WINDOW & B16 (Non-numeric...) (AREV Specific)
At 26 NOV 2001 10:31:28AM Roy Drummond wrote:
We are getting a strange instance of a B16 error in a window. The first time a window for a certain file is accessed it abends with a B16 error. The data is saved that was entered and the second time the window is accessed the error does not occur (something has been set that was not set before). All windows accessing the table get the error the first time. I am in the process of LH_VERIFYing my indexes. Hope this is not the problem. Any insights (open or othewise
?
TIA,
Roy
At 26 NOV 2001 10:54AM [url=http://www.sprezzatura.com" onMouseOver=window.status=Click here to visit our web site?';return(true)]The Sprezzatura Group[/url] wrote:
Paint the Window, toggle the Recalculate status, save. Repeat (put back to how it was), save.
Should now be ok.
Old and weird one.
World Leaders in all things RevSoft
At 26 NOV 2001 12:33PM Roy wrote:
I tried that and it did not fix the problem. Any other suggestions.
Roy
At 26 NOV 2001 12:35PM [url=http://www.sprezzatura.com" onMouseOver=window.status=Click here to visit our web site?';return(true)]The Sprezzatura Group[/url] wrote:
Did you do a FLUSH to get rid of the old version from memory?
World Leaders in all things RevSoft
At 27 NOV 2001 01:52PM Roy wrote:
You bet! OFF does a great memory flush. I think we have the same problem with the EDIT program now and then. When I try to delete a record I get the Non-numeric abend. When I check the record it has been deleted and the next delete works fine.
Roy
At 28 NOV 2001 02:34AM [url=http://www.sprezzatura.com" onMouseOver=window.status=Click here to visit our web site?';return(true)]The Sprezzatura Group[/url] wrote:
Sounds like a problem with an MFS possibly. Any custom MFSs?
World Leaders in all things RevSoft
At 28 NOV 2001 10:08AM Roy wrote:
SI.MFS is the only MFS on the file. Thanks for your continued follow up.
Roy
At 28 NOV 2001 10:16AM [url=http://www.sprezzatura.com" onMouseOver=window.status=Click here to visit our web site?';return(true)]The Sprezzatura Group[/url] wrote:
Any chance that si.mfs is evaluating a symbolic for indexing purposes which references a labelled common which has not been setup? After the crash the variable may have initialised to zero and the problem disappeared?
World Leaders in all things RevSoft
At 28 NOV 2001 10:55AM Victor Engel wrote:
How many fields do you have on the file? How many indexes? What if you change the index definition or recompile the index code?
You said SI.MFS was the only MFS on the file. Do you have a volume-level MFS on the volume that the file is on?
What if you edit a record using the system editor prior to using the window? Do you still get the error? What about setting a qfile with the same name as the file you're having problems with. This way, the window will still function but on a different table. This should help you narrow down whether the problem is with the window template or the FS.
At 29 NOV 2001 07:34AM Dan Reese wrote:
If you remove all indexing on the file (and verify that the !file is gone), does the problem still occur? If not, you might have some corruption in the index file.