[cle-release-team] [sakai2-tcc] 2.9.0 key issues [short] list

[Steve Swinsburg]

Hi all,

Those delete lines will force me to -1 the release. There are people that depend on that data and the old site stats code still uses it. There is a conversion for that (STAT-299) and its going to be forthcoming for Oracle shortly (its been available for MySQL for a while) but dumping that data will cause major issues.

cheers,
Steve


On 01/11/2012, at 7:04 AM, Seth Theriault <slt at columbia.edu> wrote:

> Hello,
> 
> I agree that the wholesale deletes are bad, but backup tables would be
> really good in this case.
> 
> Seth
> 
> On Wed, Oct 31, 2012 at 4:00 PM, Matthew Jones <matthew at longsight.com> wrote:
>> It also seems like the conversion script having the lines
>> 
>> DELETE FROM SAKAI_EVENT;
>> DELETE FROM SAKAI_SESSION;
>> 
>> Are bad ideas, especially for schools that don't archive these tables.
>> 
>> Ideally it should have had
>> CREATE TABLE SAKAI_EVENT_KNL_734_BAK AS (SELECT * FROM SAKAI_EVENT);
>> CREATE TABLE SAKAI_SESSION_KNL_734_BAK AS (SELECT * FROM SAKAI_SESSION);
>> 
>> Note these deletes are not the in mysql scripts.
>> 
>> I'm not sure if going to TIMESTAMP WITH LOCAL TIME ZONE still even requires
>> this deletion since the data isn't altered like it is with the other format?
>> 
>> On Wed, Oct 31, 2012 at 3:54 PM, Matthew Jones <matthew at longsight.com>
>> wrote:
>>> 
>>> I created this as a new issue https://jira.sakaiproject.org/browse/KNL-985
>>> and plan on changing it to TIMESTAMP WITH LOCAL TIME ZONE. I also think it
>>> would be convenient to test this over the weekend as well when the time
>>> changes if possible. If it didn't work we could change it back before the
>>> final 2.9.0 release.
>>> 
>>> 
>>> On Wed, Oct 31, 2012 at 1:06 PM, Seth Theriault <slt at columbia.edu> wrote:
>>>> 
>>>> On Wed, Oct 31, 2012 at 12:54 PM, Matthew Jones <matthew at longsight.com>
>>>> wrote:
>>>> 
>>>>> Converting to TIMESTAMP WITH LOCAL TIME ZONE could potentially resolve
>>>>> both
>>>>> clustering and DST, or it might break still break the clustering.
>>>> 
>>>> Personally, I'd like to see the above type tested since we already
>>>> have reports of breakage using TIMESTAMP WITH TIME ZONE.
>>>> 
>>>> Seth
>>> 
>>> 
>> 
> _______________________________________________
> cle-release-team mailing list
> cle-release-team at collab.sakaiproject.org
> http://collab.sakaiproject.org/mailman/listinfo/cle-release-team