Re: cvs-1.11.2 "make check" failure

[Paul Jarc]

lawrence.jones@eds.com (Larry Jones) wrote:
> Read TESTS for guidance.

Ok, check.log ends with:

PASS: tag8k-16
cvs: rcs.c:1095: rcsbuf_getkey: Assertion `ptr >= rcsbuf_buffer && ptr < 
rcsbuf_buffer + rcsbuf_buffer_size' failed.
cvs [tag aborted]: received abort signal
exit status was 1
FAIL: tag8k-17

And I see in your other message that this is already fixed.  Sorry for
the noise.

> CVS uses gmtime(), which should not be affected by the timezone, but
> on your system it apparently is.

It's affected by the choice of timescale.

> Since you've chosen to break your system (TAI is "right" for many
> things, but the system clock is not one of them), I think you just
> need to deal with it yourself.

I don't agree that it's broken; FWIW, all the problems I've run into
with this configuration that I can remember are simply test suite
failures like this one.  Anyway, adding "TZ=UTC; export TZ" to
sanity.sh would also help for this issue mentioned in TESTS:
  Also, some of the tests expect your local timezone to be an integral
  number of hours from UTC -- if you usually use a fractional
  timezone, use a different (integral) timezone when running the tests
  to avoid spurious failures.
Is there any harm in adding "TZ=UTC; export TZ"?


paul