- mice-discuss - lists.micemn.net

Re: Cacti Graph using SNMPv1?

by Steve Howard

On 05/25/2011 02:15 PM, Mike Horwath wrote: > On 5/25/11 11:39 AM, "Andrew Hoyos" <hoyosa(a)GMAIL.COM> wrote: > >> On May 25, 2011, at 11:33 AM, David Farmer wrote: >> >>> On 5/25/11 11:07 CDT, Jay Hanke wrote: >>>>> Did you check that the Data Source for the graph is also using the >>>>> 64-bit counters? >>>>> >>>>>> Nothing special in logs either showing a problem. >>>>>> >>>>>> Steve - your own monitoring showing problems at all with data >>>> Are you looking at Gi-0/2? >>> it is possible this is a problem on the Cisco side too, that port >>> Gi-0/2 on the 3560E got changed to being in a port channel sometime this >>> creates weird issues. They usually go away after a reboot. >> Yeah, I've seen this same thing on a few occasions when adding/removing >> interfaces from port channels. >> >> Also, double check you've got "snmp-server ifindex persist" set in the >> config, as you might be polling something else too. > Shit, that's a great point! > > We aways have that set. > > Steve, can you check that? > > Could very well be a different port now :( > > But once you set it to persist (if it wasn't set), I can look into things > further. > It was not there. It is now. ######################################################################## To unsubscribe from the MICE-DISCUSS list, click the following link: http://lists.iphouse.net/cgi-bin/wa?SUBED1=MICE-DISCUSS&A=1

May 25, 2011

Re: Cacti Graph using SNMPv1?

by Mike Horwath

On 5/25/11 11:39 AM, "Andrew Hoyos" <hoyosa(a)GMAIL.COM> wrote: >On May 25, 2011, at 11:33 AM, David Farmer wrote: > >> On 5/25/11 11:07 CDT, Jay Hanke wrote: >>>> Did you check that the Data Source for the graph is also using the >>>> 64-bit counters? >>>> >>>>> >>>>> Nothing special in logs either showing a problem. >>>>> >>>>> Steve - your own monitoring showing problems at all with data >>> >>> Are you looking at Gi-0/2? >> >> it is possible this is a problem on the Cisco side too, that port >>Gi-0/2 on the 3560E got changed to being in a port channel sometime this >>creates weird issues. They usually go away after a reboot. > >Yeah, I've seen this same thing on a few occasions when adding/removing >interfaces from port channels. > >Also, double check you've got "snmp-server ifindex persist" set in the >config, as you might be polling something else too. Shit, that's a great point! We aways have that set. Steve, can you check that? Could very well be a different port now :( But once you set it to persist (if it wasn't set), I can look into things further. -- Mike Horwath ipHouse - Welcome home! drechsau(a)iphouse.net The universe is an island, surrounded by whatever it is that surrounds universes. - Berkeley Fortune ######################################################################## To unsubscribe from the MICE-DISCUSS list, click the following link: http://lists.iphouse.net/cgi-bin/wa?SUBED1=MICE-DISCUSS&A=1

May 25, 2011

Re: Cacti Graph using SNMPv1?

by Mike Horwath

On 5/25/11 11:01 AM, "Justin Krejci" <jkrejci(a)USINTERNET.COM> wrote: >On Wed, 2011-05-25 at 15:53 +0000, Mike Horwath wrote: >> On 5/24/11 6:28 PM, "Justin Krejci" <jkrejci(a)USINTERNET.COM> wrote: >> >> > >> >'micemn-gateway-2.micemn.net - Traffic - Gi0/2' in Cacti is exhibiting >> >signs of using SNMPv1 counters (frequent gaps at higher bandwidth >>usage) >> > >> >Is this a known issue? If not I presume someone at iphouse can fix this >> >easily. >> > >> >The "Fix64bit" plugin for Cacti is great for correcting this condition >>on >> >the fly if my assumption on the problem is correct. >> >> Your assumption is wrong unfortunately. >> >> I wish I had made a mistake, but stats.micemn.net's Cacti config is >>64bit >> for gathering of stuff. >> >> And using SNMPv2 according to the configuration. > >Did you check that the Data Source for the graph is also using the >64-bit counters? Verified! -- Mike Horwath ipHouse - Welcome home! drechsau(a)iphouse.net The universe is an island, surrounded by whatever it is that surrounds universes. - Berkeley Fortune ######################################################################## To unsubscribe from the MICE-DISCUSS list, click the following link: http://lists.iphouse.net/cgi-bin/wa?SUBED1=MICE-DISCUSS&A=1

May 25, 2011

Re: New Member's Configured on BIRD

by Mike Horwath

On 5/25/11 11:46 AM, "James Montz" <jamesmontz(a)MANKATONETWORKS.NET> wrote: >Added IPv4 and IPv6 Configuration for City of Windomw (ASN: 46692). >Welcome! >Also noticed that IPv6 config for UMN (ASN: 57) was not in, so added that >as well. > >Maybe its a matter of semantics, but the website lists Northern Lights >GigaPoP as ASN: 57, NOT UMN. Then fix0r! You has the power! -- Mike Horwath ipHouse - Welcome home! drechsau(a)iphouse.net The universe is an island, surrounded by whatever it is that surrounds universes. - Berkeley Fortune ######################################################################## To unsubscribe from the MICE-DISCUSS list, click the following link: http://lists.iphouse.net/cgi-bin/wa?SUBED1=MICE-DISCUSS&A=1

May 25, 2011

New Member's Configured on BIRD

by James Montz

Added IPv4 and IPv6 Configuration for City of Windomw (ASN: 46692). Welcome! Also noticed that IPv6 config for UMN (ASN: 57) was not in, so added that as well. Maybe its a matter of semantics, but the website lists Northern Lights GigaPoP as ASN: 57, NOT UMN. --------------------------------------------------- *James C. Montz* Senior Consultant *Mankato Networks LLC* <http://www.mankatonetworks.net/>320 East Liberty Street <http://maps.google.com/maps?q=320+East+Liberty+Street%2CMankato%2CMN+56001&…>Mankato, MN 56001 *Phone:* *(320)-496-6689* *Email: jamesmontz(a)mankatonetworks.net* *Web: http://www.mankatonetworks.net Skype: james.montz* * http://www.linkedin.com/in/jmontz* ######################################################################## To unsubscribe from the MICE-DISCUSS list, click the following link: http://lists.iphouse.net/cgi-bin/wa?SUBED1=MICE-DISCUSS&A=1

May 25, 2011

Re: Cacti Graph using SNMPv1?

by Andrew Hoyos

On May 25, 2011, at 11:33 AM, David Farmer wrote: > On 5/25/11 11:07 CDT, Jay Hanke wrote: >>> Did you check that the Data Source for the graph is also using the >>> 64-bit counters? >>> >>>> >>>> Nothing special in logs either showing a problem. >>>> >>>> Steve - your own monitoring showing problems at all with data >> >> Are you looking at Gi-0/2? > > it is possible this is a problem on the Cisco side too, that port Gi-0/2 on the 3560E got changed to being in a port channel sometime this creates weird issues. They usually go away after a reboot. Yeah, I've seen this same thing on a few occasions when adding/removing interfaces from port channels. Also, double check you've got "snmp-server ifindex persist" set in the config, as you might be polling something else too. ######################################################################## To unsubscribe from the MICE-DISCUSS list, click the following link: http://lists.iphouse.net/cgi-bin/wa?SUBED1=MICE-DISCUSS&A=1

May 25, 2011

Re: Cacti Graph using SNMPv1?

by David Farmer

On 5/25/11 11:07 CDT, Jay Hanke wrote: >> Did you check that the Data Source for the graph is also using the >> 64-bit counters? >> >>> >>> Nothing special in logs either showing a problem. >>> >>> Steve - your own monitoring showing problems at all with data > > Are you looking at Gi-0/2? it is possible this is a problem on the Cisco side too, that port Gi-0/2 on the 3560E got changed to being in a port channel sometime this creates weird issues. They usually go away after a reboot. > Usually SNMPv1 issues will appear at right about 100 Mb/s on a five > minute average. We're pulling the po1 interface well over a Gig > without an issue. Also, the traffic will show up with spike and then > nothing (when the counter rolls over). > > Jay > > ######################################################################## > > To unsubscribe from the MICE-DISCUSS list, click the following link: > http://lists.iphouse.net/cgi-bin/wa?SUBED1=MICE-DISCUSS&A=1 -- =============================================== David Farmer Email:farmer@umn.edu Networking & Telecommunication Services Office of Information Technology University of Minnesota 2218 University Ave SE Phone: 612-626-0815 Minneapolis, MN 55414-3029 Cell: 612-812-9952 =============================================== ######################################################################## To unsubscribe from the MICE-DISCUSS list, click the following link: http://lists.iphouse.net/cgi-bin/wa?SUBED1=MICE-DISCUSS&A=1

May 25, 2011

Re: Cacti Graph using SNMPv1?

by Steve Howard

On 05/25/2011 10:53 AM, Mike Horwath wrote: > Steve - your own monitoring showing problems at all with data I haven't got around to setting up the monitoring. Yours has worked great so I didn't bother (yet). Steve ######################################################################## To unsubscribe from the MICE-DISCUSS list, click the following link: http://lists.iphouse.net/cgi-bin/wa?SUBED1=MICE-DISCUSS&A=1

May 25, 2011

Re: Cacti Graph using SNMPv1?

by Owen DeLong

On May 25, 2011, at 9:01 AM, Justin Krejci wrote: > On Wed, 2011-05-25 at 15:53 +0000, Mike Horwath wrote: >> On 5/24/11 6:28 PM, "Justin Krejci" <jkrejci(a)USINTERNET.COM> wrote: >> >>> >>> 'micemn-gateway-2.micemn.net - Traffic - Gi0/2' in Cacti is exhibiting >>> signs of using SNMPv1 counters (frequent gaps at higher bandwidth usage) >>> >>> Is this a known issue? If not I presume someone at iphouse can fix this >>> easily. >>> >>> The "Fix64bit" plugin for Cacti is great for correcting this condition on >>> the fly if my assumption on the problem is correct. >> >> Your assumption is wrong unfortunately. >> >> I wish I had made a mistake, but stats.micemn.net's Cacti config is 64bit >> for gathering of stuff. >> >> And using SNMPv2 according to the configuration. > > Did you check that the Data Source for the graph is also using the > 64-bit counters? > This would be the question of ifInOctets (32 bit) vs. ifInHCOctets (64 bit) IIRC. In case the reference is helpful. Owen >> >> Nothing special in logs either showing a problem. >> >> Steve - your own monitoring showing problems at all with data? >> > > ######################################################################## > > To unsubscribe from the MICE-DISCUSS list, click the following link: > http://lists.iphouse.net/cgi-bin/wa?SUBED1=MICE-DISCUSS&A=1 ######################################################################## To unsubscribe from the MICE-DISCUSS list, click the following link: http://lists.iphouse.net/cgi-bin/wa?SUBED1=MICE-DISCUSS&A=1

May 25, 2011

Re: Cacti Graph using SNMPv1?

by Jay Hanke

> Did you check that the Data Source for the graph is also using the > 64-bit counters? > >> >> Nothing special in logs either showing a problem. >> >> Steve - your own monitoring showing problems at all with data Are you looking at Gi-0/2? Usually SNMPv1 issues will appear at right about 100 Mb/s on a five minute average. We're pulling the po1 interface well over a Gig without an issue. Also, the traffic will show up with spike and then nothing (when the counter rolls over). Jay ######################################################################## To unsubscribe from the MICE-DISCUSS list, click the following link: http://lists.iphouse.net/cgi-bin/wa?SUBED1=MICE-DISCUSS&A=1

May 25, 2011