Comment 7 for bug 1338660

Revision history for this message
OpenStack Infra (hudson-openstack) wrote : Fix merged to trove (master)

Reviewed: https://review.openstack.org/106127
Committed: https://git.openstack.org/cgit/openstack/trove/commit/?id=e527ad8f2ef45be7afd97e9af7c01eb105a6197e
Submitter: Jenkins
Branch: master

commit e527ad8f2ef45be7afd97e9af7c01eb105a6197e
Author: Amrith Kumar <email address hidden>
Date: Thu Jul 10 13:10:27 2014 -0400

    Stop cassandra during configuration and volume migration

    The cassandra service is left running after installation.
    The next operation(s) are to rewrite the configuration, and
    migrate the data onto a volume (if provided). It is not a
    good thing for the database to be running during the latter,
    and a restart is required after the former.

    Failing to stop the database during the data migration can
    result in an rsync error 24 because files that were present
    when rsync made its list may be missing by the time rsync
    gets around to trying to sync that file. This can lead rsync
    to not make a 'true' copy of the source on the destination.

    The only way around this is to stop the database, something
    that is already done by both MongoDB and MySQL in their
    guest agents.

    Change-Id: I74ecc5c12d43eccc6295488e344bc0e75fe8116d
    Closes-bug: #1338660