Junos OS

IMPORTANT MODERATION NOTICE

This community is currently under full moderation, meaning  all posts will be reviewed before appearing in the community. Please expect a brief delay—there is no need to post multiple times. If your post is rejected, you'll receive an email outlining the reason(s). We've implemented full moderation to control spam. Thank you for your patience and participation.



Expand all | Collapse all

BGP task replication not started

  • 1.  BGP task replication not started

    Posted 08-18-2021 09:11
    Hi,
    I'd be really grateful if someone could shed some light on the following output:

    show task replication logical-system XXXXXXXXXX
    Stateful Replication: Enabled
    RE mode: Master

    Protocol Synchronization Status
    OSPF Complete
    BGP NotStarted <-------
    MPLS Complete
    EVPN Complete

    Enabling traceoptions on bgp nsr-synchronisation gives:
    bgp_rsync_listen_accept: RSYNC session from backup refused because JSR=0 not active or nonstop-routing=1 not enabled

    Any thoughts/ Obvious mistakes/ ways to investigate further?

    Thanks in advance



    ------------------------------
    BEN BIGGIE
    ------------------------------


  • 2.  RE: BGP task replication not started

     
    Posted 08-19-2021 05:35
    Hi Ben,

    Do you have this BGP in a logical system while other protocols (OSPF etc.) elsewhere? Please ensure you have this enabled:
    set logical-systems xxxxx routing-options nonstop-routing

    Also, please check logs (show log messages) for the following:

    <TimeStamp> rpd[16033]: %DAEMON-3: task_jsr_alloc: File exists
    <TimeStamp> kernel: %KERN-3: jsr_sdrl_setup_primary: Socket already has replication setup (state 0x1)
    <TimeStamp> kernel: %KERN-3: jsr_sdrl_pri_alloc: failed 17

    If these logs are seen, please check if the Junos version has this fix:  https://prsearch.juniper.net/InfoCenter/index?page=prcontent&id=PR1552603
    Restarting RPD or reboot of the RE's should also work this around temporarily.

    Hope this helps.
    Regards,
    -r.
    --------------------------------------------------
    If this solves your problem, please mark this post as "Accepted Solution."
    Kudos are always appreciated :).



  • 3.  RE: BGP task replication not started

     
    Posted 08-19-2021 05:36

    Hi Ben,

     

    Do you have this BGP in a logical system while other protocols (OSPF etc.) elsewhere? Please ensure you have this enabled:

    set logical-systems xxxxx routing-options nonstop-routing

     

    Also, please check logs (show log messages) for the following:

     

    <TimeStamp> rpd[16033]: %DAEMON-3: task_jsr_alloc: File exists

    <TimeStamp> kernel: %KERN-3: jsr_sdrl_setup_primary: Socket already has replication setup (state 0x1)

    <TimeStamp> kernel: %KERN-3: jsr_sdrl_pri_alloc: failed 17

     

    If these logs are seen, please check if the Junos version has this fix:  https://prsearch.juniper.net/InfoCenter/index?page=prcontent&id=PR1552603

    Restarting RPD or reboot of the RE's should also work this around temporarily.

     

    Hope this helps.

    Regards,

    -r.

    --------------------------------------------------

    If this solves your problem, please mark this post as "Accepted Solution."

    Kudos are always appreciated :).


    Juniper Business Use Only






  • 4.  RE: BGP task replication not started

    Posted 08-19-2021 08:29
    Hi (thought I'd posted a reply)

    To answer you questions:
    I do have non-stop routing set within the LS
    All protocols (OSPF+SR, MPLS, BGP running on same LS
    No matches on any of those messages 
    I did try and restart routing on the LS in both REs (not in production yet,
    With regards to rev: I'm currently on 19.3R2-S3.4 - I'm intending to get to latest GA before going live.
    Thanks again

    ------------------------------
    BEN BIGGIE
    ------------------------------



  • 5.  RE: BGP task replication not started

    Posted 08-24-2021 05:26
    Hi, I just thought I'd say I've sorted this issue as I was able to do a full reboot because we're not in production yet.

    Thanks for your help

    ------------------------------
    BEN BIGGIE
    ------------------------------