"Disturbing somewhat this thread is." -- Yoda I try to keep my P2P to a minimum, because it does introduce additional points of failure, but sometimes it is either unavoidable, or much cleaner/more efficient than alternatives. Besides, doesn't the "D" in DCS stand for "distributed"? The limitations on P2P in most systems I've worked on (capacity, in my experience usually, not reliability) are somewhat frustrating, particularly when accompanied with inflexibility in mapping I/O points to arbitrary control processors (which is also often the case). This also seems to me to illustrate a(nother) problem with exception-based reporting, at least when apparently implemented with the philosophy on the sink end of, "value hasn't changed, so it must be good". I guess I will be getting better acquainted with rsom. Corey Clingo BASF dirk.pauwels@xxxxxxxxxx Sent by: foxboro-bounce@xxxxxxxxxxxxx 09/06/2010 08:50 AM Please respond to foxboro@xxxxxxxxxxxxx To foxboro@xxxxxxxxxxxxx cc Subject Re: [foxboro] P2P links locked Hi Sander, Once you know the link is bad, there are probably several ways of fixing it, the problem is how to detect a bad link, before the programming which uses the link, fails or executes incorrectly because of the incorrect input. Perhaps there is a way of running an rsom and capturing the error messages at regular intervals, thus generating an alarm when errors are reported, which would draw the operator or engineer's attention. (just thinking aloud) I know in the perfect Invensys world one would not have P2P links, but in real life it's almost unavoidable, so why is there so little (none except for the dbvu, which doesn't even report errors of this kind) alarming on this P2P stuff We know of 2 bad links, maybe there are more,...... Rgds, Dirk sander@xxxxxxxxxxx Sent by: foxboro-bounce@xxxxxxxxxxxxx 06/09/2010 15:07 Please respond to foxboro@xxxxxxxxxxxxx To foxboro@xxxxxxxxxxxxx cc Subject Re: [foxboro] P2P links locked Dirk, We've had the same issue some time ago. Connections between different CPs. Also fixed it relaying the connection to a different parameter, checkpointing en switching the connection to the original parameter. With us it did seem to happen only the connections with a CALC as the source. Regards, Sander _______________________________________________________________________ This mailing list is neither sponsored nor endorsed by Invensys Process Systems (formerly The Foxboro Company). Use the info you obtain here at your own risks. Read http://www.thecassandraproject.org/disclaimer.html foxboro mailing list: //www.freelists.org/list/foxboro to subscribe: mailto:foxboro-request@xxxxxxxxxxxxx?subject=join to unsubscribe: mailto:foxboro-request@xxxxxxxxxxxxx?subject=leave