[OPEN-ILS-GENERAL] Holds for ILLs on Template bib records

[Tina Ji (Project Sitka)]

Hi all,

Mike's "injest" theory may explain my recent experience.

I merged two bib records, one with wait-for-copy holds but no  
available copy, the other with available copy. In my first case, in 10  
min a target copy had been assigned. In my second case, it took 2 days.

I have heard of that after a hold is initially targetted with no  
success, it will be retargetted once only everyday in the following  
days. I am not sure whether this is Sitka particular. But it helps  
explain why it takes longer time.


Tina Ji
Trainer/Help Desk Specialist
BC SITKA Evergreen Implementation



Quoting Mike Rylander <mrylander at gmail.com>:

> On Thu, Aug 20, 2009 at 6:36 AM, Sharp,
> Chris<csharp at georgialibraries.org> wrote:
>> Hi Marlene,
>>
>>> We are trying to place Title holds, and they are for the same org unit
>>> owning the item (we do have branches). We were aware that the
>>> "targeter" was on a 15 minute delay, but the delay in finding the item
>>> is much more than that, sometimes a day or two days
>>
>> The targeter program sometimes takes much longer than every 15   
>> minutes.  The script looks to see if the targeter is already   
>> running (by checking for the existence of the   
>> /tmp/hold_targeter-LOCK file), and if the lockfile is present, it   
>> checks again 15 minutes later.  In PINES, depending on the volume   
>> of holds placed, this can take hours rather than the expected 15   
>> minutes.
>>
>
> The 15 minute time (well, strictly speaking, the configured
> re-targeting run time) is just for re-targeting after the targeting
> interval has expired.  Initial targeting occurs when the hold is
> placed, so something else is going on here.
>
> I suspect that this is another side-effect of the intermittent ingest
> issue that has shown up recently.  If the record fails to ingest
> properly then some required data (extracted from the fixed fields, and
> used to determine the record type) won't be there and can cause the
> targeting to exit early.  There is a more aggressive workaround than
> the hourly cleanup (Marlene, this has been proposed to your
> consortium, in fact) that we believe will address this until we have
> completely pinned down the root cause of this intermittent issue.
>
> --
> Mike Rylander
>  | VP, Research and Design
>  | Equinox Software, Inc. / The Evergreen Experts
>  | phone:  1-877-OPEN-ILS (673-6457)
>  | email:  miker at esilibrary.com
>  | web:  http://www.esilibrary.com
>