Switching

Expand all | Collapse all

ex4300-48t - pfe_khms_spurious_wakeup - weird logs after upgrade

  • 1.  ex4300-48t - pfe_khms_spurious_wakeup - weird logs after upgrade

    Posted 24 days ago
    Hi All,

    I have got a weird case on JunOS - 19.1R3-S5.3  -  somehow 6 months ago release 19.1 was recommended, we started than an upgrade plan in company (around 100 logical switches).
    In two cases I have noticed weird behavior in logs (both switches work in different networks/sites).

    show log messages is flooded by logs like follows:

    Jun 29 07:45:01.225 2021 SWITCH-SITE-A eventd[1351]: EVENTD_KERNEL_LOGS_DROPPED: Due to excessive logging, (83) kernel events dropped by eventd
    Jun 29 06:29:08.042 2021 SWITCH-SITE-A /kernel: peer_master_work: 5631: pfe_khms_spurious_wakeup: 100739411 Peer (class: 0, type: 17, index: 0, vksid: 1, state: 1 pfe_stats_spurious_wakeup: 100739411 )
    Jun 29 07:45:01.225 2021 SWITCH-SITE-A eventd[1351]: EVENTD_KERNEL_LOGS_DROPPED: Due to excessive logging, (3) kernel events dropped by eventd
    Jun 29 06:29:08.042 2021 SWITCH-SITE-A /kernel: peer_master_work: 5631: pfe_khms_spurious_wakeup: 100739412 Peer (class: 0, type: 17, index: 0, vksid: 1, state: 1 pfe_stats_spurious_wakeup: 100739412 )
    Jun 29 07:45:01.227 2021 SWITCH-SITE-A eventd[1351]: EVENTD_KERNEL_LOGS_DROPPED: Due to excessive logging, (3) kernel events dropped by eventd
    Jun 29 06:29:08.069 2021 SWITCH-SITE-A /kernel: peer_master_work: 5631: pfe_khms_spurious_wakeup: 100739519 Peer (class: 0, type: 17, index: 0, vksid: 1, state: 1 pfe_stats_spurious_wakeup: 100739519 )
    Jun 29 07:45:01.227 2021 SWITCH-SITE-A eventd[1351]: EVENTD_KERNEL_LOGS_DROPPED: Due to excessive logging, (3) kernel events dropped by eventd
    Jun 29 06:29:08.069 2021 SWITCH-SITE-A /kernel: peer_master_work: 5631: pfe_khms_spurious_wakeup: 100739520 Peer (class: 0, type: 17, index: 0, vksid: 1, state: 1 pfe_stats_spurious_wakeup: 100739520 )
    (it is repeating constantly , few time per second, and flooding entire log file)

    Jun 29 07:19:27.259 2021 SWITCH-SITE-B /kernel: peer_master_work: 5631: pfe_khms_spurious_wakeup: 3511584113 Peer (class: 0, type: 17, index: 0, vksid: 1, state: 1 pfe_stats_spurious_wakeup: 3511584113 )
    Jun 29 08:32:15.620 2021 SWITCH-SITE-B eventd[1349]: EVENTD_KERNEL_LOGS_DROPPED: Due to excessive logging, (106) kernel events dropped by eventd
    Jun 29 07:19:27.259 2021 SWITCH-SITE-B /kernel: peer_master_work: 5631: pfe_khms_spurious_wakeup: 3511584114 Peer (class: 0, type: 17, index: 0, vksid: 1, state: 1 pfe_stats_spurious_wakeup: 3511584114 )
    Jun 29 07:19:27.286 2021 SWITCH-SITE-B /kernel: peer_master_work: 5631: pfe_khms_spurious_wakeup: 3511584221 Peer (class: 0, type: 17, index: 0, vksid: 1, state: 1 pfe_stats_spurious_wakeup: 3511584221 )
    Jun 29 08:32:15.647 2021 SWITCH-SITE-B eventd[1349]: EVENTD_KERNEL_LOGS_DROPPED: Due to excessive logging, (106) kernel events dropped by eventd
    Jun 29 07:19:27.286 2021 SWITCH-SITE-B /kernel: peer_master_work: 5631: pfe_khms_spurious_wakeup: 3511584222 Peer (class: 0, type: 17, index: 0, vksid: 1, state: 1 pfe_stats_spurious_wakeup: 3511584222 )
    Jun 29 08:32:15.649 2021 SWITCH-SITE-B eventd[1349]: EVENTD_KERNEL_LOGS_DROPPED: Due to excessive logging, (106) kernel events dropped by eventd
    Jun 29 07:19:27.314 2021 SWITCH-SITE-B /kernel: peer_master_work: 5631: pfe_khms_spurious_wakeup: 3511584329 Peer (class: 0, type: 17, index: 0, vksid: 1, state: 1 pfe_stats_spurious_wakeup: 3511584329 )
    Jun 29 08:32:15.664 2021 SWITCH-SITE-B eventd[1349]: EVENTD_KERNEL_LOGS_DROPPED: Due to excessive logging, (3) kernel events dropped by eventd
    (it is repeating constantly , few time per second, and flooding entire log file)


    1) In both cases these are Virtual-Chassis built of 2 physical ex4300-48t
    2) On SWITCH-SITE-A we have got Junos: 19.1R3-S5.3 & on the SWITCH-SITE-B Junos: 19.1R3-S4.6
    3) in both cases logs appeared after upgrade to this version, it was simple upgrade with full reboot
    4) Please make have a look on the TIMESTAMPS, there is different time "events" and different for "/kernel" and difference is around 1 hour and 15 minutes.
    5) There are other switches around these, which works in the same LAN's where issues have not appeared after upgrades.
    6) there was no any configuration changes during an upgrade
    ​​​​​​

    Has anyone similar symptoms?


  • 2.  RE: ex4300-48t - pfe_khms_spurious_wakeup - weird logs after upgrade

    Posted 22 days ago
    This seems to be a log indicating something internal to the platform, which is hard to troubleshoot unless one has access to internal Juniper resources to find out what the log really means. I guess, open a JTAC case and see what details they can share on this.

    Elvin