Preventative maintenance questions *MASTER RECORD*
- Product: Aleph
- Product Version: 20, 21, 22, 23
- Relevant for Installation Type: Dedicated-Direct, Direct, Local, Total Care
Description:
There have never been procedures set up and documented here for preventive maintenance that should be run on a weekly basis. I'm in the process of trying to figure this out and I have a few questions.
I've been referring to the System Administrator's Guide-Preventative Maintenance and Recovery Policy as well as the System Administrator's Guide-Oracle and the page on UTIL X from version 18's system administrator guide.
From the Preventative Maintenance doc, section 2.1:
I should be monitoring the tablespaces weekly. What I have done is run UTIL O/14/1 to see all the tablespaces and then use UTILo/17/5 to view the total/free tablespace by tablespace. First of all, is there an easier way to look at this? It appears to me that we don't currently have any problems--could you confirm that?
From Preventative Maintenance doc, section 2.2:
a. Use UTIL A to clean temporary data tables. It appears that I should be running A/12, A/13, A/14 and A/16 on a weekly basis. Is that everything or am I missing something?
b. Clearing vir01--Util X/8. This says to run when the application is not active. Does aleph_shutdown* stop this application?
From Preventative Maintenance doc, section 7:
Use Util X (1-7 and 9-10 in addition to 8 above). I am assuming I run these after running aleph_shutdown*, correct?
General questions
a. As these have not been run for a long time, will they take a long time to run?
b. Another question related to log files. I've searched the knowledge base and documentation but I do not see information about log files that I should be checking on a regular basis.
Resolution:
Section 2.1: I don't know of easier way to view tablespace space than util O.
Section 2.2: a. You should *not* be doing util A/13, A/14 and A/16 on a regular basis. They are only for exceptional cases. A/12 should be included in the job_list as “util_a_12_c” (to run daily).
b. You don't want or need to clear vir01 using util x/8. You have the following entry in your $alephe_tab/job_list which is already handling this:
W3 05:00:00 Y VIR01 clear_vir01 VIR01
Section 7: Do util x 1, 2, and 9 only. Save the most recent month of $data_print. The most recent week of others. Do not do util x/3, 6, 7, 8, or 10 on a regular basis. Not necessary. See the util x section of the "ALEPH UTIL Menus" doc in the /19 folder on the Doc Portal.
Your $alephe_root/aleph_startup script already does cleanup of $TMPDIR and $LOGDIR:
#*******************************************
# Cleanup TMPDIR/LOGDIR
#*******************************************
if (! $?KEEP_LOGDIR) then
setenv KEEP_LOGDIR 7 # default for keeping logs is 7 days
endif
if (! $?KEEP_TMPDIR) then
setenv KEEP_TMPDIR 2 # default for keeping tmp is 2 days
endif
echo "Cleaning temporary files more that $KEEP_TMPDIR days old..."
echo " "
So doing util x/4 and 5 manually is unnecessary.
The Doc Portal (How_to_from_Support directory) document "How to add UTIL X to the job list" decribes how to add the util x functions to the job_list.
General questions:
a. These utilities take seconds (or, at the most, a few minutes).
b. You do not need to check log files on a regular basis. You may need to check a log file in connection with certain problems, as described in certain Knowledge Base records.
Most important: Do
> df -h
weekly to monitor the space in each filesystem -- especially /exlibris. You're at 68%. If it gets to 80%, start checking what's using it.
See also Knowledge Base records: 3976, 3975, and 3992. And the "Aleph Down-System Troubleshooting" document.
As suggested in KB's 3966 and 16384-19025, in $alephe_root/pc_server_defaults, we suggest changing:
setenv create_statistics Y
setenv pc_transactions_log Y
to:
setenv create_statistics N
setenv pc_transactions_log N
This will reduce potential space problems. The util_a_13_b entry in your job_list should then be removed.
Per KB 8192-2055, you should drop/disable the *bib* (xxx01) z35 which can grow to a very large size and is no longer needed.
Additional Info
Sites may want to delete older records from the xxx50 z35 (Circ events), z30h (item history, z36h (loan history), and z37h (request history), and the xxx01 z69 (OPAC transaction) tables. This would be done once all the statistics for a particular time-period have been gathered. Prior to version 22.1, it would need to be done via SQL, but with the 22.1 Minor release, rep_change 2020, util x/13 has been added do this.
- Article last edited: 12-Mar-2016