p_serial_44: Oracle error: fetch z30; variable not in select list

• Article Type: General
• Product: Aleph
• Product Version: 19.01

Description:
In our testing of v. 19, when we ran the monthly reports on Feb. 1, one of the 28 schools got the Oracle error:
Oracle error: fetch z30
ORA-01007: variable not in select list
in the p_serial_44 job. The error was not seen in any school but JKL. We've not gotten this error before. We then re-ran the reports on 2/7, and again got the errors:

KB 4045 does not seem to apply here (we do not want to drop the z30 table!). We're wondering if there is a file missing in the this ADM.

Resolution:
Diagnostics show that the "Oracle error: fetch z30" message is coming from the ./com/expand_docx program:

./butil/b_serial_44_c_2 calls

./form_fill/form_fill_bib_info_xml, which calls

./com/get_bib_line_x, which calls

./com/expand_docx.

get_bib_line_x is calling expand_docx with the xxx01 as the active library and with a label of "HOLDING". The abc01 tab_expand has this for HOLDING:

HOLDING expand_doc_adm_bib
! testing for myshelf
HOLDING expand_doc_bib_loc_1_a
HOLDING expand_doc_bib_loc_1_b
HOLDING expand_doc_bib_loc_1_c
HOLDING expand_doc_sort_loc_b
HOLDING expand_doc_bib_loc_2_a
HOLDING expand_doc_bib_loc_3_a

I'm pretty certain that if these lines were commented out, you would not see this error -- but I suspect that you would not want to do that since these HOLDING lines serve other purposes (such as, myshelf).

The expand_docx HOLDING is being called so that the expanded fields can be used with the bib format specified in the ADM bib_format table. The jkl50 bib_format file has "012" for the "serials-claim-report". And the abc01 edit_paragraph.eng table has this for 012:

! Author-Title for Bib Inf on ITEM Items
012 1#### D ^:^
012 245## 9
012 260## W ^
012 300## D ^

As you can see, there are no holdings fields in this edit_paragraph entry. Thus, the fact that the preceding tab_expand has not included any PST lines is not a problem. And, in fact, we see that proper entries *are* produced on the report for all 7 of the items which get the fetch error. (Each "fetch error" is triggered by one particular item.)

It's possible that we will eventually arrive at change which prevents the fetch error. If that happens, I will certainly let you know. But since there is no practical consequence to the error, I am closing this SI.

[See KB 4045 for a discussion of Oracle fetch errors in general.]