[c-nsp] 7507 GEIP Controller
Rodney Dunn
rodunn at cisco.com
Wed Jul 25 17:26:23 EDT 2007
On Thu, Jul 26, 2007 at 01:26:46AM +0500, Masood Ahmad Shah wrote:
> Thanks for your reply..I have not tried without cable, I will check it.
>
> Does cable effects even if the interface is shutdown?
For packet switching it shouldn't but if the driver is still sending
an interrupt it may.
> Why I can't see the process taking this high cpu usage while I do show proc
> cpu etcc.?
Because it's interrupt level work the CPU is doing. We could try
profiling the CPU and see what it says.
It may or may not help.
>
> Here is the output for show diag, show stacks and show align
>
> Router#show diag 5 ( this is GEIP controller slot )
> Slot 5:
> Physical slot 5, ~physical slot 0xA, logical slot 5, CBus 1
> Microcode Status 0x4
> Master Enable, LED, WCS Loaded
> Board is analyzed
> Pending I/O Status: None
> EEPROM format version 1
> GEIP controller, HW rev 2.02, board revision A0
> Serial number: 27588119 Part number: 73-2167-05
> Test history: 0x00 RMA number: 00-00-00
> Flags: cisco 7000 board; 7500 compatible
>
> EEPROM contents (hex):
> 0x20: 01 21 02 02 01 A4 F6 17 49 08 77 05 00 00 00 00
> 0x30: 50 00 00 01 00 00 00 00 00 00 00 00 00 00 00 00
>
> Slot database information:
> Flags: 0x4 Insertion time: 0x33C0 (4d00h ago)
>
> Controller Memory Size: 128 MBytes DRAM, 4096 KBytes SRAM
>
> PA Bay 0 Information:
> Gigabit-Ethernet PA, 1 ports
> EEPROM format version 1
> HW rev 1.01, Board revision A0
> Serial number: 29339093 Part number: 73-3144-05
>
> --Boot log begin--
>
> Cisco IOS Software, VIP Software (SVIP-DW-M), Version 12.4(3a), RELEASE
> SOFTWARE (fc2)
> Technical Support: http://www.cisco.com/techsupport
> Copyright (c) 1986-2005 by Cisco Systems, Inc.
> Compiled Fri 30-Sep-05 07:50 by hqluong
> Image text-base: 0x6001104C, data-base: 0x60720000
>
> Port Statistics for unclassified packets is not turned on.
>
> --Boot log end--
>
>
>
> Router# #show stacks
>
> Minimum process stacks:
> Free/Size Name
> 2340/3000 RSP memory size check
> 11280/12000 DHCPD Receive
> 5340/6000 Inspect Init Msg
> 59136/60000 script background loader
> 5224/6000 CDP Protocol
> 5028/6000 Clock Server
> 11004/12000 Router Init
> 3220/12000 Init
> 4184/6000 Update prst
> 4976/6000 DIB error message
> 5152/6000 RADIUS INITCONFIG
> 5188/6000 DRMI Master Reg. Slave Process
> 5136/6000 CEF Reloader
> 5312/6000 MDFS Reload
> 34844/36000 TCP Command
> 5332/6000 BGP Accepter
> 4892/6000 BGP Open
> 2276/3000 Rom Random Update Process
> 9236/12000 IPS SDF Loader
> 8220/12000 Virtual Exec
> 8220/12000 Exec
> 10884/12000 SSH Process
>
> Interrupt level stacks:
> Level Called Unused/Size Name
> 1 2213677378 6892/9000 Network Interrupt
> 2 5585183 7676/9000 Network Status Interrupt
> 3 0 8692/9000 OIR interrupt
> 4 0 9000/9000 PCMCIA Interrupt
> 5 171491 8596/9000 Console Uart
> 6 0 9000/9000 Error Interrupt
> 7 86692017 8604/9000 NMI Interrupt Handler
Get a couple of sh stacks and look at the interrupt level calls and
see which one is going up the most.
It may be a bad card.
>
> Router# #show align
>
> No alignment data has been recorded.
>
> No spurious memory references have been recorded.
>
>
> Regards,
> Masood Ahmad Shah
>
>
> -----Original Message-----
> From: Rodney Dunn [mailto:rodunn at cisco.com]
> Sent: Thursday, July 26, 2007 1:14 AM
> To: Masood Ahmad Shah
> Cc: cisco-nsp at puck.nether.net
> Subject: Re: [c-nsp] 7507 GEIP Controller
>
> What does 'sh diag' say and sh stack and show align.
>
> It's as if interrupts are getting sent to the RSP for some reason.
>
> Does the same thing happen if no cables are attached.
>
> 'sh diag'?
>
> On Thu, Jul 26, 2007 at 12:47:43AM +0500, Masood Ahmad Shah wrote:
> > I have strange issue with 7507. I have added a new GEIP controller into
> it.
> > By adding this GEIP controller the CPU usage has been increased up to 20%.
> > The strange thing is that CPU usage is always normal + 20% and there
> > is no traffic on this interface ( even if interface is shutdown the
> > usage remains same ). And if there is traffic it's same J
> >
> >
> >
> > I can't see the process taking this high CPU utilization if I look
> > into CPU usage stats.
> >
> >
> >
> > What the heck has happened to this? J
> >
> >
> >
> > Here are CPU stats:
> >
> >
> >
> > Router# #show proc cpu | excl 0.00%__0.00%__0.00%
> >
> > CPU utilization for five seconds: 15%/13%; one minute: 15%; five minutes:
> > 13%
> >
> > PID Runtime(ms) Invoked uSecs 5Sec 1Min 5Min TTY Process
> >
> > 5 518320 41974 12348 0.00% 0.11% 0.11% 0 Check heaps
> >
> >
> > 20 66004 924869 71 0.08% 0.01% 0.00% 0 IPC Seat
> > Manager
> >
> > 22 204088 1466063 139 0.00% 0.03% 0.02% 0 ARP Input
> >
> >
> > 31 7080 1609 4400 0.98% 1.44% 0.59% 2 Virtual
> Exec
> >
> >
> > 52 117672 5843 20138 0.00% 0.02% 0.00% 0 Per-minute
> > Jobs
> >
> > 73 3586004 17990635 199 0.82% 0.60% 0.49% 0 IP Input
> >
> >
> > 189 38964 213398 182 0.00% 0.01% 0.00% 0 IP SNMP
> >
> >
> > 191 56184 104952 535 0.00% 0.01% 0.00% 0 SNMP ENGINE
> >
> >
> > 196 160560 1397067 114 0.08% 0.01% 0.00% 0 BGP Router
> >
> >
> > 197 49200 250208 196 0.00% 0.01% 0.00% 0 BGP I/O
> >
> >
> >
> > IOS version:
> >
> >
> >
> > IOS Version 12.4(3a)
> >
> >
> >
> > Controller
> >
> > Router# sh ver | inc GEIP
> >
> >
> >
> > 1 GEIP controller (1 GigabitEthernet).
> >
> > 1 GEIP controller (1 GigabitEthernet)
> >
> >
> >
> >
> >
> > Regards,
> >
> > Masood Ahmad Shah
> >
> >
> >
> > _______________________________________________
> > cisco-nsp mailing list cisco-nsp at puck.nether.net
> > https://puck.nether.net/mailman/listinfo/cisco-nsp
> > archive at http://puck.nether.net/pipermail/cisco-nsp/
More information about the cisco-nsp
mailing list