We were able to collect some more traces today.

In the first "plc shutting down.tra", our app starts up at 15:37:02.816 (I believe the entries prior to that were caused by one read subscription from an instance of the EasyOPC.NET Demo Application).

The system operates normally until 15:40:30.519 when the first (and all subsequent) writes fail.

At 15:40:30.784 OnDataChange to the EasyOPC.NET Demo Application fires with an error.

At 15:40:30.802, our app receives the equivalent OnDataChange error.

After that point, in addition to a bunch of failed writes, it looks like the system is repeatedly calling AddItems with various tags which it was previously writing to. I don't understand this.

---

The second trace 'app restart after plc restart.tra', continues like the first one left off.

At 15:50:55.713 our app was shut down (the last message in the log was occurred at that time, shutdown was requested slightly prior).

15:51:16.219 is the time of the first entry in the log after restart. After restart, data reads and writes normally.

I don't know exactly when the PLC reboot completed, but I believe it was before our app shutdown at 15:50:55.713.

---

I did not try synchronous writes instead of asynchronous (even if writes worked after the PLC restart, we would still have the issue of the read subscription not functioning properly).

Please review and if you still believe the OPC vendor is at fault for both reads and writes not working post-plc-restart, I will direct them to this thread.

Regards,

Shea

The PLC was power cycled earlier today (the customer did it without my knowing). The attached trace contains OPC Analyzer data of failed writes.

Does the trace contain enough information to troubleshoot the problem or do we need a trace of before/during/after?

Thanks,

Shea

Sounds good - I will follow your suggestion of running the analyzer, triggering the problem, then reverting back to the original system configuration.

Do you have a link to download the analyzer?

A google search yielded this link but it would require setting up an OPC Foundation account (which I am okay doing if necessary):

www.opcfoundation.org/DownloadFile.aspx?CM=3&RI=359&CU=1

Thanks,

Shea

In the steady state we are updating 1 tag every 500ms (a heartbeat) as well as data from 3 thermocouples each of which typically update every 2-3s. There are 15-20 additional tags which update infrequently.

There are reads (subscriptions) which stop updating and do not work again until the app is restarted (I didn't think to check the exception or inner exception on the reads but will do so next time it happens).

I am open to installing an OPC analyzer between the server and our application. I assume there will not be any side effects?

Thanks,

Shea

The problem just occurred again (it only happens when they power cycle their PLC, which has been relatively infrequently lately):

The outer exception contains the error message:
'An OPC operation failure with error code -2147467259 (0x80004005) occurred, originating from 'OpcLabs.EasyOPCRaw.DataAccess.RawEasyDAClient'. The inner exception contains details about the problem.'

The inner exception contains the error message:
'Unspecified error'

Please advise.

Thanks,

Shea