I am new to WebSphere.    We are using WebSphere Application Server 7.0 and JMS Queues.    The server has large websphere file.  It keeps growing.      The file is found in this path.    .../databases/com.ibm.ws.sib/mlwlms20Node01.wmserver-wmserver_ILS_QUEUE_CONNECTOR.sib.84A49A866203A375/seg0.    This directory has 98 files.  The file names are c*.dat.    The big boy is c450.dat.   The size is 67 GB.    Too big to view the file.   I was able to cat the file.   I found XML records in the file.    

What is this big boy?   Can it be purge or deleted?     I need to do something before this file system runs out of space.   

Hi denny,

  The path is the default repository of SIB Bus(es) message/transactions logs stores.
 
  You have configured one SIB Bus and a ME (Message Engine) on it, JMS Queues are in this ME.  
 
  Check your datastores configuration to know the max limit of the ME store.
 
  Check through adminconsole:
  Service Integration - Buses - BUS_NAME - Message Engines - YOUR_ME - datastores
 
  JMS Queues have persistent information? Maybe your application is "storing" useless information.
 
  DON'T DELETE IT, it have your JMS Queues information.
 
  Hope this helps.
 
regards

python is only tool I have to display jms queues attributes.    I found sample script to list queues and depth.     

# Print Queue Stats
print "Print Queue Stats"
scope = AdminConfig.getid('/Node:%s/Server:%s' % (nodeName, serverName))allqueues = AdminTask.listSIBJMSQueues(scope).split()
print "sequence number, queue name, current queue depth"
i = 0
for queue in allqueues:
    i = i + 1
    identifier = AdminControl.getAttribute(queue,'identifier')
    depth = AdminControl.getAttribute(queue,'depth')
    print i, identifier, depth

Results..................

Print Queue Stats
sequence number, queue name, current queue depth
WASX7017E: Exception received while running file "/apps/scope/home/wasadmin/scripts/test12.py"; exception informati
on: com.ibm.ws.scripting.ScriptingException: WASX7025E: Error found in String "WebSphere:jms.queue.INVENTORY_INBOUN
D(cells/WM_JCKQCell/nodes/mlwlms20Node01/servers/wmserver|resources.xml#J2CAdminObject_1369057142046)"; cannot crea
te ObjectName.

Why it script failing with message WASX7017E?    The value queue has valid jms queue name.    It it because variable identfier and depth values are null?  What am I missing?

Hi denny,

  I think that you are not getting the correct Objectname....
 
  You can check Depth and Messages in runtime from adminconsole.
 
    Service integration > Buses > YOURBUS > Messaging engines > YOURME > In runtime tab you can check Queues
    
  I have used with Version 6.1 the next utility:

    Service Integration Bus Performance
    www.ibm.com/developerworks/community/gro...
    
  I'm testing with version 7. I tell you if work.
 
  Hope this helps.
 
Kind regards

[quote author=91692909 post=535080853]Hi denny,

  I think that you are not getting the correct Objectname....
 
  You can check Depth and Messages in runtime from adminconsole.
 
    Service integration > Buses > YOURBUS > Messaging engines > YOURME > In runtime tab you can check Queues
    
  I have used with Version 6.1 the next utility:

    Service Integration Bus Performance
    www.ibm.com/developerworks/community/gro...
    
  I'm testing with version 7. I tell you if work.
 
  Hope this helps.
 
Kind regards[/quote]

I can confirm that the tool works on WAS 7.0.0.17

Regards

Sorry it took me so long to post response.   We found solution to clear our large WebSphere files.    The process took me number of steps to identify the problem.   

Our company purchased new Warehousing Application System.     JMS queues are used to send data from our mainframe system to new Warehouse System.     Our JMS Queues are configured on AIX 7 and WebSphere 8.0.0.2.  JMS is using the SIB (System Integration Bus).     I found scripts to query the JMS Queue depth, but I had problems putting the pieces together.     These were the steps.  

1. Enable PMI message process counters


2. Created Jython scripts to monitoring JMS queue depth


3. Installed Hermes to browse messages in the JMS queues.

Message processor counters.   The service integration bus messages processor counters are part of the performance monitoring infrastructure (PMI), which provides server-side monitoring and client-side API to retrieve performance information.      These counters need to be enabled to collect message queue depth.  

Signon to the WebSphere Console.    Enable PMI to collection message queue depth

Performance Monitoring Infrastructure (PMI) > wmserver > Custom monitoring level

-         SIB Service

• SIB Messaging Engines
  
  
  
  • *
    
    
    
    • Destinations
      
      
      
      • Queues
        
        
        
        • QueueStats.AvailableMessageCount
        
        
        • QueueStats.UnavailableMessageCount   
        
        
        
        
      
      
      
      
    
    
    
    
  
  
  
  

My first jython scripts failed.  I was using the wrong port number.   My scripts work using SOAP_CONNECTOR_ADDRESS port number.    Found this jython script to list the ports being used by WebSphere. 

printports.sh

/manh/3rdparty/swtools/WebSphere8/AppServer/bin/wsadmin.sh -conntype SOAP -host mlwlms20.woolworth.com  -port 11035 -username system -password password -lang jython -f /apps/scope/wms/utilities/scripts/printports.py

printports.py

formatString = '''

        Node: %s

        Server: %s

 

 Port |  EndPoint Name

------+---------------'''

for node in AdminConfig.list( 'Node' ).splitlines() :

        nodeName = AdminConfig.showAttribute( node, 'name' )

        for se in AdminConfig.list( 'ServerEntry', node).splitlines() :

                serverName = AdminConfig.showAttribute( se, 'serverName' )

                print formatString % (nodeName, serverName )

                seps = AdminConfig.showAttribute( se, 'specialEndpoints' )

                for sep in seps[ 1:-1 ].split( ' ' ) :

                        ep = AdminConfig.showAttribute(sep,'endPoint')

                        port = AdminConfig.showAttribute(ep,'port')

                        name = AdminConfig.showAttribute(sep,'endPointName')

                        print '%5d | %s' % (port, name)

                print

Print Port Results

        Node: mlwlms20Node02

        Server: server1

 

 Port |  EndPoint Name

------+---------------

11034 | BOOTSTRAP_ADDRESS

11035 | SOAP_CONNECTOR_ADDRESS

11040 | ORB_LISTENER_ADDRESS

11037 | SAS_SSL_SERVERAUTH_LISTENER_ADDRESS

11038 | CSIV2_SSL_SERVERAUTH_LISTENER_ADDRESS

11039 | CSIV2_SSL_MUTUALAUTH_LISTENER_ADDRESS

11031 | WC_adminhost

11030 | WC_defaulthost

11041 | DCS_UNICAST_ADDRESS

11033 | WC_adminhost_secure

11032 | WC_defaulthost_secure

11046 | SIP_DEFAULTHOST

11047 | SIP_DEFAULTHOST_SECURE

11042 | SIB_ENDPOINT_ADDRESS

11043 | SIB_ENDPOINT_SECURE_ADDRESS

11044 | SIB_MQ_ENDPOINT_ADDRESS

11045 | SIB_MQ_ENDPOINT_SECURE_ADDRESS

11036 | IPC_CONNECTOR_ADDRESS

 

        Node: mlwlms20Node02

        Server: wmserver

 

 Port |  EndPoint Name

------+---------------

11004 | BOOTSTRAP_ADDRESS

11005 | SOAP_CONNECTOR_ADDRESS

11014 | ORB_LISTENER_ADDRESS

11010 | SAS_SSL_SERVERAUTH_LISTENER_ADDRESS

11011 | CSIV2_SSL_SERVERAUTH_LISTENER_ADDRESS

11012 | CSIV2_SSL_MUTUALAUTH_LISTENER_ADDRESS

11002 | WC_adminhost

11000 | WC_defaulthost

11013 | DCS_UNICAST_ADDRESS

11003 | WC_adminhost_secure

11001 | WC_defaulthost_secure

11015 | SIP_DEFAULTHOST

11016 | SIP_DEFAULTHOST_SECURE

11017 | IPC_CONNECTOR_ADDRESS

11006 | SIB_ENDPOINT_ADDRESS

11007 | SIB_ENDPOINT_SECURE_ADDRESS

11008 | SIB_MQ_ENDPOINT_ADDRESS

11009 | SIB_MQ_ENDPOINT_SECURE_ADDRESS

These scripts will list the JMS queues

alljmsqueues.sh

/manh/3rdparty/swtools/WebSphere8/AppServer/bin/wsadmin.sh -conntype SOAP -host mlwlms20.woolworth.com  -port 11005 -username system -password password -lang jython –tracefile  /apps/scope/wms/utilities/scripts/logs/alljmsqueues_trace.log -appendtrace false  -f /apps/scope/wms/utilities/scripts/alljmsqueues.py  > /apps/scope/wms/utilities/scripts/logs/alljmsqueues.log

Script alljmsqueues.py

import time

import os

print "Start alljmsqueues.sh script      ",time.ctime()

 

formatString = '''

   Queue                                           | Queue Depth |       State  |  Threshold

   ------------------------------------------------+-------------+--------------+-----------'''

 

flag = 'N'

allqueues = AdminControl.queryNames('WebSphere:*,type=SIBQueuePoint').split('\n')

print formatString

for queue in allqueues:

    identifier = AdminControl.getAttribute(queue,'identifier')

    depth = AdminControl.getAttribute(queue,'depth')

    state = AdminControl.getAttribute(queue,'state')

    highMessageThreshold = AdminControl.getAttribute(queue,'highMessageThreshold')

    print '%50s | %12s| %12s | %10s' % (identifier, depth, state, highMessageThreshold)

    if depth '0':

        flag = 'Y'

    continue

 

if flag == 'Y':

    cmd = 'mailx -s ' + '"QA All JMS Queues Depth"' + 'address@emailserver < /apps/scope/wms/utilities/scripts/logs/alljmsqueues.log'

   

    os.system(cmd)

Partial results of alljmsqueues.py

WASX7209I: Connected to process "wmserver" on node mlwlms20Node02 using SOAP connector;  The type of process is: UnManagedProcess

Start alljmsqueues.sh script       Wed Oct 30 07:00:17 2013

   Queue                                                          | Queue Depth |       State        | Threshold

   ---------------------------------------------------+----------------+----------------+ -------------

    JMS.QUEUE.MA.LM.MASTER.IN         |                     0|       ACTIVE |      50000

    JMS.QUEUE.MA.LM.MASTER.OUT     |                     0|       ACTIVE |      50000

    JMS.QUEUE.MA.LM.RESPONSE.OUT |                     0|       ACTIVE |      50000

    JMS.QUEUE.MA.LM.TRAN.IN               |                    0|       ACTIVE |      50000

    JMS.QUEUE.MA.LM.TRAN.OUT           |                    0|       ACTIVE |      50000

    JMS.QUEUE.MA.WM.MASTER.IN        |                    0|       ACTIVE |      50000

    JMS.QUEUE.MA.WM.MASTER.OUT    |                    1|       ACTIVE |      50000

After all of this we did not find queue with large number of messages.   There was queue with 30,000+ messages.   We did not have utility to browse the messages in the JMS queue.  We downloaded and installed Hermes JMS Browser 1.15.      After the install Hermes fails to start.   FYI there is typo in the hermes.bat file.     Change no to not.  

            :typeHermesCfg

            If no exist      

            Change to

            If not exist

 

Hermes 1.15 started.   Successfully configure session.    Configuring Hermes was not easy for me.     I have screen prints available.   Hermes JMS Browser was use to looks at problem JMS queue.   The queue was not the problem.   The Jython scripts and Hermes gave us enough information to contact our vendor.   The vendor gave us instruction to clear the big file.     The solution was simple.     

1. Login to App server and shutdown the application
2. Go to the WM JMS DB directory (Ex for QA - $cd /apps/scope/wms/profile-root/WM_JCKQ/databases/com.ibm.ws.sib )
3. Move all the contents of the above directory to a backup location (empty the com.ibm.ws.sib directory). This backup can be deleted later once changes are tested and verified.
4. Now start up the application and test.    The restart of the system will rebuild the WebSphere Files.   

The vendor Manhattan Assoicates said the large file (69GB) found  in ../com.ibm.ws.sib directory were jms xml records that failed to update.   The problem XML records have been fixed.     We no longer have the large file.
Once per week this file is clear.  I am still WebSphere newbie.   Not sure if this solution will work for everybody.    

 

 

 

 

Congratulations denny!!! great and hard work

glad to hear that the problem is solved.

Regards