Host Engineering Forum
General Category => Do-more CPUs and Do-more Designer Software => Topic started by: OrionHE on May 10, 2016, 05:00:11 PM
-
I'm using an H2-DM1E and getting an error message:
(https://forum.hosteng.com/proxy.php?request=http%3A%2F%2Ftoolofgod.com%2Fwp-content%2Fuploads%2F2016%2F05%2FLast-Errors.png&hash=d8fab6b0a282e0f77bd6cf7843017f60b3b5bcd1)
In the "MSG:" field it says RCPT TO:<""> which made me think that the double-quotes is causing some sort of unrecoverable error, but when I try to clear the double-quotes, Do-more designer tells me this message:
(https://forum.hosteng.com/proxy.php?request=http%3A%2F%2Ftoolofgod.com%2Fwp-content%2Fuploads%2F2016%2F05%2Fempty-cell.png&hash=afd8d8c1cd90be40831a1bb7d3d4441ac8dac526)
which makes me think that double-quotes are ok.
The big problem is that under certain conditions,
1. the PLC will try to send an email
2. the email will be undeliverable (this is ok, as the double-quotes are a placeholder for a real email address)
3. the PLC will shutdown and restart itself, setting $firstScan and messing me up
Ideas? Automation Direct support sent me here because they couldn't explain my issue.
I'm happy to provide more details.
-
More information, I've put in a good email address and the PLC is still under certain conditions restarting itself. The latest error message is here:
(https://forum.hosteng.com/proxy.php?request=http%3A%2F%2Ftoolofgod.com%2Fwp-content%2Fuploads%2F2016%2F05%2Fnew-error.png&hash=3750caaf159aa717923705fef890209d9c6e2df1)
Why would a simple TCP connection issue restart the PLC?
-
I suspect the PLC is restarting because of the I/O errors and the email-related stuff is just warning messages.
-
I suspect the PLC is restarting because of the I/O errors and the email-related stuff is just warning messages.
Thank you for your suggestion. This particular instance happened on my bench-test PLC, but it happens at multiple facilities that use an exact copy of this program in multiple counties across the state of California. This should rule out I/O, Network, and PLC Hardware, leaving programming mistakes and software glitches. I am however open to revisiting any of these topics.
-
Hmmm, that's odd. I can't believe an email error would restart the PLC, unless it gets into some kind of loop retrying and locks everything up. I did that one time doing a DNS lookup too frequently.
-
Unexpected response could be a clue. Could be a crash due to an un-caught exception condition.
There was a bug in the email instruction that was fixed one or two releases ago. I think it would crash and reboot. What firmware version are you running?
As an FYI, the default hardware watchdog behavior is to reboot the CPU. It can optionally be changed to lock down, requiring a power cycle. To get a sense of frequency, you can monitor $WatchdogReboots, which is a retentive counter of the number of times the system has been rebooted by the watchdog handler.
-
I had to put some time into testing, but the problem doesn't seem to have occurred since I've updated the OS version, is this what is meant by the firmware?
Thanks for the suggestion, hopefully the many months of frustration and "blaming the programmer" can come to an end. I knew my logic was sound!
-
I had to put some time into testing, but the problem doesn't seem to have occurred since I've updated the OS version, is this what is meant by the firmware?
Yep, Firmware = Bootloader + OS, and in this case I did mean OS. There was an error in a buffer, where if the length of an email changed (got shorter, I think) on a subsequent email, it could crash. It was fixed a while back.
-
I remember this one well. The new OS should take care of it. Not a hitch since we updated ours.
-
I remember this one well. The new OS should take care of it. Not a hitch since we updated ours.
Pretty sure you reported it. ;)