Path: utzoo!utgpu!jarvis.csri.toronto.edu!mailrus!umich!samsung!brutus.cs.uiuc.edu!psuvax1!psuvm!HMCVAX!NED From: NED@HMCVAX.CLAREMONT.EDU (Ned Freed, Postmaster) Newsgroups: bit.listserv.pmdf-l Subject: RE: Pmdf_relaying and vms logicals and master_debug Message-ID: Date: 9 Feb 90 07:40:00 GMT Sender: PMDF Distribution List Reply-To: PMDF Distribution List Lines: 30 Approved: NETNEWS@PSUVM Gateway Errors-to: postmast@YMIR.BITNET X-Envelope-to: PMDF-L@IRLEARN.BITNET X-VMS-To: IN%"DMPM@DUKEMC.BITNET" X-VMS-Cc: IPMDF X-To: DMPM@DUKEMC, INFO-PMDF@YMIR James Dryfoos writes: > I have made Ned's changes to pmdf_mail.pas and now can use the > pmdf_relaying feature. However there are a few things you should > know. First, if you turn mater_debug on for the local channel > I am told it is supposed to say pmdf_relaying flag is set. It does > not do this. The output from master_debug is below. I added the code to output this information after I posted the change to info-pmdf. Sorry about that. I do tend to change these discretionary patches around as I fold them into my development copy of PMDF. > Also, when you define the logical pmdf_relaying you must define > it in executive mode (def/exec) and have sysnam priv to do this. > If you do not have sysnam it automatically (w/o warning) creates > the logical under supervisor mode (do a show log/fu pmdf_relaying). > Here is the catch. If you notice this and then give yourself > sysnam priv you can do the def/exec again and it will say > that the previous value has been superseded. However, it was superseded > in supervisor mode and not exec. You must first deass the logical > and then do the def/exec. Isn't this a lovely "feature" of DEFINE? I wish it did what you told it to, and complained if it cannot! However, it is probably way too late to change... I ran into this (as well as some other idiosyncratic behavior) recently while writing a little logical name editor. SHOW LOGICAL/FULL can be your friend. I know it is mine. Ned