[Koha-devel] Out of sort memory, consider increasing server sort buffer size

[Philippe Blouin]

A month later...

Maryse found a (recommanded) solution for the aqspent.pl crashes:

-    aqorders.biblionumber, aqorders.basketno, aqorders.ordernumber,
-    quantity-quantityreceived AS tleft,
-    budgetdate, entrydate,
-    aqbasket.booksellerid,
+    MAX(aqorders.biblionumber) AS biblionumber, MAX(aqorders.basketno) 
AS basketno, aqorders.ordernumber,
+    MAX(quantity)-MAX(quantityreceived) AS tleft,
+    MAX(budgetdate) AS budgetdate, MAX(entrydate) AS entrydate,
+    MAX(aqbasket.booksellerid) AS booksellerid,
      GROUP_CONCAT(DISTINCT itype SEPARATOR '|') as itypes,
-    title,
-    aqorders.invoiceid,
-    aqinvoices.invoicenumber,
-    quantityreceived,
-    unitprice_tax_included,
-    datereceived,
-    aqbooksellers.name as vendorname
+    MAX(title) AS title,
+    MAX(aqorders.invoiceid) AS invoiceid,
+    MAX(aqinvoices.invoicenumber) AS invoicenumber,
+    MAX(quantityreceived) AS quantityreceived,
+    MAX(unitprice_tax_included) AS unitprice_tax_included,
+    MAX(datereceived) AS datereceived,
+    MAX(aqbooksellers.name) as vendorname

We'll hack that in all our distributions.  But I'd really like to have 
an official distributed solution to the problem.  To me, it's not that 
much uglier than the fix adding all the column to avoid the problem with 
ONLY_FULL_GROUP_BY.

What's your take on it?

Regards,

Philippe Blouin,
Directeur de la technologie

Tél.  : (833) 465-4276, poste 230
philippe.blouin at inLibro.com <mailto:philippe.blouin at inLibro.com>

inLibro | pour esprit libre | www.inLibro.com <http://www.inLibro.com>
On 2019-12-13 11:25 a.m., Philippe Blouin wrote:
>
> minuscule:
>
>   * 2000 biblio, 12000 items
>   * 8 aqbudgets, 200 invoices, 25 booksellers, 60 baskets, 240 orders.
>
> That makes little sense...
>
> Philippe Blouin,
> Directeur de la technologie
>
> Tél.  : (833) 465-4276, poste 230
> philippe.blouin at inLibro.com <mailto:philippe.blouin at inLibro.com>
>
> inLibro | pour esprit libre | www.inLibro.com <http://www.inLibro.com>
> On 2019-12-13 9:49 a.m., Jonathan Druart wrote:
>> How big is your tables?
>> As you can see SQL query is really ugly, and can be good to split it anyway...
>>
>> Le ven. 13 déc. 2019 à 15:26, Philippe Blouin
>> <philippe.blouin at inlibro.com>  a écrit :
>>> Thanks,
>>>
>>> Ok, I've never seen the "$field is not in group by" error.  Interesting.
>>>
>>> I've followed the bread crumbs: 21622 -> 20182 -> 21723 -> 22260... read it all.  It's "damned if you do, damned if you don't".
>>>
>>> ONLY_FULL_GROUP_BY mode is the SQL Standard, I don't think it would be wise to stray from that.  (I can do it locally, but I don't think Koha should, philosophically)
>>>
>>> But my client can't click on spent.pl.  And no memory setting seems to fix it.  So what's my solution?  Could the query be reworked in two steps?  This is not a screen where 0.2s will make a difference in the user experience.
>>>
>>> Suggestions welcomed.
>>>
>>> Philippe Blouin,
>>> Directeur de la technologie
>>>
>>> Tél.  : (833) 465-4276, poste 230
>>> philippe.blouin at inLibro.com
>>>
>>> inLibro | pour esprit libre |www.inLibro.com
>>> On 2019-12-13 4:33 a.m., Jonathan Druart wrote:
>>>
>>> Hi Philippe,
>>>
>>> See bug 21622.
>>> It is necessary to not explode with "$field is not in group by" (when
>>> sql_mode has ONLY_FULL_GROUP_BY)
>>> However we may need to discuss the need of this flag if it brings us
>>> performance issues.
>>>
>>> Cheers,
>>> Jonathan
>>>
>>> Le jeu. 12 déc. 2019 à 23:23, Philippe Blouin
>>> <philippe.blouin at inlibro.com>  a écrit :
>>>
>>> Good morning Koha,
>>>
>>> Newbie question: why do we have 16 fields in GROUP BYs ?  Are they really all necessary to eliminate duplicate lines in budgets?  Genuine question.
>>>
>>> When in acqui-home.pl, clicking certain entries linking to spent.pl just crashes with the error listed in the mail's object: Out of memory.  Although I hate that solution, I went with it, increased sort_buffer_size 64x fold (I'm a generous guy).  Still not enough.
>>>
>>> Removing most lines in GROUP BY of spent.pl solves it all, of course.
>>>
>>> --
>>> Philippe Blouin,
>>> Directeur de la technologie
>>>
>>> Tél.  : (833) 465-4276, poste 230
>>> philippe.blouin at inLibro.com
>>>
>>> inLibro | pour esprit libre |www.inLibro.com
>>> _______________________________________________
>>> Koha-devel mailing list
>>> Koha-devel at lists.koha-community.org
>>> https://lists.koha-community.org/cgi-bin/mailman/listinfo/koha-devel
>>> website :http://www.koha-community.org/
>>> git :http://git.koha-community.org/
>>> bugs :http://bugs.koha-community.org/
>
> _______________________________________________
> Koha-devel mailing list
> Koha-devel at lists.koha-community.org
> https://lists.koha-community.org/cgi-bin/mailman/listinfo/koha-devel
> website : http://www.koha-community.org/
> git : http://git.koha-community.org/
> bugs : http://bugs.koha-community.org/
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://lists.koha-community.org/pipermail/koha-devel/attachments/20200110/7e62969c/attachment.html>