Re: SMTP Mail from z/OS

2019-03-25 Thread Bruce C Baxter
I should have qualified my statements acknowledging that sendmail can often be used as an open relay. In this case, we're sending email only outbound, and the IPSEC firewall pretty much precludes anyone from off platform connecting to the SMTP mail daemon on z/OS USS. You assert that CSSMTP

Re: SMTP Mail from z/OS

2019-03-22 Thread Seymour J Metz
What is EXIM, chopped liver? I had a few gigs at organizations that used the SPOOL to mail facility, one of which migrated to CSSMTP. IBM is like Quantum Mechanics; you may be able to make accurate predictions but don't try to make sense out of the internal workings - you'll just get

Re: SMTP Mail from z/OS

2019-03-21 Thread Timothy Sipples
Allan Staller wrote: >I am not aware of any shops that use SMTP "to the mainframe". >i.e. the mainframe is not usually the target for inbound >communications. This would require a "email server" to be running >on the mainframe (HRC anyone?) Even if "mainframe" means z/OS specifically in this

Re: SMTP Mail from z/OS

2019-03-21 Thread Tony Thigpen
Allan stated: > I am not aware of any shops that use SMTP "to the mainframe". I have seen some implementations of SMTP receive on a mainframe. In one case, the receiver would take the message as jcl and submit it to run as a batch program. This was mainly to allow programmers to send compile

Re: SMTP Mail from z/OS

2019-03-21 Thread Allan Staller
CSSMTP was designed to mitigate several concerns with the SMTP client. As far as I can recall, there were some configuration options, that if not addressed, allowed the SMTP client to be a mail relay (among several others). CSSMTP was designed to eliminate those exposures. I am not aware of any

Re: SMTP Mail from z/OS

2019-03-21 Thread Paul Gilmartin
On 2019-03-21, at 10:03:08, Baxter, Bruce (ITS) wrote: > I'm going to cross post this to mvs-oe as well as here... > > The folks that support Comm Server in-house have approached me about > replacing the SMTPD (which is apparently a USS / OE facility) with CSSMTP, > and I've got some thoughts