upl@puff.cs.wisc.edu (Future Unix Gurus) (09/10/89)
Hello,
Since people have been posting things about uucp lately, here's my
problem. I'm new to this uucp thing, so if this is a stupid error,
one that everyone gets, well I'm sorry for not knowing, but I need help
anyway.
This is what happens: I try to poll tbaas, but I get a Handshake error.
Any ideas? Here's my LOGFILE
tbaas!uucp (9/10-1:44:25) (C,4188,0) C.tbaas (bldfl)
tbaas!uucp (9/10-1:44:55) (C,4188,0) OK (DIAL ph0 P**phone # deleted** 23)
tbaas!uucp (9/10-1:45:03) (C,4188,0) SUCCEEDED (call to tbaas )
tbaas!uucp (9/10-1:45:05) (C,4188,0) Shere=tbaas (HERE)
tbaas!uucp (9/10-1:45:10) (C,4188,0) harrie (send Myname)
tbaas!uucp (9/10-1:45:14) (C,4188,0) HANDSHAKE FAILED (^M^M)
I'm also trying to feed another computer where I work. It is running
BNU, though (at least that's what they call it. Looks like HDB to me)
It's an IBM rt running AIX. When I try to poll my machine with it, it will
log in correctly, but after getting a "Valid Shere=harrie" I get an error,
(you guessed it) Handshake failed.
What is going wrong, and what can I do to correct it? Thanks in advance!!
(you guys know everything about how to fix my machine)
- sparkie
p.s. please e-mail responses to the following address, NOT the address
listed above:
___ ___ ___ ___ _ _ _ ___
/ __\| . \/ . \| . \| |/ /|_|| _ |
\___\| __/| || _ /| < | || _[ sparkie@uhura.cs.wisc.edu
\___/|_| |_|_||_|\\|_|\_\|_||___|kls@ditka.UUCP (Karl Swartz) (09/12/89)
In article <2955@puff.cs.wisc.edu> upl@puff.cs.wisc.edu (Future Unix Gurus) writes: >I'm also trying to feed another computer where I work. It is running >BNU, though (at least that's what they call it. Looks like HDB to me) >It's an IBM rt running AIX. BNU (Basic Networking Utilities) and HDB (HoneyDanBer) are one and the same. And if your RT is running AIX 2.2.1, the current release for the RT, you've got HDB. Or something that started out as HDB. Somehow IBM managed to introduce an enormous number of bugs into it (most of the rest of AIX 2.2.1 too). I'm not saying that this is your problem in this case, but I've spent many hours trying to coax AIX 2.2.1's HDB into doing what any decent uucp should be able to do without any problem, and given a problem that involved it (or any other part of 2.2.1) I would *always* suspect the IBM software first. -- Karl Swartz |UUCP {ames,lll-winken}!pacbell!ditka!kls 1-505/672-3113 |Internet kls@rt1.lanl.gov |BIX kswartz "I never let my schooling get in the way of my education." (Twain)
randy@cctb.wa.com (Randy Orrison) (09/13/89)
In article <4184@ditka.UUCP> kls@ditka.UUCP (Karl Swartz) writes: | In article <1218@mitisft.Convergent.COM> dold@mitisft.Convergent.COM (Clarence Dold) writes: | >If two systems are logging in using two different LOGNAMEs, they must | >have different user id's. | Wrong. Each of my uucp connections has a separate login, e.g. | Ufoo for site foo and Ubar for site bar, all sharing the same | uid. Works fine with HDB, worked fine (as much as anything | worked fine) with the brain-dead stock uucp. So far, Karl is right. All uucp logins *can* have the same uid, there's no problem with this. However, | In fact, each uucp login must have the *same* uid, as many | files which uucico needs to read are only readable by uucp. This isn't true, since uucico is setuid (or should be) to uucp. It is quite permissable for each login to have a different uid. This is true for both the stock uucp and HDB. -randy -- Randy Orrison Chemical Computer Thinking Battery uunet!microsoft!cctb!randy randy@cctb.wa.com "Paradise is exactly like where you are right now, only much, much, better."
kls@ditka.UUCP (Karl Swartz) (09/15/89)
In article <1989Sep13.160258.16400@cctb.wa.com> randy@cctb.wa.com (Randy Orrison) writes: >In article <4184@ditka.UUCP> kls@ditka.UUCP (Karl Swartz) writes: >| In fact, each uucp login must have the *same* uid, as many >| files which uucico needs to read are only readable by uucp. >This isn't true, since uucico is setuid (or should be) to uucp. Indeed, you're quite correct. I spaced the fact that uucico (and uuxqt) are setuid to uucp and thus there's no reason for each uucp login to have the same uid. -- Karl Swartz |UUCP {ames,lll-winken}!pacbell!ditka!kls 1-505/672-3113 |Internet kls@rt1.lanl.gov |BIX kswartz "I never let my schooling get in the way of my education." (Twain)
les@chinet.chi.il.us (Leslie Mikesell) (09/16/89)
In article <1989Sep13.160258.16400@cctb.wa.com> randy@cctb.wa.com (Randy Orrison) writes: >| >If two systems are logging in using two different LOGNAMEs, they must >| >have different user id's. >| Wrong. Each of my uucp connections has a separate login, e.g. >| Ufoo for site foo and Ubar for site bar, all sharing the same >| uid. Works fine with HDB, worked fine (as much as anything >| worked fine) with the brain-dead stock uucp. > >So far, Karl is right. All uucp logins *can* have the same uid, there's >no problem with this. However, My experience with HDB is that different systems using different LOGNAMEs must have different uid's if you want to match the correct entry in the PERMISSIONS file. That is, the PERMISSIONs entry will be found by the first match of the uid in the passwd file instead of matching the actual login name (which could be found by the LOGNAME environment variable but would be trivial to fake). Otherwise, and with non-HDB it shouldn't make any difference since uucico is set-uid and is normally started by random users sending mail etc. Les Mikesell