pkt, rec_oct, rec_err, snd_pkt, snd_oct, > snd_err, coll; > int i, fd; > conf_if_list *if_ptr; > - const char *scan_line_2_2="%lu %lu %lu %*lu %*lu %*lu %*lu %*lu %lu > %lu %lu %*lu %*lu %lu"; > - const char *scan_line_2_0="%lu %lu %*lu %*lu %*lu %lu %lu %*lu %*lu > %lu"; > + const char *scan_line_2_2="%llu %llu %llu %*llu %*llu %*llu %*llu > %*llu %llu %llu %llu %*llu %*llu %llu"; > + const char *scan_line_2_0="%llu %llu %*llu %*llu %*llu %llu %llu > %*llu %*llu %llu"; > const char *scan_line_to_use; > > #endif > > > /Niels > > -- > Niels Baggesen - @home - Århus - Denmark - [EMAIL PROTECTED] > -- All people smile in the same language -- > > _______________________________________________ > Net-snmp-coders mailing list > [email protected] > https://lists.sourceforge.net/lists/listinfo/net-snmp-coders >
