Random Path Pilot crashes

[jnack95]

I've been getting random crashes with path pilot since I've owned my machine (several years). It will just randomly crash about once a week (sometimes less) in the middle of a job. The screen freezes and the machine stops (luckily). I then need to cut power to the controller (hard reset) and reset the machine, re-find origin and re-run the program again. This has happened with both PP1.9.x and 2.0.x . I have no tool changer and it just seems to be unrelated to any specific thing the machine is doing. I thought this might be a power issue and put the controller on a UPS but it still happens. I think it might be bios related but I can't seem to access the bios? I need to check dmsg next time it crashes for clues. Anybody have any ideas?

Similar Threads:

• path pilot and fusion
• Got my Path Pilot
• Path Pilot vs Mach 3
• Welcome to the Path Pilot forum!
• Problem- Post Processor crashes with long path names

[jwatte]

It could be power related, but on something like the USB bus or the controller interface wires.
Also, do you have it on network? It could be a network thing.
It could also be some bug that slowly leaks memory, and then runs out. You might be able to check memory use over time using some system recording tool like "sar"
Using dmesg, and checking the syslogs, seem like good things to try, too.

[syscore]

When you say "once a week", do you mean that PP is up and running continuously for the whole week? Or do you start it each day, and once a week during the day, it crashes?

[PCW_MESA]

I've been getting random crashes with path pilot since I've owned my machine (several years). It will just randomly crash about once a week (sometimes less) in the middle of a job. The screen freezes and the machine stops (luckily). I then need to cut power to the controller (hard reset) and reset the machine, re-find origin and re-run the program again. This has happened with both PP1.9.x and 2.0.x . I have no tool changer and it just seems to be unrelated to any specific thing the machine is doing. I thought this might be a power issue and put the controller on a UPS but it still happens. I think it might be bios related but I can't seem to access the bios? I need to check dmsg next time it crashes for clues. Anybody have any ideas?

If you get a hard crash that requires power cycle or reset to recover I would suspect problems with the PC motherboard or memory.

[rowbare]

When I started using PP I had problems with lockups if my compressor kicked in while I was using the jog wheel. I removed the USB extenders I was using and rerouted the cables and that solved the problem. Oh and I also installed the Tormach suppressor kit. This is in an upgraded Series 1.

bob

[tmarks11]

...I would suspect problems with the PC motherboard or memory.

If your PP controller is like mine, and came with only a single 1 GB RAM stick in it, then considering throwing in another stick. PP2+ and the new OS takes more memory.

[jnack95]

When you say "once a week", do you mean that PP is up and running continuously for the whole week? Or do you start it each day, and once a week during the day, it crashes?

No, it's not continuous. I maybe use it 4 or 5 hours a day. Thanks everyone for the responses. I've upgraded the underlying Linux mint os and plugged the keyboard into a regular USB2 slot on the controller instead of the recommended USB3. It seems the crashes usually occurred when a usb key was already inserted....maybe. I've also started using a smb drive on the controller to eliminate the usb. So far so good. No crashes yet. If that doesn't work then maybe upgrading my 1 gb of RAM is next but right now I think it's either the USB system or possibly the video firmware (because X seems to freeze when the crash happens). I've had problems with both the video drivers and USB on a different Linuxcnc machine.....

[jnack95]

Well it just crashed again. This time the machine wasn't running and the computer had been idle for abouyt 15 minutes while I was doing something else. The following are the only errors is dmesg.

i915 0000:00:02.0: fb0: inteldrmfb frame buffer device
[ 1.398994] ata1: SATA link up 1.5 Gbps (SStatus 113 SControl 300)
[ 1.404069] ACPI Error: [DSSP] Namespace lookup failure, AE_NOT_FOUND (20170119/psargs-363)
[ 1.404110] ACPI Error: Method parse/execution failed [\_SB.PCI0.SAT0.SPT0._GTF] (Node ffff9b98004b6898), AE_NOT_FOUND (20170119/psparse-543)
[ 1.404163] ata1.00: ATAPI: ASUS DVD-E818AAT, 1.01, max UDMA/100
[ 1.405632] ACPI Error: [DSSP] Namespace lookup failure, AE_NOT_FOUND (20170119/psargs-363)
[ 1.405669] ACPI Error: Method parse/execution failed [\_SB.PCI0.SAT0.SPT0._GTF] (Node ffff9b98004b6898), AE_NOT_FOUND (20170119/psparse-543)
[ 1.405721] ata1.00: configured for UDMA/100
[ 1.408841] scsi 0:0:0:0: CD-ROM ASUS DVD-E818AAT 1.01 PQ: 0 ANSI: 5
[ 1.410988] ata6: SATA link up 3.0 Gbps (SStatus 123 SControl 300)
[ 1.411700] ACPI Error: [DSSP] Namespace lookup failure, AE_NOT_FOUND (20170119/psargs-363)
[ 1.411737] ACPI Error: Method parse/execution failed [\_SB.PCI0.SAT0.SPT5._GTF] (Node ffff9b98004b6af0), AE_NOT_FOUND (20170119/psparse-543)
[ 1.411942] ata6.00: ATA-8: WDC WD2500AAKX-00ERMA0, 15.01H15, max UDMA/133
[ 1.411943] ata6.00: 488397168 sectors, multi 16: LBA48 NCQ (depth 31/32), AA
[ 1.412723] ACPI Error: [DSSP] Namespace lookup failure, AE_NOT_FOUND (20170119/psargs-363)
[ 1.412759] ACPI Error: Method parse/execution failed [\_SB.PCI0.SAT0.SPT5._GTF] (Node ffff9b98004b6af0), AE_NOT_FOUND (20170119/psparse-543)
[ 1.412992] ata6.00: configured for UDMA/133
[ 1.438463] sr 0:0:0:0: [sr0] scsi3-mmc drive: 48x/48x cd/rw xa/form2 cdda tray
[ 1.438464] cdrom: Uniform CD-ROM driver Revision: 3.20

[shred]

This on a Tormach controller or homebrew?

[jnack95]

This on a Tormach controller or homebrew?

Tormach controller bought several years ago..

[jwatte]

Those ACPI errors seem to happen during start-up (1.5 seconds into the boot) and are probably not related.

The most common computer gremlins are bad drivers/hardware for some intermittently-used device (USB, network) or electrical noise entering either through power supply, or more commonly through things like USB or network or controller cabling.

[PCW_MESA]

Since this is a Tormach controller and it fails on both Pathpilot 1.9.x and 2.0.x (which have completely different drivers) , My bet would be on bad hardware...

[jnack95]

So, I booted linux from a live CD and ran memtest for about an hour. No errors. I also reinstalled path pilot 2 and updated to the current version of 2.0.4. I've also flashed a new version of the bios. I have also determined that the ACPI errors I was getting are related to the SATA mode in the bios. Switching from AHCI to IDE eliminated them and the system seems to run fine in IDE mode (any drawbacks?). So far so good. Thanks everyone for your help.

[jnack95]

Two weeks now and no crashes. I think it was the bios upgrade. I left the bios SATA setting on AHCI.

[jnack95]

Good to hear. Your last post says things were working good using IDE and now you’re happy leaving it on AHCI?

Nowadays most HD’s run in ACHI mode while IDE mode is mostly used for CDROM/DVD drives – these are separate settings in the BIOS.

Putting the SATA in IDE mode got rid of the errors, but I decided they were of no consequence (i get similar ones on other linux computers) and changed it back to AHCI mode. I've been AHCI mode for the past several weeks.