HI,
I would like to know all the possible reasons which cause CMS Service go down. I searched in BOB, could find few basic reasons that cause service go down. But I would like check whether I am missing any key reasons.
My environment is like below.
OS: Win3K Server
BO: BOXIR1 SP1
DB: Oracle 10G R1
APL-Server: IBM Http server6.0
Here’s my 2 cents: the most common reason the CMS goes down, in my experience is that the database goes down. The CMS will check the system audit database at regular intervals. If the db goes down for a backup or any other reason, the CMS will try to connect and then give up after 10 minutes or so. Then you start getting user calls…
You can check your system log in windows and you should, with a little digging, find out if the CMS is going down at similar times as the database. Some one posted a great script that restarts the cms every morning. You can also make the cms start dependent on the database.
Hello, do you know where the script is located. Because I am having the same trouble my CMS is stopping all the time now just last week I had to restart my CMS service and today also. Would you say I need to reinstall the CMS again? I am not sure why this is happening… hmmmm help!!!
Steve, sorry for delay., thanks for the useful info.
By the way, I am just wondering what all other (main) reasons to let CMS go down other than DB service?
It would be great if you can let me know the link where I can find the script (to start CMS everyday), you were referring in your mail.
Right now we are implementing BusinessObjects in our enterprise.
There is only one CMS we have and people access Universe and reports from that CMS.
If the CMS goes down for some reason I am sure that they will not have access. How do I make sure that should have access to Unvierse and reports eventhough the CMS goes down.
yeah, you can create cluster environment with multipal CMS. so if one CMS is down that load can be transfer to second one. but reason for CMS down time is Data Base connectivity. then both CMS will be down.
Yeah you are right.
For example accidently the CMS server shutdown during the night.
The reports and the universe should work as usual from the secondary CMS. How can I have the similar reports and Universe in the secondary CMS? The most important things how the schedule takes place for instance?
Info seeker. I found the script on BOB, but I can’t find it right now…
I wrote up a document for this, so I will upload. Ignore any specific server name references or any SQL Server references Daily Batch File to restart the CMS.doc (78.0 KB)
I’d similar problem on Unix environment - the CMS wend down unexpectedly after the start up. In my case the DB part has been changed (ip, instance name and DB client moved to another dir).
Even after the cmsdbsetup.sh configured to the right DB, the cms process appeared and die.
Thanks to SteveBickerton, his idea of DB reason directs me to check the relation between BO and Oracle.
As result, have a look the .profile of the bob user on the OS and found that ORACLE_HOME and LD_LIBRARY_PATH were pointed to the old version.
I too have a similar problem. The Web_IntelligenceReportServer is getting stopped unexpected.
Can anyone please throw light on how to find the status of Web_IntelligenceReportServer whether it is down or up without manully checking in CMC/CCM. I mean a mail has to be triggered indicating the status of the server or through some other means…
Hi Krystal, the Webi server is a completely different piece of work, so I would suggest creating a new thread on this one.
However, functionality like the one you are after is pretty much 3rd party stuff, but I have seen a script that one of the posters here, CBussel, suggested here: Script to restart CMS server or orhter servers in XIr2
Back to the main topic - the generic reason the CMS will go down is when it’s middleware cannot contact the DB anymore.
So when the oracle client software cannot contact the Oracle DB, the CMS goes down after 3 tries. (I am only talking about the System DB here, not your reporting DB of course).
This could be when the DB is down, when there is a network issue (like the DB getting a new IP adddress and the middleware cannot connect), or when the DB is upgraded to a version that the middleware cannot connect to.
The CMS will in theory also go down when there is database corruption but I have not experienced this.