[1494] in Kerberos_V5_Development
Re: util/db2 fails make check on Linux
daemon@ATHENA.MIT.EDU (Sam Hartman)
Sun Aug 4 01:30:41 1996
To: Mark Eichin <eichin@MIT.EDU>
Cc: Sam Hartman <hartmans@MIT.EDU>, krbcore@MIT.EDU, marc@MIT.EDU
From: Sam Hartman <hartmans@MIT.EDU>
Date: 04 Aug 1996 01:30:31 -0400
In-Reply-To: Mark Eichin's message of 03 Aug 1996 20:10:18 -0400
>>>>> "Mark" == Mark Eichin <eichin@MIT.EDU> writes:
Mark> This was a problem with the old db as well. With the old db,
Mark> in fact, some tests would fail if /usr/dict/words (or one of
Mark> the several it looked for in its place) was under 25,000
Mark> lines, or some threshhold like that. We should just find a
Mark> sufficiently free wordlist and ship it as part of the test.
I would like to clarify that the test still failed after I
installed a word list. The point of mentioning the word list was to
explain why I had not run the tests earlier.
Is there some way we could construct a test file from the
Kerberos source code itself or something; we're shipping enough text
around that it is kind of silly to go adding another dictionary.
