Hi

I am using FAST ESP 5.3 SP5 searchengine patch02. It is a multinode installation and below is the list of modules running on each server

AdninServer
Non-Admin Server1 : Indexing Dispatcher, nctrl, qrserver, completionserver, indexer, search-1, topfdispatch
Non-Admin Server2 : nctrl, indexer, search-1

Also sometimes "non-admin" nodes information is not available under "system management" tab after doing refresh it becomes available.

All the modules on three servers are in running state. But we are getting below warnings in the log on Admin GUI.

configserver   <admin server>>   16005   systemmsg   Module (Clarity) at <admin server>>:16098 is not responding
configserver   <admin server>>   16005   systemmsg   Module (Search Dispatcher) at <<Non-Admin RTS Node1>>:15102 is not responding  
configserver   <admin server>>   16005   systemmsg   Module (WebAnalyzer) at <admin server>>:16700 is not responding  
configserver   <admin server>>   16005   systemmsg   Module (Search Engine) at <<Non-Admin RTS Node1>>:15674 is not responding  
configserver   <admin server>>   16005   systemmsg   Module (Search Engine) at <<Non-Admin RTS Node2>>:15703 is not responding  
configserver   <admin server>>   16005   systemmsg   Module (ProcessorServer) at <admin server>>:16210 is not responding  
configserver   <admin server>>   16005   systemmsg   Module (ProcessorServer) at <admin server>>:16220 is not responding  
configserver   <admin server>>   16005   systemmsg   Module (ProcessorServer) at <admin server>>:16200 is not responding  
configserver   <admin server>>   16005   systemmsg   Module (NodeControl) at <admin server>>:16015 is not responding  
configserver   <admin server>>   16005   systemmsg   Module (NodeControl) at <<Non-Admin RTS Node1>>:16015 is not responding  
configserver   <admin server>>   16005   systemmsg   Module (NodeControl) at <<Non-Admin RTS Node2>>:16015 is not responding  
configserver   <admin server>>   16005   systemmsg   Module  (NodeControl) at <<Non-Admin RTS Node1>>:16015 is not responding  
configserver   <admin server>>   16005   systemmsg   Module  (NodeControl) at <<Non-Admin RTS Node2>>:16015 is not responding
configserver   <admin server>>   16005   systemmsg   Module (fdmworker) at <admin server>>:16722 is not responding
configserver   <admin server>>   16005   systemmsg   Module (WebAnalyzer) at <admin server>>:16700 is not responding
configserver   <admin server>>   16005   systemmsg   Module (WaLinkStorerReceiver) at <admin server>>:16710 is not responding  

procserver   <admin server>>   16215   systemmsg   RegisterCapabilities failed: pyxmlrpcClient.Fault: 1: ConfigServerExceptions.ModuleError: No ProcessorServer registered

at '<admin server>>:16215' (in src/ConfigServerConfig.py:RegisterProcessorClasses line 1901) 

logtransformer   <admin server>>   16010   systemmsg   Unable to find the QRServer node for <<Non-Admin RTS Node1>> Is the QRServer registered?
logtransformer   <admin server>>   16010   systemmsg   An exception was thrown whilst downloading files

Thanks

Ashish

Components to register with configserver upon start up. If you are getting "modules not responding" warnings and nctrl shows components running, I would recommend checking any network/communication related issues between the three nodes.

Hi Ashish,

Is this running on Windows or Linux?  For either platform, I would verify that the admin node can successfully perform a forward nslookup (nslookup FQDN) and reverse nslookup (nslookup IP_address) to each node, and vice versa.  Also, if ESP is running on Windows, I would verify the below two items:

-       Confirm that the Fast processes have been configured as excluded in the antivirus exclusions as well, as outlined in http://technet.microsoft.com/en-us/library/ff381239.aspx in the section entitled Configure anti-virus configuration

-       Confirm that you have disabled TCP Task and IP offloading on all ESP servers, per KB article http://support.microsoft.com/kb/2570111?

Let us know the results of the above, and if you have any questions.

Thanks!

Rob Vazzana | Sr Support Escalation Engineer | US Customer Service & Support

Customer Service   & Support                            Microsoft| Services

• Proposed as answer by Rob Va - MSFTMicrosoft employee, Moderator Friday, November 15, 2013 9:05 PM

I have the same issue on a Linux platform. I added a new row in several FAST clusters. Each cluster has 3 nodes per row. I have this one node in one of the clusters that won't register. I checked nslookup and I'm fine. Our network team checked the network and it's fine. The 2 other nodes are fine as well. I've reinstalled and reconfigured 3 times and am at a loss.

What I do notice on all 9 nodes is that nctrl stop times out and I have to manually stop those tasks.

Thanks for any advice :-)

Gina

Hi Gina and Ashish,

The behavior you describe usually points to some type of timeout.  If your network team was not able to identify the issue, I would recommend opening a new case with our Technical Support team to further investigate the issue and review the environment.

Thanks!

Rob Vazzana | Sr Support Escalation Engineer | US Customer Service & Support

Customer Service   & Support                            Microsoft| Services

• Proposed as answer by Rob Va - MSFTMicrosoft employee, Moderator Tuesday, December 31, 2013 5:02 PM