Forum Discussion

K1974_98730's avatar
K1974_98730
Icon for Nimbostratus rankNimbostratus
May 10, 2018

Devices are not synchronizing configuration to each other

Hi folks,

 

I´m new on Big IP and started my studies on LTM. I´ve set up two VEs running 13.1.0.5 software and I´m facing some issues in regards of synchronization of the configuration. I tried to create device-groups manually, rebuild the trust domain, push the last config from both devices, force an specific device to be the leader sync, restart processes (mcpd, devmgmtd, ntpd). Configsync-ips are correct and responding, mcpd connection in Established, I´ve already load conf to default (both VEs), read a lot of forums/discussion and at this time I´m still have no idea about why the status is Always “Not all Devices Synced. Most likely I´m missing something, did something wrong (or both).

 

Thanks in advance for the help.

 

  • I have also run into this caveat. I am running 13.1.1.4 on VMware Workstation 15 on Fedora 29 Linux kernel 5.0.4

     

    Thanks to the help of F5 FSE, I have managed to resolve it by adding the following DB variables on each VM on WMware Workstation 15.x:

     

    tmsh modify /sys db tm.tcplargereceiveoffload value disable

     

    tmsh modify /sys db tm.tcpsegmentationoffload value disable

     

     

    It appears that WMware loses packets when offload is enabled.

     

  • Snl's avatar
    Snl
    Icon for Cirrostratus rankCirrostratus

    make sure time is sync in both devices

     

  • Can you check your logs? Post any errors you see here to diag this issue? Did you configure a sync vlan....can you telnet to each other device on ports 443,22? Are both devices part of device trust? run the below commands

     

    grep -i cmi /var/log/ltm

     

    grep -i configsync /var/log/ltm

     

  • Thanks for your reply, snl! I´ve set up a NTP server on both devices and also I tried to set up date and time manually however, it didn´t work.

     

    Regards!

     

  • I am using VM workstation as well...and I never see the kind of behavior you are describing. So I am just curious. Anyway good luck.

     

  • I have the same issues. Have you had a solution for that? Thanks in advance.

     

    • Krystian_Baniak's avatar
      Krystian_Baniak
      Icon for Altocumulus rankAltocumulus

      I have also run into this caveat. I am running 13.1.1.4 on VMware Workstation 15 on Fedora 29 Linux kernel 5.0.4

       

      Thanks to the help of F5 FSE, I have managed to resolve it by adding the following DB variables on each VM on WMware Workstation 15.x:

       

      tmsh modify /sys db tm.tcplargereceiveoffload value disable

       

      tmsh modify /sys db tm.tcpsegmentationoffload value disable

       

       

      It appears that WMware loses packets when offload is enabled.

       

      • loi's avatar
        loi
        Icon for Nimbostratus rankNimbostratus

        Thank you so much Krysitan Baniak, i've tried to run 2 above commands on each VM, it worked.

         

    • K1974_98730's avatar
      K1974_98730
      Icon for Nimbostratus rankNimbostratus

      Hello DNTDS,

       

      Not at all... I´ve removed all the configuration from the VMs, installed again, licensed again and it didn´t work as well. I also tried to use 13.1.0.6 which is not working either. It´s very strange as 13.1.0.5 was working before.

       

      The status in VEs keep on changing from "changes pending" to "awaiting initial config" and so on, other day the traffic-group-1 was have no devices on it...

       

      Not sure what is happening... that´s sad!

       

    • DNDTS_361701's avatar
      DNDTS_361701
      Icon for Nimbostratus rankNimbostratus

      Have you tried another older version? I tried version 13.0.0.0.0.1645 and it worked.

       

  • gha's avatar
    gha
    Icon for Nimbostratus rankNimbostratus

    issue still exists in version BIG-IP 14.0.0.2.

     

    I haven't been able to successfully sync two VEs running in VMware Workstation since 13.0.

     

    Last version this worked was 12.1.x.

     

    I do loads of demos, trainings, workshops etc. using the vLab environment. It's a pity I can't show DSC on any current version.

     

  • I can partially answer this question. I was able to successfully configure DSC for VE v13.0.0 but with LAB license. Each VE required 8 GB of RAM and 4 CPUs (ESXi environment). Not sure if this works for Trial or Evaluation licenses as far I didn't manage to configure DSC with 4 GB of RAM and 2 CPUs per VE using both of those license types (VMware Fusion environment).