Re: [PATCH] t3306,t5304: avoid clock skew issues

[Michael J Gruber <git@drmicha.warpmail.net>]

Junio C Hamano venit, vidit, dixit 08.04.2011 22:20:
> Michael J Gruber <git@drmicha.warpmail.net> writes:
> 
>> On systems where the local time and file modification time may be out of
>> sync (e.g. test directory on NFS) t3306 and t5305 can fail because prune
>> compares times such as "now" (client time) with file modification times
>> (server times for remote file systems). I.e., these are spurious test
>> failures.
>>
>> Avoid this by setting the relevant modification times to the local time.
>>
>> Noticed on a system with as little as 2s time skew.
>>
>> Signed-off-by: Michael J Gruber <git@drmicha.warpmail.net>
>> ---
>> I don't think we can safeguard prune itself properly. A filesystem may be out
>> of time sync sporadically (and produce incorrect time stamps) but then be in
>> sync when "git prune" is run so that we can't detect it.
>>
>>  t/t3306-notes-prune.sh |    4 ++++
>>  t/t5304-prune.sh       |    3 ++-
>>  2 files changed, 6 insertions(+), 1 deletions(-)
>>
>> diff --git a/t/t3306-notes-prune.sh b/t/t3306-notes-prune.sh
>> index c428217..3114972 100755
>> --- a/t/t3306-notes-prune.sh
>> +++ b/t/t3306-notes-prune.sh
>> @@ -20,6 +20,10 @@ test_expect_success 'setup: create a few commits with notes' '
>>  	git add file3 &&
>>  	test_tick &&
>>  	git commit -m 3rd &&
>> +	COMMIT=$(git rev-parse HEAD) &&
>> +	COMMIT_FILE=.git/objects/$(echo $COMMIT | sed "s/^../&\//") &&
> 
> Hmm, the remainder of the test script seems to know this commit is 5ee1c35
> all over the place.  Do the above two lines worth it?

I found that use of concrete sha1s a bit strange (I didn't run blame...)
but have no objection against keeping the new bit in line with the
existing test, i.e.

+	COMMIT_FILE=.git/object/5e/e1c35e83ea47cd3cc4f8cbee0568915fbbbd29 &&

instead of the 2 lines above.

Michael