Grouper Users - Open Discussion List

[Baron Fujimoto <>]

Thanks, this looks promising. I got as far as running 
edu.internet2.middleware.grouper.misc.SyncPITTables().syncAllPITTables(), but 
encountered syntax errors[*] trying to create GROUPER_CHANGE_LOG_TEMP_V... 
possibly due to our older version of MySQL. I'll see if I can sort them out 
and find workarounds to continue.

[*] E.g.:

ERROR 1064 (42000) at line 2: You have an error in your SQL syntax; check the 
manual that corresponds to your MySQL server version for the right syntax to 
use near 'BEQUEATH DEFINER
AS
SELECT gcle.created_on,
gclt.change_log_categor' at line 33


On Fri, Jan 25, 2019 at 08:16:10PM +0000, Shilen Patel wrote:
>Does it seem like this issue?
>
>https://spaces.at.internet2.edu/display/Grouper/Change+log+and+point+in+time+ordering+issue
>
>If so, it was fixed in a 2.3 patch.  But the wiki tells you how to update 
>the timestamp in the temp changelog to get past it if you don’t have the 
>patch.
>
>- Shilen
>
>On 1/25/19, 2:57 PM, 
>"
> on behalf of Baron Fujimoto" 
><
> on behalf of 
>>
> wrote:
>
>    Every so often when running 
> loaderRunOneJob("CHANGE_LOG_changeLogTempToChangeLog") from gsh, we 
> encounter errors aprently due to table/entry inconsistencies that I have to 
> manually try to hunt down and clean up (I don't know what the root cause of 
> these inconsistencies might be).
>    
>    Today, the current exception is:
>    
>    // Error: unable to evaluate command: Sourced file: inline evaluation 
> of: ``loaderRunOneJob("CHANGE_LOG_changeLogTempToChangeLog");'' : Error 
> invoking compiled command: : Error in compiled command: 
> java.lang.RuntimeException: java.lang.RuntimeException: Active PITMember 
> with sourceId=4d799e6a04594a128af363db89478f3b not found,
>    Problem in HibernateSession: HibernateSession (4087f108): new, 
> notReadonly, READ_WRITE_NEW, notActiveTransaction, session (1e769d81)
>    
>    I'm not sure where to find this offending entry though. I've attempted 
> to search the grouper_pit_* tables for a source_id column with that value 
> without success, but perhaps/probably it's in some other table under 
> different column that is being used as the source_id? Any tips tracking 
> this down?

-- 
Baron Fujimoto 
<>
 :: UH Information Technology Services
minutas cantorum, minutas balorum, minutas carboratum desendus pantorum