[c-nsp] VTEMPLATE Backgr eating CPU

Matteo Sgalaberni sgala at sgala.com
Sun Jul 28 06:54:09 EDT 2013


Hello @ all!

I have a 7206 with NPE-G2 that acting as BRAS and authenticate ATM users via PPP.

Some hours ago I saw that my SNMP graphs are not graphed and some users are not able to re-connect in case of disconnection.

I found that my CPU is at 99%:
bras-mi#show process cpu sort
CPU utilization for five seconds: 99%/22%; one minute: 99%; five minutes: 99%
 PID Runtime(ms)     Invoked      uSecs   5Sec   1Min   5Min TTY Process 
 205  1053768396    16755861      62890 73.54% 74.97% 74.97%   0 VTEMPLATE Backgr 
 284    59478052  1797600834         33  0.71%  0.64%  0.62%   0 PPP Events       
   5   859433372    42828786      20066  0.49%  0.36%  0.41%   0 Check heaps      
  74    14959840   676286131         22  0.42%  0.36%  0.36%   0 ATM OAM TIMER    
  73   257007368  2110983772        121  0.35%  0.30%  0.30%   0 ATM OAM Input    
  51   158636476    34669242       4575  0.21%  0.23%  0.23%   0 Compute load avg 

    111111111111111111111111111111111111111111111111111111111111
    000000000000000000000000000000000000000000000000000000000000
    000000000000000000000000000000000000000000000000000000000000
100 ############################################################
 90 ############################################################
 80 ############################################################
 70 ############################################################
 60 ############################################################
 50 ############################################################
 40 ############################################################
 30 ############################################################
 20 ############################################################
 10 ############################################################
   0....5....1....1....2....2....3....3....4....4....5....5....6
             0    5    0    5    0    5    0    5    0    5    0
               CPU% per minute (last 60 minutes)
              * = maximum CPU%   # = average CPU%

this is my memory:
bras-mi#show processes memor sort
Processor Pool Total:  794527124 Used:  636161608 Free:  158365516
      I/O Pool Total:   67108864 Used:   12087520 Free:   55021344
Transient Pool Total:   33554432 Used:   22811680 Free:   10742752

 PID TTY  Allocated      Freed    Holding    Getbufs    Retbufs Process
 295   0 3486590308 1554152256  371435688      26780          0 BGP Router      
 205   0  969275248 2751907344  112024860          0          0 VTEMPLATE Backgr
 114   0  784318992 3152652872   82767240          0          0 IP RIB Update   
   0   0  793304348  720536068   63065088          0          0 *Init*          
  39   0  270306004     187080    8727936     603620          0 Net Background  
 276   0 2214867672 2005198696    3909096          0          0 PPPoA Manager   

my logs say that there are not so much reconnection to justify this load average on VTEMPLATE 
.Jul 28 10:49:38.046: %LINK-3-UPDOWN: Interface Virtual-Access882, changed state to down
.Jul 28 10:49:38.830: %LINK-3-UPDOWN: Interface Virtual-Access1079, changed state to up
.Jul 28 10:49:38.850: %LINK-3-UPDOWN: Interface Virtual-Access1218, changed state to up
.Jul 28 10:49:39.614: %LINK-3-UPDOWN: Interface Virtual-Access699, changed state to up
.Jul 28 10:49:39.618: %LINK-3-UPDOWN: Interface Virtual-Access1218, changed state to down
.Jul 28 10:49:40.390: %LINEPROTO-5-UPDOWN: Line protocol on Interface Virtual-Access1079, changed state to up
.Jul 28 10:49:40.390: %LINK-3-UPDOWN: Interface Virtual-Access699, changed state to down
.Jul 28 10:49:41.178: %LINK-3-UPDOWN: Interface Virtual-Access723, changed state to up
.Jul 28 10:49:41.286: %LINK-3-UPDOWN: Interface Virtual-Access955, changed state to up
.Jul 28 10:49:41.290: %LINK-3-UPDOWN: Interface Virtual-Access723, changed state to down
.Jul 28 10:49:41.958: %LINK-3-UPDOWN: Interface Virtual-Access1391, changed state to up
.Jul 28 10:49:41.958: %LINK-3-UPDOWN: Interface Virtual-Access955, changed state to down
.Jul 28 10:49:41.966: %LINK-3-UPDOWN: Interface Virtual-Access1391, changed state to down


I'm running a 12.4(24)T6.

What do you think?

Thanks!

Matteo


More information about the cisco-nsp mailing list