04-30-2009 07:43 AM - edited 03-04-2019 04:35 AM
Hi,
we are currently using the router 7206 VXR NPEG1 which is going to HIGH CPU Utllization at some times so now we planning to upgrade the existing router to NPEG2
this will make change in the handling CPU utlization or the same as the existing hardware
as per my understanding the packet handling capacity is double from the existing router
i have given the traffic flow in one of our interface
GigabitEthernet0/3 is up, line protocol is up
Internet address is xx.xx.xx.17/24
MTU 1500 bytes, BW 1000000 Kbit, DLY 10 usec,
reliability 255/255, txload 57/255, rxload 71/255
Encapsulation ARPA, loopback not set
Keepalive set (10 sec)
Full-duplex, 1000Mb/s, media type is RJ45
output flow-control is XON, input flow-control is XON
ARP type: ARPA, ARP Timeout 04:00:00
Last input 00:00:00, output 00:00:00, output hang never
Last clearing of "show interface" counters 27w3d
Input queue: 0/75/15679/924143 (size/max/drops/flushes); Total output drops: 3
9315
Queueing strategy: fifo
Output queue: 0/40 (size/max)
5 minute input rate 280168000 bits/sec, 384472 packets/sec
5 minute output rate 223970000 bits/sec, 311684 packets/sec
1405404492 packets input, 3683894657 bytes, 670 no buffer
Received 0 broadcasts, 0 runts, 0 giants, 10968 throttles
0 input errors, 0 CRC, 0 frame, 13275561 overrun, 0 ignored
0 watchdog, 55464 multicast, 0 pause input
0 input packets with dribble condition detected
3168067588 packets output, 3916374664 bytes, 0 underruns
0 output errors, 0 collisions, 0 interface resets
0 babbles, 0 late collision, 0 deferred
1 lost carrier, 0 no carrier, 0 pause output
0 output buffer failures, 0 output buffers swapped out
04-30-2009 10:00 AM
In theory, yes the G2 should provide 2x the forwarding performance of the G1. However, before jumping into a new faster processor, you might verify your existing configuration supports optimal performance and/or your high CPU is interrupt CPU.
04-30-2009 03:08 PM
Hi,
Thanks for your reply
you mean to say some configuration in our existing router is making CPU utlization high
we are using all 5 interface in the existing router asper the traffic flow analysis
1. including all 5 interface the traffic rate will be around 450 mb
which factor will increase the CPU utlization
please find small config
interface GigabitEthernet0/1
ip address 192.168.191.1 255.255.255.0
duplex auto
speed auto
media-type rj45
no negotiation auto
!
interface GigabitEthernet0/2
ip address 192.168.190.1 255.255.255.0
duplex auto
speed auto
media-type rj45
no negotiation auto
!
interface GigabitEthernet0/3
ip address xx.xx.xx.17 255.255.255.0
duplex auto
speed auto
media-type rj45
no negotiation auto
!
interface FastEthernet2/0
ip address xx.xx.35.17 255.255.255.0
no ip redirects
no ip proxy-arp
duplex auto
speed auto
no cdp enable
!
interface FastEthernet2/1
ip address xx.xx.xx.42 255.255.255.252
no ip redirects
no ip proxy-arp
duplex full
speed 100
no cdp enable
!
router bgp xxxxx
no synchronization
bgp log-neighbor-changes
network xx.xxx.0.0 mask 255.255.248.0
network xx.xx.xx.0
redistribute connected
neighbor xx.xx.xx.45 remote-as 41847
neighbor xx.xx.xx.45 description NEWER STEALTH 21032007
neighbor xx.xx.xx.45 shutdown
neighbor xx.xx.xx.45 next-hop-self
neighbor xx.xx.xx.45 send-community
neighbor xx.xx.xx.45 soft-reconfiguration inbound
neighbor xx.xx.xx.45 distribute-list 56 in
neighbor xx.xx.xx.45 prefix-list MY_ROUTES out
neighbor xx.xx.xx.45 route-map STEALTH-IN in
neighbor xx.xx.xx.45 route-map STEALTH-OUT out
neighbor xx.xx.xx.41 remote-as 8757
neighbor xx.xx.xx.41 description NEW STEALTH
neighbor xx.xx.xx.41 shutdown
neighbor xx.xx.xx.41 next-hop-self
neighbor xx.xx.xx.41 send-community
neighbor xx.xx.xx.41 soft-reconfiguration inbound
neighbor xx.xx.xx.41 distribute-list 56 in
neighbor xx.xx.xx.41 prefix-list MY_ROUTES out
neighbor xx.xx.xx.41 route-map STEALTH-IN in
neighbor xx.xx.xx.41 route-map STEALTH-OUT out
neighbor xx.xx.xx.252 remote-as 15830
neighbor xx.xx.xx.252 description NEW TELECITY
neighbor xx.xx.xx.252 next-hop-self
neighbor xx.xx.xx.252 send-community
neighbor xx.xx.xx.252 soft-reconfiguration inbound
neighbor xx.xx.xx.252 prefix-list MY_ROUTES out
neighbor xx.xx.xx.252 route-map TELECITY-IN in
neighbor xx.xx.xx.252 route-map TELECITY-OUT out
neighbor xx.xx.xx.251 remote-as 15830
neighbor xx.xx.xx.251 next-hop-self
neighbor xx.xx.xx.251 send-community
neighbor xx.xx.xx.251 soft-reconfiguration inbound
neighbor xx.xx.xx.251 prefix-list MY_ROUTES out
neighbor xx.xx.xx.251 route-map TELECITYBAK-IN in
neighbor xx.xx.xx.251 route-map TELECITYBAK-OUT out
no auto-summary
!
ip default-gateway xx.xx.xx.252
ip classless
ip route 10.39.0.0 255.255.0.0 192.168.190.2
!
!
!
ip prefix-list MY_ROUTES seq 5 permit xx.xx.0.0/21
ip prefix-list MY_ROUTES seq 10 permit xx.xx.141.0/24
!
04-30-2009 03:47 PM
Some factors to check for high CPU can be found here: http://www.cisco.com/en/US/products/hw/routers/ps133/products_tech_note09186a00800a70f2.shtml
However, if you have about 450 Mbps, the traffic alone could run your G1 high.
04-30-2009 07:52 PM
Hello,
If you issue a "sh ip bgp summ" on your router, what are the total routes going into your router?
04-30-2009 08:09 PM
hi,
i attached the show command details
#sh ip bgp summary
BGP router identifier xx.xx.xx.17, local AS number xxx404
BGP table version is 1442283, main routing table version 1442283
285671 network entries using 34280520 bytes of memory
571332 path entries using 29709264 bytes of memory
96870/48424 BGP path/bestpath attribute entries using 12011880 bytes of memor
88246 BGP AS-PATH entries using 2405928 bytes of memory
1 BGP community entries using 24 bytes of memory
0 BGP route-map cache entries using 0 bytes of memory
0 BGP filter-list cache entries using 0 bytes of memory
Bitfield cache entries: current 1 (at peak 2) using 32 bytes of memory
BGP using 78407648 total bytes of memory
285661 received paths for inbound soft reconfiguration
BGP activity xx8168/xx2493 prefixes, 1716814/1145482 paths, scan interval 60
s
Neighbor V AS MsgRcvd MsgSent TblVer InQ OutQ Up/Down State/Pfx
xxx.xx.170.45 4 xx847 0 0 0 0 0 never Idle (Adm
xxx.xx.129.41 4 xx757 0 0 0 0 0 never Idle (Adm
xxx.xx.34.252 4 xx830 155002 1392 1442296 0 0 03:26:42 xx5661
xxx.xx.35.251 4 xx830 0 0 0 0 0 never Active
#sh proc cpu
CPU utilization for five seconds: 49%/49%; one minute: 50%; five minutes: 49%
PID Runtime(ms) Invoked uSecs 5Sec 1Min 5Min TTY Process
05-01-2009 12:58 AM
Hello Vinoth,
to see what processes use more of cpu resources you can use
sh proc cpu sorted 1min
However, from what you have attached you can see that
CPU utilization for five seconds: 49%/49%; one minute: 50%; five minutes: 49%
all cpu usage at 5 seconds is caused by cpu interrupts (second number after / is 49 as the one before).
This is likely caused by packet forwarding activity.
You can verify if the most efficient forwarding method CEF is in use
see
Be aware that there have been other threads of people that have reported that after having moved from NPE-G1 to NPE-G2 the cpu usage was reduced but less then how much expected.
More specifically port adapters look like to require more resources then built-in GE ports.
Hope to help
Giuseppe
05-01-2009 03:48 AM
Agree with Giuseppe. Both about high CPU most likely caused by packet forwarding, and G2 doesn't always seem to offer the performance boost expected (reason for my "in theory" within my original post).
However, now seeing that it looks like this router might have full Internet BGP routes, and considering your remark about "HIGH CPU Utllization at some times" in your OP, I'm wondering whether you're also seeing 100% spikes about once a minute that are not interrupt CPU, and/or input queue drops. If so, that's likely to be the BGP scanner.
05-01-2009 04:34 AM
Hi,
Thanks for your reply
Now we replaced the router to NPEG2 and also facing an same issue that cpu going high
# sh proc cpu sorted 1min
CPU utilization for five seconds: 92%/90%; one minute: 91%; five minutes: 90%
#sh interfaces
GigabitEthernet0/1 is up, line protocol is up
5 minute input rate 177982000 bits/sec, 247547 packets/sec
5 minute output rate 224304000 bits/sec, 303500 packets/sec
3276901558 packets input, 1599562081 bytes, 0 no buffer
Received 5804 broadcasts, 0 runts, 0 giants, 0 throttles
4 input errors, 0 CRC, 0 frame, 0 overrun, 4 ignored
0 watchdog, 10921 multicast, 0 pause input
0 input packets with dribble condition detected
4006620133 packets output, 3049708805 bytes, 0 underruns
0 output errors, 0 collisions, 0 interface resets
5 unknown protocol drops
39 unknown protocol drops
FastEthernet0/2 is administratively down, line protocol is down
5 minute input rate 0 bits/sec, 0 packets/sec
5 minute output rate 0 bits/sec, 0 packets/sec
0 packets input, 0 bytes
Received 0 broadcasts, 0 runts, 0 giants, 0 throttles
0 input errors, 0 CRC, 0 frame, 0 overrun, 0 ignored
0 watchdog
0 input packets with dribble condition detected
0 packets output, 0 bytes, 0 underruns
0 output errors, 0 collisions, 0 interface resets
1 unknown protocol drops
7 unknown protocol drops
GigabitEthernet0/3 is up, line protocol is up
)
5 minute input rate 260050000 bits/sec, 357435 packets/sec
5 minute output rate 209131000 bits/sec, 293521 packets/sec
523524606 packets input, 2153785717 bytes, 182 no buffer
Received 19954 broadcasts, 0 runts, 0 giants, 0 throttles
7 input errors, 0 CRC, 0 frame, 0 overrun, 7 ignored
0 watchdog, 50 multicast, 0 pause input
0 input packets with dribble condition detected
3957494965 packets output, 3670444482 bytes, 0 underruns
0 output errors, 0 collisions, 0 interface resets
47 unknown protocol drops
474 unknown protocol drops
FastEthernet3/1 is up, line protocol is up
5 minute input rate 29593000 bits/sec, 45253 packets/sec
5 minute output rate 34373000 bits/sec, 53008 packets/sec
670566917 packets input, 2892638133 bytes
Received 2223 broadcasts, 0 runts, 0 giants, 0 throttles
0 input errors, 0 CRC, 0 frame, 0 overrun, 0 ignored
0 watchdog
0 input packets with dribble condition detected
798594375 packets output, 7697882 bytes, 0 underruns
0 output errors, 0 collisions, 0 interface resets
3 unknown protocol drops
23 unknown protocol drops
and also i have seen some unknown protocol drops what it means
Can i know how much packet traffic can handle by this router NPEG2 since we are facing an issue regulary
suggest us some solution
# sh version
Cisco IOS Software, 7200 Software (C7200P-SPSERVICESK9-M), Version 12.4(15)T8, R
ROM: System Bootstrap, Version 12.4(12.2r)T, RELEASE SOFTWARE (fc1)
BOOTLDR: Cisco IOS Software, 7200 Software (C7200P-K91P-M), Version 12.2(31)SB10
, RELEASE SOFTWARE (fc2)
Cisco 7206VXR (NPE-G2) processor (revision A) with 917504K/65536K bytes of memor
y.
Processor board ID 21268035
MPC7448 CPU at 1666Mhz, Implementation 0, Rev 2.2
6 slot VXR midplane, Version 2.0
3 FastEthernet interfaces
3 Gigabit Ethernet interfaces
2045K bytes of NVRAM.
250880K bytes of ATA PCMCIA card at slot 2 (Sector size 512 bytes).
65536K bytes of Flash internal SIMM (Sector size 512K).
Configuration register is 0x2102
05-01-2009 04:34 PM
"Now we replaced the router to NPEG2 and also facing an same issue that cpu going high "
Assuming the interrupt CPU is still most of the CPU consumed, and assuming you've checked against Cisco suggestions for high CPU usage and optimal configuration, then you just might still have more load than the G2 can carry. I.e. a faster platform is needed.
"and also i have seen some unknown protocol drops what it means "
Might be non-IP packets, but considering the very, very small count vs. your total traffic, might not be worth worrying about.
PS:
The only other "engine" that might offer you more performance with your 7206VXR, would be the discontinued NSE-1. Besides being discontined, it might only 256 MB.
05-01-2009 11:32 PM
Hi,
I seen that BGP scanner is taking high amount of CPU
CPU utilization for five seconds: 79%/75%; one minute: 76%; five minutes: 75%
PID Runtime(ms) Invoked uSecs 5Sec 1Min 5Min TTY Process
194 2895768 11181 258990 3.61% 0.53% 0.45% 0 BGP Scanner
74 8128 664655 12 0.07% 0.08% 0.02% 2 Virtual Exec
68 346736 507309 683 0.07% 0.05% 0.06% 0 IP Input
2 16840 28017 601 0.00% 0.02% 0.00% 0 Load Meter
3 1052 3733 281 0.00% 0.00% 0.00% 0 BGP Open
5 3061640 155746 19657 0.00% 0.42% 0.49% 0 Check heaps
13 232 138466 1 0.00% 0.00% 0.00% 0 IPC Deferred Por
16 18128 2483827 7 0.00% 0.01% 0.00% 0 EnvMon
10 56 2325 24 0.00% 0.00% 0.00% 0 IPC Dynamic Cach
18 0 1 0 0.00% 0.00% 0.00% 0 Crash writer
12 356 138466 2 0.00% 0.00% 0.00% 0 IPC Periodic Tim
20 9392 144200 65 0.00% 0.00% 0.00% 0 ARP Background
33 536 139099 3 0.00% 0.00% 0.00% 0 GraphIt
19 65528 121370 539 0.00% 0.00% 0.00% 0 ARP Input
40 49756 43371 1147 0.00% 0.01% 0.00% 0 Net Background
43 700 138460 5 0.00% 0.00% 0.00% 0 TTY Background
44 8536 139098 61 0.00% 0.00% 0.00% 0 Per-Second Jobs
63 2476 4187671 0 0.00% 0.02% 0.02% 0 ACCT Periodic Pr
64 11332 39157 289 0.00% 0.00% 0.00% 0 CDP Protocol
84 678104 2838 238937 0.00% 0.06% 0.05% 0 IP Background
85 25332 9372 2702 0.00% 0.00% 0.00% 0 IP RIB Update
86 12460 202602 61 0.00% 0.01% 0.00% 0 CEF process
163 400 138389 2 0.00% 0.00% 0.00% 0 trunk conditioni
179 120 980 122 0.00% 0.00% 0.00% 0 Net Input
180 460 28017 16 0.00% 0.00% 0.00% 0 Compute load avg
181 54392 2386 22796 0.00% 0.00% 0.00% 0 Per-minute Jobs
182 28 1169 23 0.00% 0.00% 0.00% 0 CC-API_VCM
183 2688 1365784 1 0.00% 0.01% 0.00% 0 CCPROXY_CT
187 63092 2394 26354 0.00% 0.00% 0.00% 0 Shelf Hello Proc
190 1044 140281 7 0.00% 0.00% 0.00% 0 NTP
191 40 2330 17 0.00% 0.00% 0.00% 0 DHCPD Database
192 112964 333488 338 0.00% 0.00% 0.00% 0 BGP Router
193 11548 36066 320 0.00% 0.00% 0.00% 0 BGP I/O
195 212 3 70666 0.00% 0.00% 0.00% 0 BGP Event
Can any one help on this
05-02-2009 03:59 AM
I mentioned the BGP scanner in my 3rd post.
There's not much you can do about its CPU consumption, but further explanation and how to mitigate some of its effects can be found here: http://www.cisco.com/en/US/tech/tk365/technologies_tech_note09186a00809d16f0.shtml
PS:
The BGP scanner will commonly spike your CPU, but this doesn't tend to drive the long term CPU average high. In your case, you might have a long term high interrupt CPU. Again, if this is due to just traffic volume, you'll need a better performing device.
05-03-2009 10:51 PM
Thanks all for your suggestion
can anyone suggest us For traffic forwaring upto 1Gig which router will handle this much of traffic
Provided we will go for the upgraded platform
05-04-2009 02:33 AM
On the 7200, the discontinued NSE-1 might. Next up would be a 7304 with NSE-150, then the ASR series.
05-06-2009 05:06 AM
as per cisoc specfication the CPU processor concern its double from the earlier version
NPEG1 - Processor
700 MHz Broadcom BCM1250 processor.
NPEG2 - Processor
1.67-GHz Motorola Freescale 7448 processor
But still after upgrading to the NPEg2 also the cpu utlization is same
Suggest me the solution
The traffic handlig capacity per interface is
G0/1 - 153MB
G0/2 - 35MB
G0/3 - 225MB
Fa 3/1 - 2 MB
IP cef
router bgp xx04
no synchronization
bgp log-neighbor-changes
network xx.xx.0.0 mask 255.255.248.0
network xxx.xx.141.0
redistribute connected
neighbor xx.34.xx.45 remote-as xx847
neighbor xx.34.xx.45 description NEWER STEALTH 21032007
neighbor xx.34.xx.45 shutdown
neighbor xx.34.xx.45 next-hop-self
neighbor xx.34.xx.45 send-community
neighbor xx.34.xx.45 soft-reconfiguration inbound
neighbor xx.34.xx.45 distribute-list 56 in
neighbor xx.34.xx.45 prefix-list MY_ROUTES out
neighbor xx.34.xx.45 route-map STEALTH-IN in
neighbor xx.34.xx.45 route-map STEALTH-OUT out
neighbor xx.58.xx.41 remote-as xx57
neighbor xx.58.xx.41 description NEW STEALTH
neighbor xx.58.xx.41 shutdown
neighbor xx.58.xx.41 next-hop-self
neighbor xx.58.xx.41 send-community
neighbor xx.58.xx.41 soft-reconfiguration inbound
neighbor xx.58.xx.41 distribute-list 56 in
neighbor xx.58.xx.41 prefix-list MY_ROUTES out
neighbor xx.58.xx.41 route-map STEALTH-IN in
neighbor xx.58.xx.41 route-map STEALTH-OUT out
neighbor XX.20.XX.252 remote-as xx30
neighbor XX.20.XX.252 description NEW TELECITY
neighbor XX.20.XX.252 next-hop-self
neighbor XX.20.XX.252 send-community
neighbor XX.20.XX.252 soft-reconfiguration inbound
neighbor XX.20.XX.252 prefix-list MY_ROUTES out
neighbor XX.20.XX.252 route-map TELECITY-IN in
neighbor XX.20.XX.252 route-map TELECITY-OUT out
neighborXX.20.XX.251 remote-as xx30
neighborXX.20.XX.251 description BACKUP TELECITY 27-10-2006
neighborXX.20.XX.251 next-hop-self
neighborXX.20.XX.251 send-community
neighborXX.20.XX.251 soft-reconfiguration inbound
neighborXX.20.XX.251 prefix-list MY_ROUTES out
neighborXX.20.XX.251 route-map TELECITYBAK-IN in
neighborXX.20.XX.251 route-map TELECITYBAK-OUT out
no auto-summary
!
ip default-gateway xx.20.34.xx2
no ip forward-protocol nd
!
!
!
ip access-list extended CIRPACK
permit ip host xxx.x.97.132 any
permit ip host xx.x.47.228 any
deny ip any any log
ip access-list extended GigaBit0_1
permit ip host XX.XXX.0.203 any
permit ip any host XX.XXX.0.203
!
!
ip prefix-list MY_ROUTES seq 5 permit xx.xx.0.0/21
ip prefix-list MY_ROUTES seq 10 permit xxx.xx.141.0/24
the Cef is enabled for all interfaces
Discover and save your favorite ideas. Come back to expert answers, step-by-step guides, recent topics, and more.
New here? Get started with these tips. How to use Community New member guide