ue_21 stops creating z00p's

Last updated
Save as PDF
Share
1. Share
2. Tweet
3. Share

Product: Aleph
Product Version: 20, 21, 22, 23
Relevant for Installation Type: Dedicated-Direct, Direct, Local, Total Care

Description:
Since 5 March, the regularly scheduled p-publish-06 updates for Primo have failed to create a tar file. I managed to recuperate all updated records by using a date range, but I am unable to update the records for subsequent publications. In order to set the update flag to Y, one must specify the entire range of documents or the last handled date.

This is for our abc01 PRIMO-ABC dataset.

Is my only option at this point to run p_publish_06 on the entire range of documents with the update flag set to Y?

We have no idea why publish_06 has failed to produce any tar files since 5 March.

Resolution:
This is like the article "Aleph Z00P records not Written when xxx01 Records updated" (KB 16384-14146).

The $aleph_root/tab/tab_publish_timestamp has:
...
PRIMO-ABC 201003042038413
...

The following SQL shows that no PRIMO-ABC Z00Ps have been updated since 2010-03-04:

SQL-ABC50>> select count(*) from z00p where Z00P_SET = 'PRIMO-ABC' and substr (Z00P_TIMESTAMP,1,8) > '20100304';
COUNT(*)
----------
0

So the problem is not p_publish_06 but rather the ue_21 publishing daemon (which updates Z00P).

First. Is ue_21 running in the xxx01 and xxx30 libraries? It runs in these libraries, *not* in the $usr_library.

We see that the abc01 tab_publish has a correct entry:!

1 2 3 4 5
!!!!!!!!!!!!!!!!!!!!-!!!!!!!!!!!!!!!!!!!!-!-!!!!!-!!!!!!!!!!!!!!!
PRIMO-ABC ABC01PUB N PRIMO MARC_XML

but we see this in the abc01 $data_scratch:

ABC01-ALEPH>> ls -lrt run_e_21*
-rw-rw-r-- 1 aleph exlibris 6863720 Dec 10 12:08 run_e_21.29656
-rw-rw-r-- 1 aleph exlibris 6504630 Dec 26 14:18 run_e_21.10254
-rw-r--r-- 1 aleph exlibris 5585755 Jan 22 19:13 run_e_21.9948
-rw-rw-r-- 1 aleph exlibris 38067189 Mar 4 15:38 run_e_21.4023

ue_21 stopped processing records on 4 March.

We see this in the abc00 usr_library:

SQL-ABC00>> select count(*) from z00p;
COUNT(*)
----------
1121357

Though ue_21 shows as running in the abc01 UTIL-C-1, it clearly has not been processing records.

This command shows that it has been running since 23 Jan:

ABC01-ALEPH>>ps -ef | grep e_21 | grep ABC01
aleph 4097 1 0 Jan23 ? 05:19:37 /exlibris/aleph/a18_1/aleph/exe/rts32 ue_21_a ABC01.a18_1

I suggest that you stop ue_21 (UTIL-E-22) and then restart it (UTIL-E-21). That *should* get it going again. If not, then kill the process shown above in the ps -ef and *then* restart it with UTIL-E-21.

As described in article "How often should Aleph be shut down/restarted?" (KB 16384-21105), we recommend (even if you are not doing a cold backup) that you stop/restart Aleph once each week. We recommend this to prevent exactly the kind of situation you are seeing.

Article last edited: 12-Mar-2016