• Article Type: General
• Product: Aleph
• Product Version: 20

Description:
We have reached 200 Active GUI users (our limit) -- and are getting "License limit exceeded" messages.

We had a maintenance and had trouble bringing up the db. Is there anyway to check if these users are Idle or can they be flushed?

Resolution:
util a/17/14 for the vir01 z65 (pc sessions) table showed that the z65_id Oracle index was not present. Trying to build it by recreating the z65 via util a/17/1 failed due to duplicates. I was able to succeed by using util a/17/18.

The z65_id is supposed to be built as part of the re-initialization of the z65 by clear_vir01.

The clear_vir01 logs in $alephe_scratch show clear_vir01 running, as scheduled, at 05:15 on Aug. 1, 2, 3, and 4. Then we see no run again until today (Aug 20) at 11 am when we see 16 runs all at once:

-rw-r--r-- 1 aleph aleph 12536 Aug 1 05:15 vir01_clear_vir01.00777
-rw-r--r-- 1 aleph aleph 12536 Aug 2 05:15 vir01_clear_vir01.00778
-rw-r--r-- 1 aleph aleph 12536 Aug 3 05:15 vir01_clear_vir01.00779
-rw-r--r-- 1 aleph aleph 12536 Aug 4 05:15 vir01_clear_vir01.00781
-rw-r--r-- 1 aleph aleph 12544 Aug 20 11:06 vir01_clear_vir01.00782
-rw-r--r-- 1 aleph aleph 12900 Aug 20 11:06 vir01_clear_vir01.00783
-rw-r--r-- 1 aleph aleph 12544 Aug 20 11:07 vir01_clear_vir01.00784
-rw-r--r-- 1 aleph aleph 12877 Aug 20 11:07 vir01_clear_vir01.00785
-rw-r--r-- 1 aleph aleph 12544 Aug 20 11:08 vir01_clear_vir01.00786
-rw-r--r-- 1 aleph aleph 12544 Aug 20 11:08 vir01_clear_vir01.00787
-rw-r--r-- 1 aleph aleph 12698 Aug 20 11:09 vir01_clear_vir01.00788
-rw-r--r-- 1 aleph aleph 12544 Aug 20 11:09 vir01_clear_vir01.00789
-rw-r--r-- 1 aleph aleph 12544 Aug 20 11:10 vir01_clear_vir01.00790
-rw-r--r-- 1 aleph aleph 12704 Aug 20 11:11 vir01_clear_vir01.00791
-rw-r--r-- 1 aleph aleph 12704 Aug 20 11:11 vir01_clear_vir01.00792
-rw-r--r-- 1 aleph aleph 12544 Aug 20 11:12 vir01_clear_vir01.00793
-rw-r--r-- 1 aleph aleph 12544 Aug 20 11:12 vir01_clear_vir01.00794
-rw-r--r-- 1 aleph aleph 13020 Aug 20 11:13 vir01_clear_vir01.00795
-rw-r--r-- 1 aleph aleph 12544 Aug 20 11:13 vir01_clear_vir01.00796
-rw-r--r-- 1 aleph aleph 13027 Aug 20 11:14 vir01_clear_vir01.00797

This problem occurred because the vir01 batch queue was stopped sometime on Aug. 4. The clear_vir01 jobs accumulated and ran all at once when the vir01 batch queue (lib_batch) was started again at 11 today.

KB 8192-482 describes how running clear_vir01 at a time when staff users are trying to access the server can result in the z65_id failing to be built.

The root problem is the stopping of the vir01 batch queue on Aug. 4. This is to be strictly avoided. The vir01 batch queue needs to be running whenever Aleph is up.

p.s. Checking the final clear_vir01 log (at 11:14), I saw that the z63_id1 and z63_id2 Oracle indexes on the z63 www_server sessions had also failed -- which, no doubt, would have caused OPAC problems soon --, so I did util a/17/18 for those indexes as well.

• Article last edited: 10/8/2013