The loop with RPRLSCI is a regretful bug which I thought I had fixed in 1.5h but no, I forgot that a blank line is sent to the job output file after completion of a command, even if it is ignored, so an output file was generated even though I had flushed all the lines that had been printed to it... The problem with this kind of things is that it is very difficult to test it for real unless you accept to run the risk of starting a REAL mailing loop :-( Anyway I've fixed it. I have also added an option to DISTRIBUTE to allow you to temporarily turn it off during peak hours, and fixed a stupid problem which caused servers with release = 1.05x to accept DISTRIBUTE jobs (that started when I changed the 'default' release number for unlisted servers from from 1.3c to 1.5f to solve the problems that you know). Reference to a human person: I agree with whomever said he hates receiving stupid inquiries from Joe users. I had 211 pieces of mail in my reader this morning, 80 of which were of this category and I DON'T WANT any reference to my userid in error messages from LISTSERV. People who REALLY need to contact me will find out, painful weenies will let the matter drop. MAIL-L being linked to NODMGT-L: I really don't understand what's happening, but it seems that every piece of mail sent to MAIL-L is forwarded to NODMGT-L, automatically, although the lists are in no way linked. Conclusion: some server (not a FRECP11-LISTSERV, I mean some other server) who's getting MAIL-L stuff is forwarding it to NODMGT-L. Eric