Controllerless Networks

 View Only
last person joined: 13 hours ago 

Instant Mode - the controllerless Wi-Fi solution that's easy to set up, is loaded with security and smarts, and won't break your budget
Expand all | Collapse all

Virtual cluster offline, but still pingable

This thread has been viewed 26 times
  • 1.  Virtual cluster offline, but still pingable

    Posted Nov 26, 2021 10:29 AM
      |   view attached
    Hi all,
    We have a situation in one of our sites. Let me try to describe the situation.
    We have in one site 3 clusters of IAP's managed by Airwave (3 clusters, because in total there are 185 AP's). All AP's are IAP515 on 8.6.0.14-8.6.0.14_81691.
    Now one of those clusters went completely offline. However, I checked a few AP's and they are still pingable + I can still login to the web interface of those AP's.
    When I log in to the AP's themselves, I see that they are still managed by Airwave, but they also show that there are 0 AP's in the cluster + there are no "Alerts"... See screenshot in attachment.
    what I already did was this:
     - I checked which AP was the last known master AP in Airwave and I rebooted it by removing the PoE on the switchport + put it back after 10 seconds or so
     - I rebooted all AP's on that same switch (around 10) using the same method as my previous try.

    None of the above seems to help.

    Does anyone have an idea how I can fix this?

    Thanks!


  • 2.  RE: Virtual cluster offline, but still pingable

    EMPLOYEE
    Posted Nov 26, 2021 01:46 PM
    If you can SSH into any one of the APS, type "show log system" to see if you can have a clue why they cannot see each other.  I am assuming that they all are in the same subnet.

    ------------------------------
    Any opinions expressed here are solely my own and not necessarily that of Hewlett Packard Enterprise or Aruba Networks.
    ------------------------------



  • 3.  RE: Virtual cluster offline, but still pingable

    Posted Nov 29, 2021 03:19 AM
    Hi,
    I could login to one of the AP's through SSH.

    Below is a copy of 3 minutes of logs.

     Swarm State Change from Startup to Found
     PMM: cli_process_pwr_change now swarm state is swarm_master_found
     Swarm State Change from Startup to Found
     Cluster security is enabled, not attempting recovery
     Swarm State Change from Startup to Found
     PMM: cli_process_pwr_change now swarm state is swarm_master_found
     Swarm State Change from Startup to Found
     PMM: cli_process_pwr_change now swarm state is swarm_master_found
     Swarm State Change from Startup to Found
     Cluster security is enabled, not attempting recovery
     PMM: cli_process_pwr_change now swarm state is swarm_master_found
     Swarm State Change from Startup to Found
     PMM: cli_process_pwr_change now swarm state is swarm_master_found
     Swarm State Change from Startup to Found
     PMM: cli_process_pwr_change now swarm state is swarm_master_found
     Swarm State Change from Startup to Found
     Cluster security is enabled, not attempting recovery
     Swarm State Change from Startup to Found
     PMM: cli_process_pwr_change now swarm state is swarm_master_found
     Swarm State Change from Startup to Found
     PMM: cli_process_pwr_change now swarm state is swarm_master_found
    
     Swarm State Change from Startup to Found
     PMM: cli_process_pwr_change now swarm state is swarm_master_found
     Swarm State Change from Startup to Found
     Cluster security is enabled, not attempting recovery
     Swarm State Change from Startup to Found
     PMM: cli_process_pwr_change now swarm state is swarm_master_found
     Swarm State Change from Startup to Found
     PMM: cli_process_pwr_change now swarm state is swarm_master_found
     Swarm State Change from Startup to Found
     Cluster security is enabled, not attempting recovery
     Swarm State Change from Startup to Found
     PMM: cli_process_pwr_change now swarm state is swarm_master_found
     Swarm State Change from Startup to Found
     PMM: cli_process_pwr_change now swarm state is swarm_master_found
     Swarm State Change from Startup to Found
     PMM: cli_process_pwr_change now swarm state is swarm_master_found
     Swarm State Change from Startup to Found
     Cluster security is enabled, not attempting recovery
     Swarm State Change from Startup to Found
     PMM: cli_process_pwr_change now swarm state is swarm_master_found
     Swarm State Change from Startup to Found
     PMM: cli_process_pwr_change now swarm state is swarm_master_found
     Swarm State Change from Startup to Found
     Cluster security is enabled, not attempting recovery
     Swarm State Change from Startup to Found
     PMM: cli_process_pwr_change now swarm state is swarm_master_found
     Swarm State Change from Startup to Found
     PMM: cli_process_pwr_change now swarm state is swarm_master_found
     Swarm State Change from Startup to Found
     PMM: cli_process_pwr_change now swarm state is swarm_master_found
     Swarm State Change from Startup to Found​
    I tried the "show swarm ...." commands, but I couldn't find which AP is the master of the swarm.

    Do you know (if there is) a CLI command to know which AP is the master?


  • 4.  RE: Virtual cluster offline, but still pingable

    Posted Dec 01, 2021 07:50 AM
    with the below commands I could find that the master IAP was not reachable
    show election
    show cluster-security
    show datapath session

    Eventually, I found that there was a bad uplink between 2 switches. The master AP of this VC happened to be behind this bad uplink. I tried shutting down the interface, so that particular switch went unreachable, but nothing changed. So I needed to reboot all ap's in this cluster (while the bad switch was still unreachable) and all came online again. After the bad uplink was fixed, I asked our local guys to reboot that switch so that all AP's there were also rebooted. All is ok now