Please find bellow the results of yesterday's grooming On 10/03/2016 06:30 AM, Tristan Cacqueray wrote: > Hello, > > Next grooming we should focus on backup and metrics. > > For backup, I'd like to propose the following changes: > > * Convert per component backup scripts into ansible backup action > Like we did for update and get_logs... > * Make each component backup its data to a local bup store instead > of using remote managesf bup store > * Make each component restore task use the local bup store too > * Add a new backup playbook that would call the backup task of each > roles, and then rsync all the bup stores back to the install server > * Add a new restore playbook that would rsync the bup store to each > instance and call the restore task of each roles > > * Modify the managesf backup controller to do: > * Call the backup playbook > * Create an archive of all bup stores > * The restore action would: > * Extract the archive on the install-server > * Call the restore action of all services >
First story to refactor using ansible: http://softwarefactory-project.io/redmine/issues/1567 Second story to investigate improvements: http://softwarefactory-project.io/redmine/issues/1568 > > And for metrics, we have a couple of options: > > * Investigate proper gnocchi integration and/or > * Revert back to graphite > Investigate metrics solutions: http://softwarefactory-project.io/redmine/issues/1569 Regards, -Tristan
signature.asc
Description: OpenPGP digital signature
_______________________________________________ Softwarefactory-dev mailing list [email protected] https://www.redhat.com/mailman/listinfo/softwarefactory-dev
