1 person found this helpful
I have an open ticket with Oracle on the same issue but for Windows. I hadn't seen this on Linux as of yet but you can try a couple of temporary workarounds.
First, I always get this issue when I have the standard hosts entry of 127.0.0.1 localhost. Not sure why (Oracle is investigating) but it's guaranteed to fail on my laptop. If you get rid of that entry and add an entry for your machine name to 127.0.0.1, and re-start Endeca Server, I think you should be in better shape.
Good hosts file
Bad hosts file (note: a lot of Windows machines will have this by default)
I have to believe there's an actual fix for this coming but this will, hopefully, get you past the issue for the moment.
Thanks for your answer.
My hosts file looks like this now:
However, I am still having the same result:
[oracle@oel6 endeca-cmd]$ ./endeca-cmd create-dd testing --port 7051
SEVERE: Error while invoking endpoint "http://localhost:7051/endeca-server/ws/manage" from client
endeca-cmd encountered a problem.
The Endeca Server returned an error: OES-000169: Could not connect to Endeca Server at 192:7051
OES-000169: Could not connect to Endeca Server at 192:7051
oel6 (hostname) works properly when invoking http://oel6:7051/console but of course, not localhost:7051/console.
The server invokes localhost, so I suppose it is not possible for it to resolve it without that in the hosts file, is it?
Thank you again for your help,
Also tried following line:
[oracle@oel6 endeca-cmd]$ ./endeca-cmd create-dd testing --port 7051 --host oel6
Got same result even though new address is "http://oel6:7051/endeca-server/ws/manage"
Can you try to have 192.168.0.101 resolve to oel6 as well? So, should look something like this:
#nothing for localhost
That at least works for me temporarily on my own laptop. Still feels like a bug of some kind since I don't think the final answer is to edit the hosts file in a production deployment on Windows.
I experienced the same problem where supplying the host name to endeca-cmd still did not work. The reason is that the error is generated server side after the request gets to the engine.
I´m sorry I couldn´t try it before but, it worked!
I removed the localhost reference letting the hosts file as you suggested:
# nothing else
It works properly (endeca-cmd, bulk add/replace records...). Only "issue" is that I have to type --host xxx every time when invoking the server.
(I wrote the solution in case it is useful for someone else)
Thank you very much for your help,