Delivery-Date: Mon, 21 Mar 94 07:46:44 -0800
Message-Id: <199403211546.HAA25898@nobozo.CS.Berkeley.EDU>
To: aoki@postgres.Berkeley.EDU (Paul M. Aoki)
Subject: Re: more regression test info re:hpux port 
In-Reply-To: Your message of "Sat, 19 Mar 1994 22:31:01 MST."
             <199403200631.WAA28128@faerie.CS.Berkeley.EDU> 
Date: Mon, 21 Mar 1994 08:46:37 -0700
From: Daniel Simmons <simmdan@kenya.isu.edu>


In message <199403200631.WAA28128@faerie.CS.Berkeley.EDU> you write:
>> Before I ran the regression test again, I went in as simmdan and did
>> bmake clean all so that simmdan would own things.  While the test was
>> running, though, I noticed that postgres was ending up the owner of
>> onek.data and stud_emp.data so I changed them to be group readable and
>> writable from another window while the tests were running before it
>> got to the place near the end where it died before.  So, the tests
>> went to completion as well.  
>
>those don't have to be writable, just readable (it does a write then
>read), but the obj directory has to be writable.  i think the 
>distribution comes with 775 directories.  it should probably have 
>777 obj directories, at least in regression.

Well, they were read/write for owner only (400), and were owned by
postgres even though I was running the tests as simmdan.  It looked to
me like the files were created by the backend and therefore owned by
postgres even though the frontend was being run by simmdan.


>> I'm thinking that there's something wrong with the way the tests work,
>> though.  It doesn't real seem that I should have to jump through those
>> kinds of hoops to get them to run...
>
>oddly, i've never, ever received any problem reports about file 
>permissions during the regression tests -- most people install it
>and run it as themselves, i imagine (or ignore the warning and 
>run it as postgres, causing a different problem).  i do seem to recall 
>that nfs file systems on hpux seem to have various types of braindeath 
>with respect to permissions, though.  i would mount ultrix-exported
>file systems and be told i didn't own my own files..  groups didn't
>work right and i had to make things 777..  that kind of thing.

All of this stuff is nfs mounted, but we haven't had too many problems
of the kind you mention.  I installed everything as postgres so that
it would own it all (seemed to make sense at the time).  At any rate,
I'm not going to worry about it.  It looks like everything is working
now, and if there is some small permissions problem which comes back
to haunt me later I'll worry about it then (I doubt that will happen,
though).

>> Anyway, at the end of the message is my latest checkdiff output.  It
>> looks like it's mostly ok, but I'll let you be the final judge.
>
>looks ok to me.

Thanks so much for your help.  Now I can proceed with my project with
some amount of confidence.

Danny

-------
Daniel Simmons          electronic mail : simmdan@isu.edu
Idaho State University       voice mail : (208) 236-3199
Computer Center              snail mail : Box 8037, Pocatello