Thanks Adrian, That configuration is not what I would expect.
We generate Ellis nginx config automatically using a script. There should be four server_name entries: · Ellis’ hostname · The local IP of the server · The public IP of the server · The public hostname of the server It looks like your local IP, public IP, and public hostname are all set to the same thing. These settings are configured in your node’s local settings. This page describes the values that they should be set to: http://clearwater.readthedocs.org/en/stable/Clearwater_Configuration_Options_Reference.html#local-settings. If the settings in the nginx ellis folder do not match the settings in /etc/clearwater/local_config, I suggest you run sudo service clearwater-infrastructure restart to force the config to re-generate. If your IP and hostname settings are as you want them, it looks like our script needs to be more clever about how it generates config. I’m having trouble with GitHub at the moment, but I will raise an issue to cover this. What symptoms are you seeing as a result of these logs? Are you able to access the Ellis web UI? Chris From: Adrian Schwartz [mailto:[email protected]] Sent: 05 April 2016 06:50 To: Chris Elford (projectclearwater.org) <[email protected]> Cc: [email protected] Subject: Re: [Project Clearwater] nginx fail to start Hi Chris I went over the configuration, I believe that the content of ellis in nginx folder is the fault, it has multiple server with same address, see below. upstream http_ellis { server unix:/tmp/.ellis-sock-0; server unix:/tmp/.ellis-sock-1; # The minimum number of idle connections to keep alive to the upstream. keepalive 16; } server { listen [::]:80 default_server; # ipv6only=off is set in the base config server_name ellis.example.com<http://ellis.example.com> 10.0.0.4 10.0.0.4 10.0.0.4; location / { proxy_pass http://http_ellis; proxy_http_version 1.1; # The client may have instructed the server to close the # connection - do not forward this upstream. proxy_set_header Connection ""; # Preserve the Ellis hostname. proxy_set_header Host ellis.example.com<http://ellis.example.com>; } } On Mon, Apr 4, 2016 at 6:13 PM, Chris Elford (projectclearwater.org<http://projectclearwater.org>) <[email protected]<mailto:[email protected]>> wrote: Hi Adrian, It looks like nginx is trying to set up several servers with the same name. We haven’t seen this before. If you are seeing any problems as a result of this, I can try to help you dig into this. It would be useful to see: • The contents of /etc/nginx on your node. • The contents of /etc/clearwater on your node. • The output from running ifconfig -a on the node. Yours, Chris From: Clearwater [mailto:[email protected]<mailto:[email protected]>] On Behalf Of Adrian Schwartz Sent: 03 April 2016 15:07 To: [email protected]<mailto:[email protected]> Subject: Re: [Project Clearwater] nginx fail to start Hello all Ive uncomment the server_names_hash_bucket_size to 64 and it solved the issue. Any idea why? BTW this solves the issue Ive reported few days ago about ruuning the image over KVM Thanks Adrian On Sun, Apr 3, 2016 at 4:39 PM, Adrian Schwartz <[email protected]<mailto:[email protected]>> wrote: Hello all I'm using the all-in-one image ovf. When I copy the image to one of our computer, I see that nginx is hving trouble. What may be the root cause for such an error? Regards Adrian 2016/04/03 13:36:21 [warn] 17281#0: conflicting server name "10.0.0.4" on [::]:80, ignored 2016/04/03 13:36:21 [warn] 17281#0: conflicting server name "10.0.0.4" on [::]:80, ignored 2016/04/03 13:36:21 [emerg] 17281#0: could not build the server_names_hash, you should increase server_names_hash_bucket_size: 32
_______________________________________________ Clearwater mailing list [email protected] http://lists.projectclearwater.org/mailman/listinfo/clearwater_lists.projectclearwater.org
