Attachment 611808 Details for Bug 840478 – Restore Procedure for System Engine

Restore Procedure for System Engine

Restore.xml (text/xml), 4.67 KB, created by Dan Macpherson on 2012-09-11 15:00:05 UTC

(hide)

Description:

Filename:

MIME Type:

Creator: Dan Macpherson

Created: 2012-09-11 15:00:05 UTC

Size: 4.67 KB

patch

obsolete

><section>
>	<title>System Engine Restore Procedure</title>
>	<procedure>
>		<title>System Engine Restore Procedure</title>
>		<important>
>			<para>
>				The following process describes a full System Engine restore. This process deletes all data from a current System Engine instance. Ensure you are restoring to the correct instance. All commands are executed as <literal>root</literal> in the directory with archives created during the backup process.
>			</para>
>		</important>
>		<step>
>			<para>
>				Prepare the System Engine host for restore. This process restores the backup on the same server that generated the backup. If the original system is unavailable, provision the same configuration with the same settings (hostname, IP address, etc) and run <command>katello&ndash;configure</command> with the same options using the file from the configuration backup:
>			</para>
>			<screen># katello&ndash;configure &ndash;&ndash;answer&ndash;file=/etc/katello/katello&ndash;configure.conf</screen>
>		</step>
>		<step>
>			<para>
>				Change to the backup directory.
>			</para>
>			<screen># cd /backup</screen>
>		</step>
>		<step>
>			<para>
>				Check the names for System Engine PostgreSQL databases. Default database names are <literal>katelloschema</literal> and <literal>candlepin</literal>. Find all default values in the katello&ndash;configure manual page, or in the <filename>/usr/share/katello/install/default&ndash;answer&ndash;file</filename>. If the System Engine instance is not using the default names, check the <filename>katello&ndash;configure.conf</filename>, which stores all options provided to katello&ndash;configure command, and look for <literal>db_name</literal>. 
>				<screen># grep db_name /etc/katello/katello&ndash;configure.conf</screen>
>			</para>
>		</step>
>		<step>
>			<para>
>				Stop all services prior to restoring the databases:
>			</para>
>			<screen># service katello stop
># service katello&ndash;jobs stop
># service pulp&ndash;server stop
># service tomcat6 stop
># service postgresql stop
># service elasticsearch stop</screen>
>		</step>
>		<step>
>			<para>
>				Restore the system files. Double-check that the files extract on the correct host. Restore all system files with following commands executed from the backup directory:
>			</para>
>			<screen># tar &ndash;&ndash;selinux &ndash;xzvf config_files.tar.gz &ndash;C /
># tar &ndash;&ndash;selinux &ndash;xzvf elastic_data.tar.gz &ndash;C /
># tar &ndash;&ndash;selinux &ndash;xvf pulp_data.tar &ndash;C /</screen>
>		</step>
>		<step>
>			<para>
>				Drop the existing System Engine PostgreSQL databases if any exist:
>			</para>
>			<screen># service postgresql start
># su postgres &ndash;c "dropdb katelloschema"
># su postgres &ndash;c "dropdb candlepin"</screen>
>			<note>
>				<para>
>					There is a possibility of the following error appearing during database drop:
>				</para>
>				<screen>database xxx is being accessed by other users</screen>
>				<para>
>					Ensure all processes are stopped.
>				</para>
>			</note>
>		</step>
>		<step>
>			<para>
>				Restore System Engine PostgreSQL databases with the following commands:
>			</para>
>			<screen># su postgres &ndash;c "pg_restore &ndash;C &ndash;d postgres /backup/katello.dump"
># su postgres &ndash;c "pg_restore &ndash;C &ndash;d postgres /backup/candlepin.dump"</screen>
>			<note>
>				<para>
>					For more info about PostgreSQL backups consult <command>pg_dump</command> manual page or visit <ulink url="http://www.postgresql.org/docs/8.4/static/backup.html">http://www.postgresql.org/docs/8.4/static/backup.html</ulink>.
>				</para>
>			</note>
>		</step>
>		<step>
>			<para>
>				Ensure MongoDB is running and delete the old data:
>			</para>
>			<screen># service mongod start
># echo 'db.dropDatabase();' | mongo pulp_database</screen>
>		</step>
>		<step>
>			<para>
>				Execute the following command while provide the <literal>&ndash;&ndash;username</literal> and <literal>&ndash;&ndash;password</literal> when in the directory with backup archives:
>			</para>
>			<screen># mongorestore &ndash;&ndash;host localhost mongo_dump/pulp_database/</screen>
>			<note>
>				<para>
>					For more information consult mongorestore man page or visit <ulink url="http://www.mongodb.org/display/DOCS/Backups"/>.
>				</para>
>			</note>
>		</step>
>		<step>
>			<para>
>				Restart all System Engine processes:
>			</para>
>			<screen># service postgresql start
># service tomcat6 start
># service pulp&ndash;server start
># service elasticsearch start
># service katello&ndash;jobs start
># service katello start</screen>
>		</step>
>		<step>
>			<para>
>				Check log files for errors, check <filename>audit.log</filename> for denials and attemp to ping the System Engine instance:
>			</para>
>			<screen># katello &ndash;u admin -p admin ping</screen>
>		</step>
>	</procedure>
></section>

Actions: View

Attachments on bug 840478: 611806 | 611808