[Building Sakai] Timezone for events

[David Horwitz]

As a test I did some experimenting on mysql and I think this shows that 
mysql was also vulnerable to the DST change issue in clusterevent 
tracking as oracle was (KNL-725) and theoritical the session service has 
the same issue:


mysql> create table test (field1 datetime, field2 timestamp);
Query OK, 0 rows affected (0.06 sec)

mysql> insert into test values (now(), now());
Query OK, 1 row affected (0.00 sec)

mysql> select * from test;
+---------------------+---------------------+
| field1              | field2              |
+---------------------+---------------------+
| 2011-05-23 17:12:53 | 2011-05-23 17:12:53 |
+---------------------+---------------------+
1 row in set (0.00 sec)


mysql> set time_zone = '-2:00';
Query OK, 0 rows affected (0.00 sec)

mysql> select * from test;
+---------------------+---------------------+
| field1              | field2              |
+---------------------+---------------------+
| 2011-05-23 17:12:53 | 2011-05-23 13:12:53 |
+---------------------+---------------------+
1 row in set (0.00 sec)





On 05/23/2011 03:55 PM, Stephen Marquard wrote:
> Oracle users may know their events table to be in GMT, but mysql users
> know it to be in localtime. Our SAKAI_EVENT records are in GMT+2 which
> is our local timezone.
>
> So it looks like changing it will inevitably break any external
> assumptions for at least one set of institutions.
>
> Regards
> Stephen
>
>