That means we have 50 gb for the rest. Since the 26th of December we have had several crashes in the ReportingServicesservices.exe process. Sometimes it just crash and restarts and sometimes reports starting to run slowly and in the log file you can see lot of requests are being thrown away and it starts to complain about data that is not right at different positions. Example:
processing!ReportServer_0-1!8c48!02/13/2013-09:31:29:: e ERROR: Throwing Microsoft.ReportingServices.ReportProcessing.ReportProcessingException: , Microsoft.ReportingServices.ReportProcessing.ReportProcessingException: There is no data for the field at position 15.;
After a while the process seems to stop taking care of more requests and finally it crash. In December and January I thought it had to do with monthly running reports because the issue occurred in the end of the month more frequently, but in February the crashes have occurred completely differently, several times it has crashed exactly 5 pm. I have looked in the execution log views an tables, analyzed SQL jobs at the times when it happens, looked if there are any subscriptions or snapshots at the time and analyzed which reports that have been executed just before the issue occur but I haven’t found any conclusions. I have also analyzed memory consumption and I can see that the process sometimes consume more than 20 GB ram and normally it consumes around 2 GB Ram, but I can’t say that it rush up to 20 GB ram just before it crash. I can’t see any connection between the issue and memory consumption. I can see that we have more than 30 000 report executions each day and I have tried to find some kind of recommendation from Microsoft regarding setups like ours but I can’t find a document that says that we certainly need a front end web farm or similar. Any suggestions?