Engine Daemons won't log on to the Director
Article ID: KB0086728
Updated On:
| Products | Versions |
|---|---|
| TIBCO DataSynapse GridServer | - |
| Not Applicable | - |
Description
Resolution:
Check the following in your Manager, Engine, and network configurations:
Manager Configuration
* Make sure your Manager is configured to log at the INFO log level or a finer level, so that you can see any successful logins that may occur. This is setting is on the Manager Configuration page, on the Manager tab, in the Logging section.
* Check that entries with your machine hostnames are Fully-Qualified Domain Names in the installation.properties file found in .../datasynapse/webapps/livecluster/WEB-INF/config/. The only time FQDNs are not used is on a Windows only network that does not have a functional DNS environment.
* If FQDNs are not present, reconfigure your server to use FQDNs on the Manager Reconfigure page on the Admin tab.
* Verify that your license allows the correct hostnames. If not, contact DataSynapse Support.
* Look for database exceptions and error messages in your Manager log. If any are found, see the database administration chapter of the GridServer Administration Guide.
Engine Configuration
* Confirm that your Engine is configured to point to your Director, and that the hostname used in the configuration works. On Windows, right-click on the DataSynapse toolbar icon, and select About to display the Director the Engine is using. On Linux or Solaris, use a base64 decoder to decode the x4088 Engine configuration file found in .../profiles/hostname/ and grep for Director. The values for DSPrimaryDirector and DSSecondaryDirector should be set correctly. Note that if these values are not set to a fully-qualified domain name and your network requires one to reach the Director, your Engine won't be able to log in. You can find a public domain base64 utility at http://www.fourmilab.ch/webtools/base64/
* If either entry is incorrect, stop your Engine, delete the x4088 and x4088.default files found in .../profiles/hostname/ and restart your Engine. The intranet.dat file in the root of the Engine installation specifies the Manager used to retrieve new x4088 files. If you have modified the Engine configuration, confirm that the Engine has downloaded the modified configuration by using a base64 decoder to view the contents of the x4088 Engine configuration file found in .../profiles/hostname/
* Also, look for a config.xml file in ../profiles/hostname/ If it is present, verify its contents. This file overrides configuration settings present in the normal Engine configuration. Remove the config.xml file if these override settings are not desired.
Network Configuration
* Ping the server machine from the Engine machine
* If the ping fails traceroute to the Manager machine from the Engine machine.
* If the ping succeeds, telnet to the Manager at the configured Manager port (default: 8000) from the Engine machine and enter GET / You should see something similar to the following:
[admin@myhost admin]$ telnet myhost 8000
Trying 192.168.8.34…
Connected to myhost.
Escape character is '^]'.
GET /
The URL has moved <a href="http://myhost.full.example.com:8000/livecluster/admin/control/base/login .jsp">here</a>
Connection closed by foreign host.
* When trying the ping and telnet, make sure to try a fully qualified host name or IP number to determine if there are any hostname resolution issues between the Engine and Manager.
Check the following in your Manager, Engine, and network configurations:
Manager Configuration
* Make sure your Manager is configured to log at the INFO log level or a finer level, so that you can see any successful logins that may occur. This is setting is on the Manager Configuration page, on the Manager tab, in the Logging section.
* Check that entries with your machine hostnames are Fully-Qualified Domain Names in the installation.properties file found in .../datasynapse/webapps/livecluster/WEB-INF/config/. The only time FQDNs are not used is on a Windows only network that does not have a functional DNS environment.
* If FQDNs are not present, reconfigure your server to use FQDNs on the Manager Reconfigure page on the Admin tab.
* Verify that your license allows the correct hostnames. If not, contact DataSynapse Support.
* Look for database exceptions and error messages in your Manager log. If any are found, see the database administration chapter of the GridServer Administration Guide.
Engine Configuration
* Confirm that your Engine is configured to point to your Director, and that the hostname used in the configuration works. On Windows, right-click on the DataSynapse toolbar icon, and select About to display the Director the Engine is using. On Linux or Solaris, use a base64 decoder to decode the x4088 Engine configuration file found in .../profiles/hostname/ and grep for Director. The values for DSPrimaryDirector and DSSecondaryDirector should be set correctly. Note that if these values are not set to a fully-qualified domain name and your network requires one to reach the Director, your Engine won't be able to log in. You can find a public domain base64 utility at http://www.fourmilab.ch/webtools/base64/
* If either entry is incorrect, stop your Engine, delete the x4088 and x4088.default files found in .../profiles/hostname/ and restart your Engine. The intranet.dat file in the root of the Engine installation specifies the Manager used to retrieve new x4088 files. If you have modified the Engine configuration, confirm that the Engine has downloaded the modified configuration by using a base64 decoder to view the contents of the x4088 Engine configuration file found in .../profiles/hostname/
* Also, look for a config.xml file in ../profiles/hostname/ If it is present, verify its contents. This file overrides configuration settings present in the normal Engine configuration. Remove the config.xml file if these override settings are not desired.
Network Configuration
* Ping the server machine from the Engine machine
* If the ping fails traceroute to the Manager machine from the Engine machine.
* If the ping succeeds, telnet to the Manager at the configured Manager port (default: 8000) from the Engine machine and enter GET / You should see something similar to the following:
[admin@myhost admin]$ telnet myhost 8000
Trying 192.168.8.34…
Connected to myhost.
Escape character is '^]'.
GET /
The URL has moved <a href="http://myhost.full.example.com:8000/livecluster/admin/control/base/login .jsp">here</a>
Connection closed by foreign host.
* When trying the ping and telnet, make sure to try a fully qualified host name or IP number to determine if there are any hostname resolution issues between the Engine and Manager.
Issue/Introduction
Engine Daemons won't log on to the Director
Feedback
Yes
No
Powered by
