Ok, now it is officially published. With these instructions an update of IBM Connections 4.5 including Connections COntent Manager to CR1 should no longer be a big problem 🙂
IBM Connections 4.5 CR1 Upgrade: Missing scripts
If you are missing the files “updateJVMArgs.py” and “ccmUpdate.sh” resp “ccmUpdate.bat” aftre you installed IBM Connections 4.5 CR1, then you might have forgotten one step.
After unpacking the ZIP file with all CR1 fixes you find one additional ZIP file “45-CR1-CCM-OPTIONAL-UPDATE-UNZIP-AND-PLACE-JAR-IN-FIXES-DIRECTORY.zip” which also need to be unzipped.
The resulting fix “LO75060-IC4500-CR01-CCM.jar” also needs to be installed with the Connections update manager.
Afterwards the missing files should be available in the ” /opt/IBM/Connections/addons/ccm/ccmDomainTool/” directory.
IBM Connections: Error running “setupCmdLine.sh”
While running the setupCmdLine.sh script to set your Websphere environment variables you get the error: “dirname: invalid option — ‘b'”.
This might happen if you start the script (as described in the documentation) from the directory “/opt/IBM/WebSphere/AppServer/bin/”. If you instead run it from the Deployment manager profile, then it should work:
. /opt/IBM/WebSphere/AppServer/profiles/Dmgr01/bin/setupCmdLine.sh
Do not forget the leading “. ” in front of the command!
You may check that the variables are set correctly by running
echo $WAS_HOME
This variable should now be set to you WAS install directory (e.g. “/opt/IBM/WebSphere/AppServer”)
Ausgemistet …
Ich hatte es ja schon ein paar Mal angekündigt. Heute habe ich damit begonnen, alle von mir nicht aktiv genutzten Accounts im Internet zu löschen. Gar nicht so einfach, da ich mich ja fast überall mal angemeldet hatte … und sei es nur um es zu testen 😉 Es ist auch noch ne Menge zu tun, aber ein Anfang ist gemacht.
Insbesondere meine Profile in Social Networks wie WKW, Stayfriends etc. sind nun nicht mehr vorhanden (zumindest nicht mehr zugänglich, ob die Daten wirklich gelöscht wurden, weiss ja kein Mensch :-)).
Wer möchte, kann mir gerne einen Request bei Facebook senden oder meine öffentlichen Beiträge dort abbonieren.
See you there …
IBM Connections 4.5: Error while accessing IBM Content Manager Library
A customer reported an error while accessing an IBM Content Manager Library within communities. The user got the error message “We are unable to process your request. The library you are trying to access is not available. You can try to load the library again. If this doesn’t work, the library may have been deleted or modified to no longer be visible to you.”
Within the Websphere logs you see java exceptions. The most important ones were:
“CQL5936: The server is currently unavailable. The configured object store, ICObjectStore, was not found on the P8 domain ICDomain at corbaloc:rir:/cell/clusters/Cluster1/CCM/Engine. Contact your system administrator for assistance. ”
and
“FNRCA0031E: API_UNABLE_TO_USE_CONNECTION: The URI for server communication cannot be determined from the connection object Class=com.filenet.apiimpl.core.ConnectionImpl URI=corbaloc:rir:/cell/clusters/Cluster1/CCM/Engine Parameters={}.”
Solution:
There might be two possible configuration problems:
- Make sure that the name of the SSO token is correct.In Websphere Solution Console go to “Security -> Global Security -> Web and SIP Security ->Single Sign-on (SSO)”.
Check “LTPA V1 cookie name” and “LTPA V2 cookie name”. It must be “LtpaToken” resp. “LtpaToken2”. No other names are allowed and the entry is case sensitive. So make really sure, it looks like that.
- Make sure the Generic JVM properties of all servers where the Filenet application is running is set to the same values on all nodes. It can happen that on the secondary nodes some statements are missing.In Websphere Solution Console go to “Servers -> Server Types -> Websphere Application Servers -> <name of the server where Content Manager is running, e.g. “Cluster1_server1″>”. Within the server go to “Server Infrastructure -> Java and Process Management -> Process Definiton -> Additional Properties -> Java Virtual Machines” and check the value in “Generic JVM arguments”.
It should be the same for this server on all nodes and should look like
” -Xgcpolicy:gencon -Djava.awt.headless=true -Xsoftrefthreshold16 -Xgcpolicy:gencon -Djava.awt.headless=true -DFileNet.EJB.ContextProperties=java.naming.provider.url=! -Dibm.filenet.security.vmmProvider.waltzImpl=true -DenableWaltzIdConversion=true -Dcom.ibm.connections.directory.services.j2ee.security.principal=uid”
If the statements are missing, add it. Make sure: These statements are only on the server where IBM Content Manager / Filenet applications are running. So you only need to add it there.
After doing a resync of all nodes and restart all servers, the issue should be solved.
ST 8.5.x: Updated security certificate for Push Notifications (iOS)
Please be aware that you neeed to update the security certificate for iOS push notifications on your IBM Sametime proxy servers before July 18th. Otherwise push notifications to your iOS devices will stop working. Instructions how to download and install the new certificates can be found in this technote.
Tivoli Directory Integrator: Change UI and Log Language
If you install Tivoli Directory Integrator on a Windows machine where you have set the regional settings to another language than English, the UI and all log messages of TDI appear in the local language instead of English. If you want to change that so that all messages appear in English again you need to to the following steps:
- Open “<TDI Program Dir>\ibmditk.bat” in a text editor
- Look for the two lines where “eclipsece\miadmin” is called and add “-Duser.language=en” at the end of these lines. They will then look similar to the following lines:
[codesyntax lang=”dos”]
"%TDI_HOME_DIR%\ce\eclipsece\miadmin" -tdishutdown -noSplash %* -vm "%TDI_JAVAW_PROGRAM%" -vmargs -Dcom.ibm.di.loader.IDILoader.path="%TDI_HOME_DIR%" -Duser.language=en ... start /B "Tivoli Directory Integrator" "%TDI_HOME_DIR%\ce\eclipsece\miadmin" %* -vm "%TDI_JAVAW_PROGRAM%" -vmargs -Dcom.ibm.di.loader.IDILoader.path="%TDI_HOME_DIR%" -Duser.language=en
[/codesyntax]
- Now open “<TDI Program Dir>\ibmdisrv.bat” in a text editor
- Look for the line starting with “SET LOG4J=” and add ” -Duser.language=en” at the end of this line so it will look similar to[codesyntax lang=”dos”]
set LOG_4J=-Dlog4j.configuration="file:etc\log4j.properties" -Duser.language=en
[/codesyntax]
If you now restart TDI the UI and all messages in the log files will be in English.
IBM Sametime Meeting Help not accessable in Notes Sidebar
A customer was not able to access the online help for Sametime Meeting within the Notes embedded Sametime client. Neither the “Help Topics” link in the Meeting sideshelf nor the question mark button within a Meeting room did work. The help window appeared but it was empty.
The problem was because the customer had defined a proxy server within the Notes location document but did not exclude “localhost”. The help system tries to contact “http://localhost/…” which is not accessable via the proxy server. After I added “localhost” to the exclusion list, the help pages appeared again as expected.
IBM Connections 4.5 Cumulative Refresh 1 (CR1) just arrived
You can download this fixpack from here.
More information about how to install fixpacks for IBM Connections can be found here.
You will find a list of the fixes via this link.
If you are using IBM Connections Content Manager then you will need to update Filenet to 5.2.0 FixPack 1 (FP1) as well. More information here.
Connectivity Problem with DB2 V10.1 on Suse Linux
Today I tried to install IBM DB2 V10.1 on a Suse Enterprise Linux 10.1 SP2. However, although the installation went smoothly with no errors, I was unable to connect to the DB2 server via TCP/IP. The install program told me that the server is using port 50001 but there was no connect and a “netstat -an | grep LISTEN” showed that there was no process listening on port 50001.
I checked /etc/services and there was no entry for port 50001 as well.
Running the command “db2 get dbm cfg | grep “(SVCENAME)” showed that the service name did not seem to be correctly configured as the entry was empty.
So, to solve the problem I ran the following command: “db2 update dbm cfg using svcename db2c_db2inst1” to correctly configre the service name and also added the line “db2c_db2inst1 50001/tcp” to the file /etc/services.
Afterwards I stopped DB2 (“db2stop”) and restarted it again (“db2start”).
A quick check with netstat now shoed that there was a process listening on port 50001 now and I was also able to connect via TCP/IP to the database.
However I still have no glue what caused the problem as it should normally work without these steps.
Update:
Check also (with “db2set -all”) if the variable “DB2COMM=TCPIP” is set correctly. If not, set it with “db2set DB2COMM=TCPIP” and restart DB2 afterwards.