Skip to main content

Able to connect to Vertica locally, but not from remote!

  • September 22, 2015
  • 3 replies
  • 19 views
Sandeep_Thool

Hi,

 

I have a 3 node cluster of Vertica. I am able to login into individual nodes & connect to each local node using VSQL & issue SQL commands successfully. Even via admintools tools it shows the cluster with all 3 nodes is 'UP'.

 

But I am not able to connect to this cluster from remote. There are no firewall issues to speak of, as this was a working setup (was able to connect from remote without any issues!). I also flushed local IPTABLES on all nodes, but I'm still not able to connect.

 

One important aspect that I want to point out is that, last week one of the nodes went 'out of contact' with the rest of the 2 nodes, because someone accidently updated its IPTABLES. Those IPTABLES entries have since been removed, Vertica went through a recovery process after that & now admintools says that the cluster is 'UP'.

 

Remember, I'm still able to connect to the database when i login to the RHEL box & issue VSQL commands. I can even see my data on all 3 nodes. The issue here being that i'm not able to connect from remote!

 

Please adivice or help.

 

Thanks

    3 replies

    Adrian_Oprea_1
    Forum|alt.badge.img+1
    • Adrian_Oprea_1
    • Participating Frequently
    • September 23, 2015

     Is this on AWS  ? 

     

    Rahul_Choudhary
    Forum|alt.badge.img
    • Rahul_Choudhary
    • Participating Frequently
    • September 30, 2015

    Hi Sandeep

     

    Are you using vsql client to connect to the vertica cluster from outside or some other driver?Also what error you receive when you try to make a connection remotely?

     

    Also please provide us the output of below commands:

     

    $ admintools -t list_allnodes

    $ admintools -t view_cluster

     

    vsql => select * from nodes;

     

    From all the nodes


    # iptables -L 

     

    Regards

    Rahul

    scorpioviper
    • scorpioviper
    • New Participant
    • October 2, 2015

    Thanks folks! The issue was indeed a network component dynamically modifying IPTABLES of one of these nodes, which was causing it to breakaway from the rest of the cluster. This issue has now been resolved. Thanks for your attention.

    Still scrolling?
    Ask a question
    Terms & ConditionsAccessibility statement