TM1 Forum

Hi,

I have a client running 10.2.2 FP6 IF1, which was installed almost 2 weeks weeks ago and it's been running without incident. They have a chore running 'n Savedataall every 15 minutes past the hour and I can see that the last time it ran was at 1:15 AM morning. According to the tm1server.log the server was started just after 7AM but has since stopped a few times without any entry in the log file.

In Event viewer I can see the normal "The IBM Cognos TM1 Server - SBV service terminated unexpectedly. It has done this 1/2/3 time(s)." every time that the server stopped but couldn't find anything entry between 1:15AM and 2:15AM.

The host server was restarted this morning and there is enough disk space available on the server. Below are entries between TM1 server is ready and Data directory for 5 of the unexpected stops:


6000 [] INFO 2017-04-10 05:09:21.148 TM1.Server TM1 Server is ready
4344 [] INFO 2017-04-10 05:15:36.060 TM1.Chore Chore "Save data" executed by scheduler
4344 [] INFO 2017-04-10 05:15:36.060 TM1.Process Process SaveData executed by chore Save data
4344 [] INFO 2017-04-10 05:15:36.091 TM1.Process Process "SaveData": finished executing normally, elapsed time 0.03 seconds
4344 [] INFO 2017-04-10 05:15:36.091 TM1.Chore Chore "Save data" time = 0.03 seconds
4344 [] INFO 2017-04-10 05:15:36.091 TM1.Chore Chore "Save data" finished executing
5992 [6] INFO 2017-04-10 05:48:54.765 TM1.Process Process "Populate Cancelled Orders" executed by user "Eddie F"
5992 [6] INFO 2017-04-10 05:48:56.218 TM1.Process Process "Populate Cancelled Orders": finished executing normally, elapsed time 1.45 seconds
5512 [] INFO 2017-04-10 06:31:11.317 TM1.Server Data Directory: e:\tm1\sbv\cubes - sbv



5512 [] INFO 2017-04-10 06:38:17.010 TM1.Server TM1 Server is ready
3100 [6] INFO 2017-04-10 06:43:08.233 TM1.Process Process "Populate ATM Optimization" executed by user "Tebogo"
3100 [6] ERROR 2017-04-10 06:43:10.358 TM1.Process Process "Populate ATM Optimization": finished executing with errors. Error file: <TM1ProcessError_20170410064310_31203100_Populate ATM Optimization.log> : Invalid key: Dimension Name: "Centre", Element Name (Key): "NCS"
1532 [] INFO 2017-04-10 07:02:05.181 TM1.Server Data Directory: e:\tm1\sbv\cubes - sbv



1532 [] INFO 2017-04-10 07:11:25.297 TM1.Server TM1 Server is ready
2756 [] INFO 2017-04-10 07:15:05.498 TM1.Server Data Directory: e:\tm1\sbv\cubes - sbv



2756 [] INFO 2017-04-10 07:22:25.767 TM1.Server TM1 Server is ready
4516 [5] INFO 2017-04-10 07:28:11.678 TM1.Process Process "Populate ATM Optimization" executed by user "Tebogo"
4516 [5] ERROR 2017-04-10 07:28:14.006 TM1.Process Process "Populate ATM Optimization": finished executing with errors. Error file: <TM1ProcessError_20170410072812_03704516_Populate ATM Optimization.log>
5076 [] INFO 2017-04-10 07:47:47.833 TM1.Server Data Directory: e:\tm1\sbv\cubes - sbv




5076 [] INFO 2017-04-10 07:55:12.155 TM1.Server TM1 Server is ready
1064 [] INFO 2017-04-10 08:13:46.447 TM1.Server Data Directory: e:\tm1\sbv\cubes - sbv


Does anybody have any suggestions on what is happening here?

Rgds,

George

Presumably you've checked available disk space?

Disk space is fine

Is the SaveDataAll in the Prolog or Epilog?

Ironically, just saw this in the 10.2.2 FP6 IF1 fix list - PI68901TM1 Server Crash When Running a Savedataall - guess it is not fixed for you!

I don't believe this has anything to do with SaveDataAll, since it even crashed 5 minutes after starting without any chore or process running. I merely mentioned that since it enabled me to determine when the crash happened (and no recorded event in the Event Log).

It may be worthwhile creating/changing the tm1s-log.properties to put the log into debug. This will obviously log a lot of info but may log something just before the crash. Keep an eye on disk space.

Given that you are running on Windows have you looked in the Windows Application event logs to see if the process dying is leaving any trace in there?

"Programs -> Window Administrative Tools -> Event Viewer" or something similar.

I had the same problem. We install version 10.2.2 FP6 a couple of weeks ago and when we move it to production server, it started to fall dow after save data.We saw that FP7 resolve problems with "save data" which makes falling down the server. We install it on thursday,we havent had problem since then.

Excuse my english

Thanks for all the suggestions, but in the end it was a circular reference because of an attribute.

I worked with IBM and he couldn't get the server to fall over on FP7 or after downgrading back to FP6 IF1. In the end it was just a case of removing cubes and eventually removing and adding the rules & feeders back.

maybe you use empty view in ti process ? I have this problem some times when use mtq=-1 or >1
http://www.tm1forum.com/viewtopic.php?t=12634