jca@drutx.UUCP (ArnsonJC) (12/12/85)
The problem below is becoming quite frustrating for me. Basically, I have been TOTALLY unable to send mail to machine 'osu-eddie' at Ohio State University. I have had NO problems receiving mail. The return address from OSU is always: ihnp4!cbosgd!osu-eddie!usr. When I try and send/reply to mail at OSU I consistently get mail fromm uucp saying that they have "...been unable to contact machine 'osu-eddie' since you queued your job," etc., etc. The following is the pertinent excerpts from the uucp messages. Does anybody out there know what is going on? My mail address is fairly simple: ihnp4!drutx!jca. Please respone via mail, as I do not normally read this newsboard. /*/*/*/*/*/*/*/*/*/*/*/*/*/*/*/*/*/*/*/*/*/*/*/*/*/*/*/*/*/*/*/*/*/*/*/*\ This is the mail that I receive FROM uucp: From uucp Wed Nov 6 18:55 EST 1985 >From uucp Wed Nov 6 13:18:02 1985 remote from ihnp4 From: ihnp4!uucp Full-Name: UUCP Admin Received: by ihnp4.ATT.UUCP id AA21268; 6 Nov 85 13:18:02 CST (Wed) Subject: Warning From uucp Apparently-To: drutx!jca Status: R We have been unable to contact machine 'osu-eddie' since you queued your job. osu-eddie!mail arnson (Date 11/04) The job will be deleted in several days if the problem is not corrected. If you care to kill the job, execute the following command: uustat -kosu-eddN1210 Sincerely, ihnp4!uucp ############################################# ##### Data File: ############################ >From drutx!jca Mon Nov 4 11:40:45 1985 remote from ihnp4 From: ihnp4!drutx!jca Received: by ihnp4.ATT.UUCP id AA17705; 4 Nov 85 11:40:45 CST (Mon) To: ihnp4!cbosgd!osu-eddie!arnson Subject: test /*/*/*/*/*/*/*/*/*/*/*/*/*/*/*/*/*/*/*/*/*/*/*/*/*/*/*/*/*/*/*/*/*/*/*/*\ This is the mail header, etc. that I receive FROM osu-eddie: From news Tue Dec 3 08:12 EST 1985 >From cbosgd!osu-eddie!arnson Tue Dec 3 04:10:12 1985 remote from ihnp4 Received: by ihnp4.ATT.UUCP id AA04621; 3 Dec 85 04:10:12 CST (Tue) Received: by osu-eddie (4.12/3.0) id AA06348; Mon, 2 Dec 85 23:05:20 est Date: Mon, 2 Dec 85 23:05:20 est From: ihnp4!osu-eddie!arnson (Jill C. Arnson) Message-Id: <8512030405.AA06348@osu-eddie> To: cbosgd!ihnp4!drutx!jca Subject: test1 Status: R this is a test of Mail 9:05 MST -- jill c. arnson ihnp4!drutx!jca AT&A IS, Denver (303)538-4800 Lily Tomlin says that the key element in power dressing is "...sort of like a scarf, sort of like a tie, sort of like a ruffle, and it doesn't threaten anyone because you don't look good in it." From @u(The Search for Intelligent Life in the Universe) (-:
glc@akgua.UUCP (G.L. Cleveland [Lindsay]) (12/13/85)
More than once has a site connected to "akgua" been cursed with the
problem of mail only flowing in one direction. It usually turns
out that:
1. It is a new site that is using the UUCP control tables
"USERFILE" or "Permissions" as received on the release tape.
2. It is a site that has just upgraded their UUCP to the
Honey-Danbur version (changing from "USERFILE" to "Permissions")
3. Somebody decided to "tighten up security" and started mucking
around with "USERFILE" or "Permissions".
By manipulation of the above-mentioned control table, it is quite
possible to create a set of permissions that only allow mail to be
sent but never received. The message from the "refused" site that
gets sent to the mailer ("...unable to contact machine...") is
misleading.
I suggest that the UUCP maintainers of both sites involved get on
the phone, run some tests with the "debug mode" on, i.e.
/usr/lib/uucp/uucico -r1 -x8 -scbosgd &
and watch what happens. Of course, a review of the permission
table might also reveal the problem, but that's not as much fun as
watching "uucico" run!
(Of course this could all be some other problem, like one site having
their telephones turned off during the holidays because "nobody
will be using them".)
Cheers,
Lindsay
Lindsay Cleveland (akgua!glc) (404) 447-3909 Cornet 583-3909
AT&T Technologies/Bell Laboratories ... Atlanta, Gagjm@packard.UUCP (Gary J. Murakami) (12/15/85)
While I am no longer officially supporting ihnp4, I check on its status occationally. In this case, a test to osu-eddie resulted in "login failed." Somehow ihnp4!action didn't get (or maybe install) an update for login/password changes at osu-eddie. FYI, ihnp4 has over 1500 connections and over 200 (13%) of the connections are broken! Some of these "broken" connections are new and ihnp4 just hasn't been able to get around to testing them. Some of the problems (1-2 dozen) are due to the REMOTE REJECT YOU ARE UNKNOWN TO ME problem that comes as the default "secure" configuration for HDB UUCP as mentioned by Lindsay. But by far, most of the problems are due to poor administration, e.g. administrators that dont send in updates for changes and dont return phone calls ( - note that ihnp4 hasn't changed its public access connection information in years). I used to make/answer a many dozens of phone calls and mail messages a week to try to fix broken connections. Even with that effort, there were over 100 bad connections at any given time. Currently, Kathy Andrews <ihnp4!action> juggles a multiple person job just handling the phone and mail updates from administrators. Entropy? If that is a term applicable to a small mail/UUCP site, how do you describe a major mail/UUCP gateway? A paradox? A singularity? ... More like a cross between a star and a black hole ... :-) -Gary