Switching

last person joined: yesterday 

Ask questions and share experiences about EX and QFX portfolios and all switching solutions across your data center, campus, and branch locations.

  • 1.  Endless chassism ethswitch_eth_devstop messages filling up the log file

    Posted 06-12-2014 04:21

    Hi,

     

    This is my first forum post.

     

    I could not find an answer on the internet and am trusting that you knowledgeable people can advise me on a solution.

     

    The messages below fills up our messages log so fast that all 9 messages log files are filled up within 2 hours.  We have no idea where this message comes from or how to correct, ignore or disable it.  Can any of you advise on this please?

     

    This is for an EX8208 switch running Junos 11.4R8.5.

     

    Thank you

    Jacques

     

    ...

    Jun 12 13:00:01.334 2014  OURBOXNAME_re0 /kernel: Jun 12 12:59:50.000 2014  OURBOXNAME_re0 member0-fpc2  chassism[121]: ethswitch_eth_devstop: called for port xe-2/0/3
    Jun 12 13:00:01.334 2014  OURBOXNAME_re0 /kernel: Jun 12 12:59:50.000 2014  OURBOXNAME_re0 member0-fpc2  chassism[121]: ethswitch_eth_devstop: called for port xe-2/0/7
    Jun 12 13:00:01.334 2014  OURBOXNAME_re0 /kernel: Jun 12 12:59:50.000 2014  OURBOXNAME_re0 member0-fpc2  chassism[121]: ethswitch_eth_devstop: called for port xe-2/0/8
    Jun 12 13:00:01.334 2014  OURBOXNAME_re0 /kernel: Jun 12 12:59:50.000 2014  OURBOXNAME_re0 member0-fpc2  chassism[121]: ethswitch_eth_devstop: called for port xe-2/0/12
    Jun 12 13:00:01.334 2014  OURBOXNAME_re0 /kernel: Jun 12 12:59:50.000 2014  OURBOXNAME_re0 member0-fpc2  chassism[121]: ethswitch_eth_devstop: called for port xe-2/0/13
    Jun 12 13:00:01.334 2014  OURBOXNAME_re0 /kernel: Jun 12 12:59:50.000 2014  OURBOXNAME_re0 member0-fpc2  chassism[121]: ethswitch_eth_devstop: called for port xe-2/0/18
    Jun 12 13:00:01.334 2014  OURBOXNAME_re0 /kernel: Jun 12 12:59:50.000 2014  OURBOXNAME_re0 member0-fpc2  chassism[121]: ethswitch_eth_devstop: called for port xe-2/0/25
    Jun 12 13:00:01.334 2014  OURBOXNAME_re0 /kernel: Jun 12 12:59:50.000 2014  OURBOXNAME_re0 member0-fpc2  chassism[121]: ethswitch_eth_devstop: called for port xe-2/0/26
    Jun 12 13:00:01.334 2014  OURBOXNAME_re0 /kernel: Jun 12 12:59:50.000 2014  OURBOXNAME_re0 member0-fpc2  chassism[121]: ethswitch_eth_devstop: called for port xe-2/0/27
    Jun 12 13:00:01.334 2014  OURBOXNAME_re0 /kernel: Jun 12 12:59:50.000 2014  OURBOXNAME_re0 member0-fpc2  chassism[121]: ethswitch_eth_devstop: called for port xe-2/0/35
    Jun 12 13:00:01.334 2014  OURBOXNAME_re0 /kernel: Jun 12 12:59:50.000 2014  OURBOXNAME_re0 member0-fpc2  chassism[121]: ethswitch_eth_devstop: called for port xe-2/0/36

    ...



  • 2.  RE: Endless chassism ethswitch_eth_devstop messages filling up the log file

     
    Posted 06-12-2014 04:33

    Are the interfaces mentioned on these logs flapping?

     

    What do you see if you are issuing this command:

     

    show log messages | match xe-2/0/12

     

    Do you see any service impact that you can correlate with the interfaces mentioned in the logs?

     

    These messages by themselves do not represent a problem, but I suggest you to open a JTAC case and get more information about them if you need them.

     

    If you want to hide these messages meanwhile, you can use this:

     

    set system syslog file messages match "! ethswitch_eth_devstop: called for port"

     

     

    =====

    If this worked for you please flag my post as an "Accepted Solution" so others can benefit. A kudo would be cool if you think I earned it.



  • 3.  RE: Endless chassism ethswitch_eth_devstop messages filling up the log file

    Posted 06-12-2014 04:51

    Thank you for your quick response.

     

    I have added that system ignore message statement to the messages log file.  I have also added to the processes log file as the exact same messages also appeared in there.

     

    The most recent time any interface flapped was yesterday and these logs are for current time.  Most of the interfaces flapped more than a month ago when we last rebooted the switch.

     

    Now the messages log file shows the following messages and this is the same for the 'show' command you gave:

     

    Jun 12 13:41:30.966 2014  OURBOXNAME_re0 /kernel: led for port xe-2/0/25
    Jun 12 13:41:31.075 2014  OURBOXNAME_re0 /kernel: led for port xe-2/0/12
    Jun 12 13:41:31.188 2014  OURBOXNAME_re0 /kernel: led for port xe-2/0/36
    Jun 12 13:41:31.335 2014  OURBOXNAME_re0 /kernel: led for port xe-2/0/8
    Jun 12 13:41:31.424 2014  OURBOXNAME_re0 /kernel: led for port xe-2/0/12
    Jun 12 13:41:31.553 2014  OURBOXNAME_re0 /kernel: led for port xe-2/0/18
    Jun 12 13:41:31.633 2014  OURBOXNAME_re0 /kernel: led for port xe-2/0/25
    Jun 12 13:41:31.705 2014  OURBOXNAME_re0 /kernel: led for port xe-2/0/36
    Jun 12 13:41:32.009 2014  OURBOXNAME_re0 /kernel: led for port xe-2/0/7
    Jun 12 13:41:32.033 2014  OURBOXNAME_re0 /kernel: led for port xe-2/0/8
    Jun 12 13:41:32.184 2014  OURBOXNAME_re0 /kernel: led for port xe-2/0/25
    Jun 12 13:41:32.253 2014  OURBOXNAME_re0 /kernel: led for port xe-2/0/26
    Jun 12 13:41:32.388 2014  OURBOXNAME_re0 /kernel: led for port xe-2/0/3
    Jun 12 13:41:32.474 2014  OURBOXNAME_re0 /kernel: led for port xe-2/0/12
    Jun 12 13:41:32.553 2014  OURBOXNAME_re0 /kernel: led for port xe-2/0/13
    Jun 12 13:41:32.688 2014  OURBOXNAME_re0 /kernel: led for port xe-2/0/35
    Jun 12 13:41:32.791 2014  OURBOXNAME_re0 /kernel: led for port xe-2/0/3
    Jun 12 13:41:32.883 2014  OURBOXNAME_re0 /kernel: led for port xe-2/0/12
    Jun 12 13:41:33.000 2014  OURBOXNAME_re0 /kernel: led for port xe-2/0/25
    Jun 12 13:41:33.091 2014  OURBOXNAME_re0 /kernel: led for port xe-2/0/35
    Jun 12 13:41:33.349 2014  OURBOXNAME_re0 /kernel: led for port xe-2/0/7
    Jun 12 13:41:33.424 2014  OURBOXNAME_re0 /kernel: led for port xe-2/0/26

    ...

     

     

    Are there any other debug commands I can run to investigate this issue?

     

    Thank you



  • 4.  RE: Endless chassism ethswitch_eth_devstop messages filling up the log file

     
    Posted 06-12-2014 04:57

    Nothing that we can do here.

     

    Please open a JTAC case and follow up with them.

     

     

    =====

    If this worked for you please flag my post as an "Accepted Solution" so others can benefit. A kudo would be cool if you think I earned it.



  • 5.  RE: Endless chassism ethswitch_eth_devstop messages filling up the log file
    Best Answer

    Posted 07-21-2014 03:37

    An update to all interested:

     

    The issue that triggered this message was for interfaces with a status of 'down up'.  That's right, administratively down, but with an active operational line.  All the messages were for these interfaces.

     

    Herewith the official response from Juniper:

     

    Engineering is able to reproduce the behavior and identified the cause of the posts that are made in periodic check on the status of the interfaces.
    This problem is considered not impacting on service on the continuous writing of the log has been fixed since release 11.4R10.