Emmitt, It was not the CPU cycles that were building it was memory usage that was increasing. Nearly all of my routines use temporary tables. On occasion, I have five or more temporary tables open at a time. I wonder if some of the tables are not releasing memory when I issue the drop.
BTW. I use LOAD tablename FROM filename AS CSV when I load my UPC catalog. I have not seen the hangs in that area. Each quarter I load about 30 files by that method. Some of the files contain over 300,000 records. John From: [email protected] [mailto:[email protected]] On Behalf Of Emmitt Dove Sent: Monday, November 08, 2010 4:23 PM To: RBASE-L Mailing List Subject: [RBASE-L] - RE: RBASE hangs - not responding John, If the CPU utilization of RBG8EE is incrementing, then there must be something amiss in your application. The question is, "What?" Recently I've been doing a lot of work on data conversion. A given process may create 3-5 temporary tables and load them with data from external files. One thing I found is that RBG76EE had a tendency to show difficulties if I attempted to LOAD tablename FROM filename AS CSV. The process would appear to be hung, but Task Manager showed it to be active, consuming lots of CPU. I could look at the temporary files and see them growing by leaps and bounds. And this was while attempting to load a relatively small CSV file. I found that if I took the exact same data file and put a LOAD tablename at top and END at bottom I could RUN filename without error. The inconsistency of this is that not every instance where I attempted the LOAD . AS CSV would fail. I examined the data in the CSV in detail and verified that it was all correctly formatted. The message is that it could well be something in your application that a recent update has broken. Try falling back a few updates and see if the problem persists. By the way, I readily admit that I have not attempted to do that myself; time has not permitted me to reconstruct the conditions that cause the failure. At this point I draw no conclusions as to whether a recent update is causing my issue. Emmitt Dove Manager, Converting Applications Development Evergreen Packaging, Inc. [email protected] (203) 214-5683 m (203) 643-8022 o (203) 643-8086 f [email protected] From: [email protected] [mailto:[email protected]] On Behalf Of John Engwer Sent: Monday, November 08, 2010 16:05 To: RBASE-L Mailing List Subject: [RBASE-L] - RE: RBASE hangs - not responding Emmitt, I did check task manager a few times. The first time the RBASE application was using the most memory cycles and it was incrementing up as I watched. The other times the processes were 20 or less for everything that was running. Yesterday afternoon I had the application open on my development system when I received a phone call, ten minutes later I went back to RBASE and it was hung. I looked at task manager and the memory usage (RBG8EE) was slowly incrementing upward. I watched it for another 10 minutes and then hit CTRL/ALT/DEL. Some of the users have told me that the system hangs if they go away from the keyboard for a short while. Others have told me that it is totally random and occurs when they are active typing. I do not think that it is anything in the application because there is no consistency. John From: [email protected] [mailto:[email protected]] On Behalf Of Emmitt Dove Sent: Monday, November 08, 2010 3:35 PM To: RBASE-L Mailing List Subject: [RBASE-L] - RE: RBASE hangs - not responding John, One of the questions I would ask is, "When one of the RDP sessions hangs, what does Task Manager show on the Processes tab when sorted by CPU descending?" In other words, is something consuming myriad resources? If so, is it the hung process or something else? Emmitt Dove Manager, Converting Applications Development Evergreen Packaging, Inc. [email protected] (203) 214-5683 m (203) 643-8022 o (203) 643-8086 f [email protected] From: [email protected] [mailto:[email protected]] On Behalf Of John Engwer Sent: Monday, November 08, 2010 15:20 To: RBASE-L Mailing List Subject: [RBASE-L] - RBASE hangs - not responding For several month my clients have been reporting intermittent hangs within my application. Here are my observations and comments. V8 compiled, latest version Some networks are peer-to-peer and some are server based. (total of 8 networks) The client with the most hangs is on a Dell quad core Xeon server with 8 gb memory. Average of 10 - 12 RDP users. RBASE is the only software posting errors in the event log. Scratch files set to TMP. I have done so many unloads/reloads that I am concerned that I may be wearing out the bits! No errors during unload/reload. The sites with the most hangs also use RScheduler80. RScheduler is in a standalone directory with its own DLLs (they also are the latest DLLs) Only one set of DLLs except where the client is running RScheduler. (2 sites) The hangs occur randomly throughout the application. Most sites average 1 or two hangs a day. I have seen as many as 6 per day. Sometimes the screen goes black but most of the time the OS posts a message "application not responding" and the only out is to press CTRL/ALT/DEL. If anyone has solved this type of problem in the past or has a suggestion on how to troubleshoot this problem, I would like to hear your suggestions. I have run out of ideas. Thank you. John
