Thanks Tony. Make note of the overal pps going through the box pre/post upgrade.
sh int | incl packets sh proc cpu sort | excl 0.00 Rodney On Tue, Jul 31, 2007 at 01:36:12AM -0700, Tony Varriale wrote: > We have one G1 that was just implemented with 15T1. We also have 2 running > 11T and 1 running 9T. The 2 running 11T have, IMO, very high CPU (60+% > average durning middle of day). The 9T barely runs 20%. > > I am planning on upgrading one of the 11Ts to see if the CPU issue was > resolved. I will post up when I get some data. > > tv > ----- Original Message ----- > From: "Rodney Dunn" <[EMAIL PROTECTED]> > To: "Rafi Sadowsky" <[EMAIL PROTECTED]> > Cc: <[email protected]>; "Hank Nussbacher" <[EMAIL PROTECTED]> > Sent: Monday, July 30, 2007 8:15 AM > Subject: Re: [c-nsp] 12.4(11)T QoS problems (again) > > > > On Sat, Jul 28, 2007 at 07:18:24AM +0300, Rafi Sadowsky wrote: > >> Hi Rodney, > >> > >> FYI: The router is a 7206VXR NPE-G1 with most of traffic between 2 of > >> the > >> NPE GbE interfaces > > > > Ok. Good because the CPU change if there is one significant would > > show up between the two versions. > > > >> > >> Any particular reason you mentioned a NPE-G2 ? > > > > Search the archives for the NPE-G2 measurement discussions > > we've had. > > > > Rodney > > > >> > >> > >> -- > >> Regards > >> Rafi > >> > >> [sh ver] > >> Cisco IOS Software, 7200 Software (C7200-ADVIPSERVICESK9-M), Version > >> 12.4(4)T1, RELEASE SOFTWARE (fc4) > >> Technical Support: http://www.cisco.com/techsupport > >> Copyright (c) 1986-2005 by Cisco Systems, Inc. > >> Compiled Wed 21-Dec-05 22:58 by ccai > >> > >> ROM: System Bootstrap, Version 12.3(4r)T3, RELEASE SOFTWARE (fc1) > >> BOOTLDR: 7200 Software (C7200-KBOOT-M), Version 12.3(15), RELEASE > >> SOFTWARE > >> (fc3) > >> > >> Tel-Aviv uptime is 1 week, 5 days, 1 minute > >> System returned to ROM by reload at 07:07:17 IDT Mon Jul 16 2007 > >> System restarted at 07:09:50 IDT Mon Jul 16 2007 > >> System image file is "disk2:c7200-advipservicesk9-mz.124-4.T1.bin" > >> Last reload reason: Reload Command > >> > >> Cisco 7206VXR (NPE-G1) processor (revision B) with 491520K/32768K bytes > >> of > >> memory. > >> Processor board ID 26812591 > >> SB-1 CPU at 700MHz, Implementation 1025, Rev 0.2, 512KB L2 Cache > >> 6 slot VXR midplane, Version 2.6 > >> > >> Last reset from power-on > >> > >> [snipped] > >> > >> PCI bus mb1 (Slots 1, 3 and 5) has a capacity of 600 bandwidth points. > >> [snipped] > >> PCI bus mb2 (Slots 2, 4 and 6) has a capacity of 600 bandwidth points. > >> [snipped] > >> > >> > >> 4 Ethernet interfaces > >> 2 FastEthernet interfaces > >> 3 Gigabit Ethernet interfaces > >> 8 Serial interfaces > >> 1 Packet over SONET interface > >> 509K bytes of NVRAM. > >> > >> 125952K bytes of ATA PCMCIA card at slot 2 (Sector size 512 bytes). > >> 16384K bytes of Flash internal SIMM (Sector size 256K). > >> Configuration register is 0x2102 > >> > >> -- > >> Regards, > >> Rafi > >> > >> On Fri, 27 Jul 2007, Rodney Dunn wrote: > >> > >> >Date: Fri, 27 Jul 2007 08:52:26 -0400 > >> >From: Rodney Dunn <[EMAIL PROTECTED]> > >> >To: Hank Nussbacher <[EMAIL PROTECTED]> > >> >Cc: [email protected], [EMAIL PROTECTED] > >> >Subject: Re: [c-nsp] 12.4(11)T QoS problems (again) > >> > > >> >I'd be interested to know this as well. I helped draft the FN on this > >> >one to make sure they were clear. > >> > > >> >In the fixes that went in 15T1 there was really only one, rather large, > >> >change to improve the dataplane performance. Others were mostly bug > >> >functionality problems. > >> > > >> >The NPE-G2 performance comparisons are harder given our previous > >> >discussions around how it does CPU measurements. > >> > > >> >Rodney > >> > > >> > > >> >On Fri, Jul 27, 2007 at 07:10:06AM +0300, Hank Nussbacher wrote: > >> >>Cisco Field Notice: Field Notice: FN - 62854 - Cisco IOS - Performance > >> >>Degradation and Reloads Possible When Using Quality of Service Features > >> >>in > >> >>Version 12.4(11)T - July 26, 2007 > >> >> > >> >>http://www.cisco.com/en/US/customer/products/ps6441/products_field_notice09186a008088cc2f.shtml > >> >> > >> >>"Workaround/Solution > >> >>There is no workaround that avoids all possible symptoms. A complete > >> >>solution is expected to be available in Cisco IOS version 12.4(15)T1. > >> >>To > >> >>avoid the symptoms, users may downgrade to a pre-12.4(11)T release or > >> >>upgrade to 12.4(15)T1 when it becomes available. Cisco strongly > >> >>discourages deploying an IOS 12.4(11)T release with QoS features > >> >>enabled > >> >>due to the potential severity of symptoms." > >> >> > >> >>We went to 12.4(15)T around the beginning of July (it is now a Deferred > >> >>Release) and it did *not* help so we went back to 12.4(4)T1 which does > >> >>fix > >> >>the performance problems. One can see the exact change dates via this > >> >>MRTG CPU graph: http://noc.ilan.net.il/stats/ILAN-CPU/tau-7206-cpu.html > >> >> > >> >>I see 12.4(15)T1 came out July 20. Has anyone tested it (non-Cisco) > >> >>and > >> >>found that it does solve all performance problems? > >> >> > >> >>-Hank > >> >> > >> >>_______________________________________________ > >> >>cisco-nsp mailing list [email protected] > >> >>https://puck.nether.net/mailman/listinfo/cisco-nsp > >> >>archive at http://puck.nether.net/pipermail/cisco-nsp/ > >> > > > _______________________________________________ > > cisco-nsp mailing list [email protected] > > https://puck.nether.net/mailman/listinfo/cisco-nsp > > archive at http://puck.nether.net/pipermail/cisco-nsp/ > > _______________________________________________ > cisco-nsp mailing list [email protected] > https://puck.nether.net/mailman/listinfo/cisco-nsp > archive at http://puck.nether.net/pipermail/cisco-nsp/ _______________________________________________ cisco-nsp mailing list [email protected] https://puck.nether.net/mailman/listinfo/cisco-nsp archive at http://puck.nether.net/pipermail/cisco-nsp/
