cancel
Showing results for 
Search instead for 
Did you mean: 
cancel
1481
Views
0
Helpful
0
Replies
Beginner

High cpu utilisation on mds 9124s

Hi All

Currently we are experiencing what looks like high cpu utilisation

on the MDS 9124 switches.

If i do a sh processes cpu  ---

                                                             
     152            251                       4 6    93        
    564155555655 59769066555555555654166655655375555509565555555
100                                                            
90                                                  #         
80                                                  #         
70                                             #    #         
60                  #                          #    #         
50   #              #                          #    #         
40   #              #                        # #    ##        
30   #             ##                        # #    ##        
20  ###            ##                        # #    ##        
10 ############ ###################  ##########################
    0....5....1....1....2....2....3....3....4....4....5....5....
              0    5    0    5    0    5    0    5    0    5   

               CPU% per second (last 60 seconds)
                      # = average CPU%


         1                        1 1     11  1            1   
    9 11602716812 49162492816912170602391800960 6911259181501216
    097580099693282283154079618084050065850004079683125691204181
100      *                        * *  *  **  *  *    *    *   
90 *    *    *    *    * *  *    * *  * **** *  *    * *  *   
80 *    * *  *    *    * *  *    * *  * **** *  *    * *  *   
70 *   ** * **    *    * * **   ****  * **** * **    * *  *   
60 *   ** * **    * *  * * **   ****  * ****** **    * *  *   *
50 *   ** * **    * * ** * **   ****  * ****** **   ** * **   *
40 *   ** * **   ** * ** * **   **** ** ****** **   ** * **   *
30 *   ** * **   ** * ** * **   **** ** ****** **   ** * **   *
20 * ***#***** * ****************#************ *** ***** *# ***
10 #############*#############################*################
    0....5....1....1....2....2....3....3....4....4....5....5....
              0    5    0    5    0    5    0    5    0    5   

               CPU% per minute (last 60 minutes)
              * = maximum CPU%   # = average CPU%


    11111111111111111111111111111111111111111111111                        
    00000000000000000000000000000000000000000000000                        
    00000000000000000000000000000000000000000000000                        
100 ***********************************************                        
90 ***********************************************                        
80 ***********************************************                        
70 ***********************************************                        
60 ***********************************************                        
50 ***********************************************                        
40 ***********************************************                        
30 ***********************************************                        
20 ***********************************************                        
10 ###############################################                        
    0....5....1....1....2....2....3....3....4....4....5....5....6....6....7.
              0    5    0    5    0    5    0    5    0    5    0    5    0

                   CPU% per hour (last 72 hours)
                  * = maximum CPU%   # = average CPU%

This is  what we see above - we have been on to cisco and have upgraded our entire SAN to NXOS 5.2.2d

but this is still what we get - there are a few bugs about high cpu but they should all be fixed in that release.

If i do a - sh system internal memory-status

then i get a  - MemStatus: OK

if i do a - sh system resources-

load average:   1 minute: 0.02   5 minutes: 0.08   15 minutes: 0.08

Processes   :   174 total, 2 running

CPU states  :   3.9% user,   1.0% kernel,   95.1% idle

Memory usage:    516128K total,    403000K used,    113128K free

Current memory status: OK

I found this -

08-Aug-2012 10:00 (in response to TZwanikken)

Re: High memory usage on Cisco MDS 9124e

It turns out to be a problem with the WWN manager process. Crashes of this process will result in writing error information to core files which stay loaded in memory. This memory is not getting released, which will in turn lead to high memory utilization.

Fix 1: Restart switch to free up core files from memory.

Fix 2: Register SR with Cisco. They will load a "Dplug" which removes the core files from memory without restarting the switch.

After this you have to upgrade to a new NX-OS version, to prevent the bug for claiming memory.

But TAC said nothing about a Dplug and we upgraded to NXOS 5.2.2d we still have a problem -

So is this a real issue  ---  if so any one any clues ?

Steve

Everyone's tags (6)
CreatePlease to create content