Machsupport Forum
Mach Discussion => General Mach Discussion => Topic started by: sshneider on March 26, 2008, 05:36:25 PM
-
Things are humming along nicely and all of a sudden-machine stops moving and this pops up in the statust window ...
Wed - 13:35:05 ---command killed due to buffer error.
Anyone know how to fix this?
Thanks!
Sid
-
Hi I am not quite green yet on mach.If it were my problem I would first remove or disable anything that might compete for irq hits.E.G. sound lan usb reduce colors/resolution on video.Is it happening when commands get complex Al It may be strange but does your registry still have printer entrys. Ill keep adding as I think of them next I would disable any bios cache
-
Nice to get a reply! Thanks! All good suggestions but, other than checking the registry I don't think it's any of those things since the problem just started recently and nothing else on my system has changed.
I wouldn't say the problem happens when things get 'complex' but it does happen in the latter part of programs that have over 500 lines.
It would help to know exactly what a "Buffer Error" is. Or rather "what/where is the buffer (i.e. is it related to RAM memory, DIsk Space, etc) and what affects its usgae.
Thanks,
Sid
-
I read this recently,on high end chips there is a problem with the cooler chip heat transfer as things age.I.E.hot spots on the chip If you cant solve it any other way try regreasing your chip/ cooler interface.This would fit the longer program scene. is there a temp change in your shop Al more I dont know what software you use. If your smarter than me you backed up when you were happy,If not try a restore type operation It could just be a windows fart.
-
At first I thought the same thing, that perhaps this was a hardware based problem so I had another computer at my shop laying around. I connected that computer. Loaded Mach and started to run. Same problem. I think it's definitely a software issue.
I have been able to create "buffer errors" while runing ver 3.39. If I select the 'enable THC toggle' and run a program, pressing "feedhold' will stop my computer in it's tracks and give me the 'command killed' message. I wonder if anyone else has this same problem or can verify.
I should say that this isn't a huge problem for me, I would just like to know why and if this is indicative of other issues that I might be having.
Thanks,
Sid
-
Hi, Sid
Tried it on 2 computers, One of them didn't have a problem, The other, Feed-hold wouldn't work but stop would.
Theres a gremlin in there somewhere I'd say.
Thanks, Chip
-
gremlin indeed. I bet it resembles a smurf with red eyes and small wings. Some have even described it as a Joe Cartoon fly. I've heard stories, but never really seen one (fingers tightly crossed).
-
Saw this error mentioned on the Yahoo group, guy was having a problem with a brain and this message, the brain looked at some Inputs and if wrong the Stop button was pressed. I think Brian sugested it could have been a problem with the guys input being twitchy and momentarily messing with the stop button and thus clearing the buffer. Or something like that LOL, would have to find and read it again to refresh my old brain.
Not sure if this is in any way similar to your situation though.
Hood
-
I m going to speculate and hope it make a bulb light up.Chips post means something.My brain thinks of a linux hybrid crossing paths with a invidia driver .easy to sort out, is invidia involved with these computers ?if not what is the difference between chips computers.If it could be that simple then a rewrite of the linux invidia driver would be a soulution Al A more defining question is does it stop at the same place on the same part. I am focusing on the com because I know nothing of the software. Other gremlins does your bios support print screen?and is it getting wierd commands If it stop in the same place is there some spooler remnent still counting?The no blue screen means its not real deep.Its possible its not being dumped but overwritten by a os hard wired mem address Ill throw them in the air you guys try to shoot them down AL
-
Not at the shop today but I will check out some of those items tomorrow. But I'm 99% sure that the error happens at different times/lines of the program. I'll verify though.
Thanks to all for putting on your thinking 'caps'.
Sid
-
Hi never heard if you solved it but I found this read that might be a clue.(www.lvr/jansfaq.htm) goodluck Al
-
Hi,
I am not sure if I have solved the problem yet. Last week I started having some other weird things happen and with the help of Bob @ Practical Micro we narrowed the issue down to a faulty PMDX 131 board. They sent me out a replacement and I installed last Friday. Ran it for a coule of hours and all seems well.
As I have said before, I don't really know where/which 'buffer' Mach is referring to in the error message. So, it's kind of conjecture on my part but I would imagine the PMDX has a buffer (or maybe it reads from the buffer on the PC runing Mach?) and I'm thinking that's where my problem was.
Runing the machine this week should be telling. Thanks for the help!
Regards,
Sid
-
"I don't really know where/which 'buffer' Mach is referring to in the error message."
Is there any more detail about what this particular error indicates?