[ECODP-dev] quick fix for the production database

[Bert Van Nuffelen]

Dear,

our investigations has lead to the first conclusion. The regular upload of
(new) data increased the CKAN database table activity to a high number.
Very lickely this causes the instability of the CKAN system.

On our test system (a release 00.07.01) we activated the production dump.
During the indexing (a process step required after changing the database in
the back)
the CKAN system halted. After emptying the activity table the indexing was
restarted and it finished properly. After that we have run rdf2ckan with
the latest zip you have provided.
At this moment the EEA and SANCO zips passed. Currently a Eurostat zip is
running.

In the svn you find a script to clean the appropriate tables. This is a
quick-win step towards having back a stable system.

https://webgate.ec.europa.eu/publications/svn/ODP/trunk/ckan/management_scripts/ckan_clear_activity_table.sh

Copy this script to $USERHOME/ckan_management and complete the necessary
parameters. These are the same as for a CKAN backup.

·         TARGETBACKUPDIR

·         CKAN_PWD

·         POSTGRES_HOST
The script will first backup the table and afterwards empty the tables.

The execution on our site costs +- 45 minutes and resulted in a file of 7.6
GB.

After the script execution, you have to trigger for this time the
reindexing of the database.
In $USERHOME/ckan_management

     ./ckan_reindex.sh

(It  costs +- 1hour on our site)

We keep you informed about the outcome of the Eurostat test.

ps. I inserted the description of the new script also in the OM on section
7.10.

kind regards,

Bert


-- 
Bert Van Nuffelen

Semantic Technologies Software Architect at TenForce
www.tenforce.be

Bert.Van.Nuffelen at tenforce.com
Office: +32 (0)16 31 48 60
Mobile:+32 479 06 24 26
skype: bert.van.nuffelen
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <https://lists.okfn.org/mailman/private/ecodp-dev/attachments/20130307/cea71212/attachment.html>