[Info-vax] Latest TCPIP Services seem irreparably horked...

[VAXman- at SendSpamHere.ORG]

In article <jit23p$7ld$1 at dont-email.me>, Simon Clubley <clubley at remove_me.eisner.decus.org-Earth.UFP> writes:
>On 2012-03-02, Subcommandante XDelta <vlf at star.enet.dec.com> wrote:
>> On Thu, 01 Mar 2012 15:27:18 -0500, JF Mezei
>><jfmezei.spamnot at vaxination.ca> wrote:
>>
>>>Mr VAXman, I would delete the SMTP service from that node and recreate
>>>it from scratch.
>>>
>>>The services database seems to be shared in a cluster and perhaps TCPIP
>>>services shows you the records it thinks belong to this node while POP
>>>uses either records from another node or perhaps another copy of the
>>>database.
>>>
>>>set watch file might be of use to see exactly which file TCPIP utility
>>>uses to read the SMTP service config and which file is used by the pop
>>>server.
>>>
>>>It is possible that the drones haven't been taught about logical names
>>>yet and they hardcoded the service config filename in the pop software.
>>
>> AUUGGGHH...
>>
>> ... the Horror, the Horror...
>>
>> ...Never get out of the boat - God damn right...
>>
>> Don't mind me, I was having a sensibility infarction.
>>
>> Back to normal.
>>
>> What are the last clean (classic, pure, old-school)  VMS versions, VAX
>> and AXP?
>
>I think I know where you are going with this, but that doesn't help
>you when it comes to TCP/IP services as you need to upgrade to new
>ECO versions to fix any public facing security issues found in the
>stack. You have to be able to trust that the latest ECO will work
>without introducing any new functionality/security problems.
>
>BTW, you can also add the VMS sftp client to the list of wonky software.
>
>HP are currently chasing a sftp protocol error I am experiencing (we don't
>know yet if it's the client or (non-VMS) server at fault) and it doesn't
>help matters that when a protocol error occurs the transfer is silently
>terminated and you are returned to the sftp> prompt without _any_ error
>message been output.
>
>You have to enable the detailed debugging mode in order to see what is
>going on. If the VMS sftp client were a student project (which is what it
>feels like :-(), I would have docked the student marks for not reporting
>the error back up through the various layers of the program. In VMS
>software, you simply should not be experiencing this.
>
>It's also got half finished functionality (such as the ascii mode) which
>is documented in the V5.6 ECO5 help, but is only (apparently; I have not
>tested it yet) functional in the very latest images.

I have seen an issue with sftp/ssh.  Hoever, to belie Mr. Mezei's cluster
theory, this is on a standalone node.

I'm convinced it has something to do with POP balking at the TOP HEADERS
setting (which is set).  Disabling ignore_mail11_headers gets the email
to the client; albeit, in a fashion which keeps it from decoding any of
the MIME parts.

-- 
VAXman- A Bored Certified VMS Kernel Mode Hacker    VAXman(at)TMESIS(dot)ORG

Well I speak to machines with the voice of humanity.