locked
Configserver is failing to come up RRS feed

  • Question

  • [2013-09-16 09:22:54.439] ERROR      systemmsg Failed to register with ConfigServer: pyxmlrpcClient.Fault: [Errno 111] Connection refused
    Traceback (most recent call last):
      File "src/CacheManager.py", line 198, in ?
      File "src/cache_manager.py", line 514, in __init__
      File "/apps/esp/lib/python2.3/encodings/__init__.py", line 2374, in _bind
       
      File "/apps/esp/lib/python2.3/encodings/__init__.py", line 372, in _ns_bind
       
      File "/apps/esp/lib/python2.3/encodings/__init__.py", line 328, in _create_subcontext
       
    middleware.fatal_ex: Unable to create subcontext with id: datamanager
    Monday, September 16, 2013 4:30 PM

All replies

  • [2013-09-16 09:10:08.174] CRITICAL   systemmsg Failed to start ConfigServer: fatal_ex: Unable to create subcontext with id: fds
    [2013-09-16 09:11:16.103] VERBOSE    systemmsg Starting Config Server on port 16005 (PID: 18652)
    [2013-09-16 09:11:16.359] CRITICAL   systemmsg Failed to start ConfigServer: fatal_ex: Unable to create subcontext with id: fds
    [2013-09-16 09:24:50.466] VERBOSE    systemmsg Starting Config Server on port 16005 (PID: 28071)
    [2013-09-16 09:24:50.723] CRITICAL   systemmsg Failed to start ConfigServer: fatal_ex: Unable to create subcontext with id: fds
    Monday, September 16, 2013 4:32 PM
  • Was there any changes to the system?
    May be there was issues with hard reboots or unexpected stops when there are hung processes still running(indexer or others) and upon restart, the essentially become zombie processes.  So we always recommend making sure there are no hung processes in the Task Manager before a restart.

    Is it possible to reboot the servers and restart ESP?
    Another thing I would recommend doing, while all the ESP processes are down, is to rebuild the omniorb(Nameservice).  Just go to %FASTSEARCH\var\run\nameservice folder and delete the contents of it.

    Monday, September 16, 2013 5:32 PM
  • We tried those steps it is not helping . There was no specific change to system.
    Monday, September 16, 2013 7:26 PM
  • Hello,

    I would first look at communication issues as the cause of this issue, and we can see the message indicates Connection refused.  Is your ESP environment running on Windows or Linux?  If it running on Windows, have you verified that TCP IP and Task offloading has been disabled?  Can you confirm there are no firewalls between the nodes, or other such security software that could be blocking the ports?

    Can you also confirm that you can perform an nslookup forward (i.e.: nslookup server.domain.com) and reverse (i.e.: nslookup IP_address) from the adminnode (where configserer is running) to the other ESP nodes, and from the other nodes, to the adminnode as well?  Let us know your results.

    Thanks!

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

    Customer Service   & Support                            Microsoft| Services

    • Proposed as answer by Rob Va - MSFT Monday, September 23, 2013 9:20 PM
    Monday, September 23, 2013 8:43 PM