DDS 6 update - Automatic participant index failed to initialize

3 posts / 0 new
Last post
Offline
Last seen: 2 weeks 3 days ago
Joined: 09/11/2014
Posts: 5
DDS 6 update - Automatic participant index failed to initialize

Dear RTI-Team, dear community,

we upgraded a virtual machine (Debian 9.9) from DDS 5.3.1 (perfectly running) to DDS 6

The machine is mostly used for routing service. We upated the routing service config xml according to DDS 6 (cross checked & working on another DDS 6 machine)

However when we start the routing service we get:

DDS_DomainParticipant_enableI:Automatic participant index failed to initialize. PLEASE VERIFY CONSISTENT TRANSPORT / DISCOVERY CONFIGURATION.

Even if we run the rtiddsspy we get the same message

we read:

https://community.rti.com/kb/osx510

https://community.rti.com/forum-topic/please-verify-constistent-transport-discovery-configuration

But could not extract a solution - What is special about the VM used is that works as a server for 2 VPN (OpenVPN) tunnels.

Any idea what could cause this (5.3.1 running perfect and no changes to anything else)

Thanks in advance,

Thomas

Offline
Last seen: 2 weeks 3 days ago
Joined: 09/11/2014
Posts: 5

we investigated a bit:

we found out that for the routing service for each participant this would solve it:

<participant name="1">
  <domain_id>1</domain_id>
  <participant_qos>
    <wire_protocol>
      <rtps_auto_id_kind>RTPS_AUTO_ID_FROM_IP</rtps_auto_id_kind>
    </wire_protocol>
  </participant_qos>
</participant>    
Offline
Last seen: 2 weeks 3 days ago
Joined: 09/11/2014
Posts: 5

One more addittion:

The vm that are used are cloned, different NIC-MAC but same harddrive UUID. Can this influence the way DDS 6 generates the participant ID?

Strangely one machine has this issue the other did not. Independent of wether some DDS is running or not? (we tested all combinations)

The issue is solved with a workaround for the routing service, but tools like the spy are not working still. How could we solve this one using DDS settings only? (if the cloned machines are the issue we will remove this root cause instead of having multiple sets of configs)