From epeisach@MIT.EDU Tue Oct 1 12:08:10 1996
Received: from MIT.EDU (SOUTH-STATION-ANNEX.MIT.EDU [18.72.1.2]) by rt-11.MIT.EDU (8.7.5/8.7.3) with SMTP id MAA02799 for <bugs@RT-11.MIT.EDU>; Tue, 1 Oct 1996 12:08:10 -0400
Received: from KANGAROO.MIT.EDU by MIT.EDU with SMTP
id AA14144; Tue, 1 Oct 96 12:08:09 EDT
Received: by kangaroo.mit.edu; (5.65/1.1.8.2/08Mar96-0212PM)
id AA10446; Tue, 1 Oct 1996 12:08:07 -0400
Message-Id: <9610011608.AA10446@kangaroo.mit.edu>
Date: Tue, 1 Oct 1996 12:08:07 -0400
From: epeisach@MIT.EDU
Reply-To: epeisach@MIT.EDU
To: krb5-bugs@MIT.EDU
Subject: kadm5 sapi.2 unit test fails 104/105
X-Send-Pr-Version: 3.99
System: OSF1 kangaroo.mit.edu V3.2 214 alpha
Machine: alpha
The kadm5 server unit tests fails in init-v2.exp tests 104 and
105. These are related to the ADB name and lock file.
in kadm5/alt_prof.c the stanza at KADM5_CONFIG_ADBNAME (line 376).
a) KADM5_CONFIG_ADBNAME is using a name based upon the dbname. You
cannot pass in a name to use - the dbname is used as the basis...
b) The tests try setting KADM5_CONFIG-ADBNAME and
KADM5_CONFIG_ADB_LOCKFILE. The lockfile optoin is ignored - you cannot
specify it. The config_adbname is essentially ignored as well...
c) There is a bug - if my params.mask contains KADM5_CONFIG_ADBNAME, but
not KADM5_CONFIG_DBNAME, the admin_dbname would not be set properly..
Fixes:
a & b) I believe tests 104 and 105 should be removed from the test suite. They
are no longer relevant.
c) params.admin_dbname needs to be set if KADM5_CONFIG_DBNAME or
KADM5_CONFIG_ADBNAME is set.
From: "Barry Jaspan" <bjaspan@MIT.EDU>
To: krb5-bugs@MIT.EDU, epeisach@MIT.EDU
Cc: Subject: Re: krb5-admin/46: kadm5 sapi.2 unit test fails 104/105
Date: Tue, 1 Oct 1996 12:20:09 -0400
Geez, I was sure I ran all the kadm5 unit tests after making this code
change, but I guess I didn't. The bottom line is this: admin_dbname
and admin_lockfile are no longer separately specifiable; they *always*
derive from the value of dbname. Tests which conflict with this
assertion are obsolete and should be fixed.
a) KADM5_CONFIG_ADBNAME is using a name based upon the dbname. You
cannot pass in a name to use - the dbname is used as the basis...
That's correct.
b) The tests try setting KADM5_CONFIG-ADBNAME and
KADM5_CONFIG_ADB_LOCKFILE. The lockfile optoin is ignored - you cannot
specify it. The config_adbname is essentially ignored as well...
Yup. The tests are obsolete.
c) There is a bug - if my params.mask contains KADM5_CONFIG_ADBNAME, but
not KADM5_CONFIG_DBNAME, the admin_dbname would not be set properly..
That's not a bug, because KADM5_CONFIG_ADBNAME is always ignored if
specified (see the updated api functional specification).
a & b) I believe tests 104 and 105 should be removed from the test
suite. They are no longer relevant.
Right.
c) params.admin_dbname needs to be set if KADM5_CONFIG_DBNAME or
KADM5_CONFIG_ADBNAME is set.
Wrong.
Does this all make sense to you?
From: Ezra Peisach <epeisach@MIT.EDU>
To: "Barry Jaspan" <bjaspan@MIT.EDU>
Cc: krb5-bugs@MIT.EDU
Subject: Re: krb5-admin/46: kadm5 sapi.2 unit test fails 104/105
Date: Tue, 01 Oct 1996 12:47:48 EDT
Yup - all makes sense... You care to make the change - or should I?
Ezra
From: "Barry Jaspan" <bjaspan@MIT.EDU>
To: epeisach@MIT.EDU
Cc: krb5-bugs@MIT.EDU
Subject: Re: krb5-admin/46: kadm5 sapi.2 unit test fails 104/105
Date: Tue, 1 Oct 1996 13:07:56 -0400
You are welcome to; I will do it if the it isn't done when I get
around to it, but I don't know when that will be.
Barry
State-Changed-From-To: open-closed
State-Changed-By: epeisach
State-Changed-When: Tue Oct 1 15:20:48 1996
State-Changed-Why:
I have removed tests 104 & 105 from init-v2.exp. Tests all pass.
Received: from MIT.EDU (SOUTH-STATION-ANNEX.MIT.EDU [18.72.1.2]) by rt-11.MIT.EDU (8.7.5/8.7.3) with SMTP id MAA02799 for <bugs@RT-11.MIT.EDU>; Tue, 1 Oct 1996 12:08:10 -0400
Received: from KANGAROO.MIT.EDU by MIT.EDU with SMTP
id AA14144; Tue, 1 Oct 96 12:08:09 EDT
Received: by kangaroo.mit.edu; (5.65/1.1.8.2/08Mar96-0212PM)
id AA10446; Tue, 1 Oct 1996 12:08:07 -0400
Message-Id: <9610011608.AA10446@kangaroo.mit.edu>
Date: Tue, 1 Oct 1996 12:08:07 -0400
From: epeisach@MIT.EDU
Reply-To: epeisach@MIT.EDU
To: krb5-bugs@MIT.EDU
Subject: kadm5 sapi.2 unit test fails 104/105
X-Send-Pr-Version: 3.99
Show quoted text
>Number: 46
>Category: krb5-admin
>Synopsis: kadm5 sapi.2 unit test fails 104/105
>Confidential: no
>Severity: serious
>Priority: medium
>Responsible: bjaspan
>State: closed
>Class: sw-bug
>Submitter-Id: unknown
>Arrival-Date: Tue Oct e 12:09:00 EDT 1996
>Last-Modified: Tue Oct e 15:21:33 EDT 1996
>Originator: Ezra Peisach
>Organization:
mit>Category: krb5-admin
>Synopsis: kadm5 sapi.2 unit test fails 104/105
>Confidential: no
>Severity: serious
>Priority: medium
>Responsible: bjaspan
>State: closed
>Class: sw-bug
>Submitter-Id: unknown
>Arrival-Date: Tue Oct e 12:09:00 EDT 1996
>Last-Modified: Tue Oct e 15:21:33 EDT 1996
>Originator: Ezra Peisach
>Organization:
Show quoted text
>Release: 1.0-development
>Environment:
>Environment:
System: OSF1 kangaroo.mit.edu V3.2 214 alpha
Machine: alpha
Show quoted text
>Description:
The kadm5 server unit tests fails in init-v2.exp tests 104 and
105. These are related to the ADB name and lock file.
Show quoted text
>How-To-Repeat:
gmake checkShow quoted text
>Fix:
in kadm5/alt_prof.c the stanza at KADM5_CONFIG_ADBNAME (line 376).
a) KADM5_CONFIG_ADBNAME is using a name based upon the dbname. You
cannot pass in a name to use - the dbname is used as the basis...
b) The tests try setting KADM5_CONFIG-ADBNAME and
KADM5_CONFIG_ADB_LOCKFILE. The lockfile optoin is ignored - you cannot
specify it. The config_adbname is essentially ignored as well...
c) There is a bug - if my params.mask contains KADM5_CONFIG_ADBNAME, but
not KADM5_CONFIG_DBNAME, the admin_dbname would not be set properly..
Fixes:
a & b) I believe tests 104 and 105 should be removed from the test suite. They
are no longer relevant.
c) params.admin_dbname needs to be set if KADM5_CONFIG_DBNAME or
KADM5_CONFIG_ADBNAME is set.
Show quoted text
>Audit-Trail:
From: "Barry Jaspan" <bjaspan@MIT.EDU>
To: krb5-bugs@MIT.EDU, epeisach@MIT.EDU
Cc: Subject: Re: krb5-admin/46: kadm5 sapi.2 unit test fails 104/105
Date: Tue, 1 Oct 1996 12:20:09 -0400
Geez, I was sure I ran all the kadm5 unit tests after making this code
change, but I guess I didn't. The bottom line is this: admin_dbname
and admin_lockfile are no longer separately specifiable; they *always*
derive from the value of dbname. Tests which conflict with this
assertion are obsolete and should be fixed.
a) KADM5_CONFIG_ADBNAME is using a name based upon the dbname. You
cannot pass in a name to use - the dbname is used as the basis...
That's correct.
b) The tests try setting KADM5_CONFIG-ADBNAME and
KADM5_CONFIG_ADB_LOCKFILE. The lockfile optoin is ignored - you cannot
specify it. The config_adbname is essentially ignored as well...
Yup. The tests are obsolete.
c) There is a bug - if my params.mask contains KADM5_CONFIG_ADBNAME, but
not KADM5_CONFIG_DBNAME, the admin_dbname would not be set properly..
That's not a bug, because KADM5_CONFIG_ADBNAME is always ignored if
specified (see the updated api functional specification).
a & b) I believe tests 104 and 105 should be removed from the test
suite. They are no longer relevant.
Right.
c) params.admin_dbname needs to be set if KADM5_CONFIG_DBNAME or
KADM5_CONFIG_ADBNAME is set.
Wrong.
Does this all make sense to you?
From: Ezra Peisach <epeisach@MIT.EDU>
To: "Barry Jaspan" <bjaspan@MIT.EDU>
Cc: krb5-bugs@MIT.EDU
Subject: Re: krb5-admin/46: kadm5 sapi.2 unit test fails 104/105
Date: Tue, 01 Oct 1996 12:47:48 EDT
Yup - all makes sense... You care to make the change - or should I?
Ezra
From: "Barry Jaspan" <bjaspan@MIT.EDU>
To: epeisach@MIT.EDU
Cc: krb5-bugs@MIT.EDU
Subject: Re: krb5-admin/46: kadm5 sapi.2 unit test fails 104/105
Date: Tue, 1 Oct 1996 13:07:56 -0400
You are welcome to; I will do it if the it isn't done when I get
around to it, but I don't know when that will be.
Barry
State-Changed-From-To: open-closed
State-Changed-By: epeisach
State-Changed-When: Tue Oct 1 15:20:48 1996
State-Changed-Why:
I have removed tests 104 & 105 from init-v2.exp. Tests all pass.
Show quoted text
>Unformatted:
kadm5 sapi tests fail 104/105