DSKMGR.THD --- Copyright 1987 by Phil Wheeler An original compilation of Compuserve Model 100 Forum messages for use by Forum members only. Recently a bug report on DSKMGR was uploaded, regarding its detection of defective media. It resulted in an exchange here in the Forum -- and eventually, an Easyplex message from the author, J.K. Heilman, explaining the cause of the "bug". That message concludes this file. Message range: 162788 to 162905 Dates: 12/20/87 to 12/25/87 [plus EasyPlex of 1/4/88] Sb: #DSKMGR Bug Report Fm: Stan Wong 70346,1267 To: [F] Phil Wheeler 71266,125 Don't know who to send bug reports to on the DSKMGR program. I just started using it and ran across a bug. The bug appears during a format of defective media. FLOPPY.CO reports an I/O error while DSKMGR appears to complete normally. Copying a file to the disk seems to load a correct directory entry followed by the contents of the file in the remaining directory entries. DSKMGR v3.02 should report an error if the format process fails. Fm: Tony Anderson 76703,4062 To: Stan Wong 70346,1267 Bugs should be reported to the author of the program. James Heilman, who wrote the program, can be reached by sending a message via EasyPlex to 71216,1375. Fm: Phil Wheeler 71266,125 To: Stan Wong 70346,1267 Generally, not working like FLOPPY is considered an advantage . Tony's advice is best -- reporting it to James Heilman. BUT -- what computer and which version in the DL's were you using? And did you verify that the disk was truly defective as FLOPPY reported. It is conceivable that my ASM, built by disassembly, missed something in the original JKH code (unlikely -- since bigger things would be likely to happen, but possible). Fm: Stan Wong 70346,1267 To: Tony Anderson 76703,4062 I will report the DSKMGR bug to James Heilman. It's not a real biggie since all floppies sold have "certified and tested" on the box. Well, so much for Maxell disks... It's not as if TDD-1 uses a high density format. The diskettes are supposedly single-side, double density, double-tracks 135 tpi. Anyone from Maxell out there? Fm: Stan Wong 70346,1267 To: Phil Wheeler 71266,125 I have tested the "bad" media with FLOPPY as well. FLOPPY reports an I/O error whereas DSKMGR thinks everything is hunkey-dorey. I tried both with "good" media and both work fine. I tried version 3.02 of DSKMGR, the relocatable version (from file DSKMGR.DOR). It is loaded as the highest program in memory (with your XMDPW5 just below it). I will report this to James Heilman as Tony suggested. Fm: Phil Wheeler 71266,125 To: Stan Wong 70346,1267 Please try DSKMGR.100 (non-relocating version, always at top of memory) and see if you get the same result withthe same disk. That will tell if it is James' original version or just my relocating version. Need to know if it is my problem or his -- and I may look at it ether way, just need to know where to look! Fm: Stan Wong 70346,1267 To: Phil Wheeler 71266,125 Tried the non-relocating version of DSKMGR with the same results. FLOPPY reports an I/O error during a format of defective media. DSKMGR reports normal completion. If a file is copied to the disk (via DSKMGR) the file name appears as the first directory entry. Subsequent entries have the file contents. DSKMGR and FLOPPY performs as expected for "good" media. The software was downloaded in the last couple of week and is version 3.02. Shall I report this is James Heilman? Fm: Phil Wheeler 71266,125 To: Stan Wong 70346,1267 Surely tell James about this. If I get time over the holidays, I will take a look, dn fix the REL versions (ughh -- being 3 of them that is a big job!). Providing, of course, I can find the reasom for the problem. Hmmmm... now to find a defective disk to use in testing!! Fm: Stan Wong 70346,1267 To: Phil Wheeler 71266,125 If you want the "floppy" that I used to detect the DSKMGR bug, I'll be happy to mail it to you if you plan to debug the program. I'll be happy to send it to whomever needs it. Also note that this problem occurs with a TDD-1. I don't know what would happen with a TDD-2. Fm: Phil Wheeler 71266,125 To: Stan Wong 70346,1267 Hmmm..best not send it to me, then. I don't have a TDD1 any more. JKH would, I'm sure. I will look at the code first, and see if I see any clues there -- given some time this week. I'm still looking for the last bug in XMDPW5.200. And it has eluded me pretty well; small but mighty! Date: 04-Jan-88 07:07 PST From: J.K. Heilman [71216,1375] Subj: DSKMGR Question PHIL: FYI...RESPONSE TO QUESTION FROM STAN WONG. Stan: Here's the story in a nutshell... The Format command is internal to the TDD, i.e., the Model 100 issues a single instruction to the drive and it executes a series of internal commands that does the formatting. To check format validity, FLOPPY shifts access modes and performs sector write/read. This format-validation code would have added significantly to DSKMGR's size and was not usable by any other segment of the program. Since I/O errors are reported by the Load/Save functions, I chose to drop the validity check. Based on your description, I suspect that your disk has a defect that corrupts the first sector on the disk which is used for the file directory. This is the only sector on the disk where the I/O checks are not performed. The odds of that problem are very low and, frankly, I just missed it. If you're interested in more technical information on the TDD, I wrote two articles which appeared in the November/December 1986 issues of PICO. You can get back-issue information from them at: PICO Magazine Portable Computing International P.O. Box 428 Peterborough NH 03458-0428 I hope this answers your question, if not the problem. Thanks for the letter.