Backups: Difference between revisions
Jump to navigation
Jump to search
(Created page with "Backups are run as follows: * Weekly starting Friday night: /nfs/home/ * Quarterly (or on demand) on the first day of the quarter: /nfs/db, /nfs/store * Never: /nfs/work *...") |
No edit summary |
||
| Line 7: | Line 7: | ||
We have four save packs that we rotate for /nfs/home. We set aside one of these savepacks quarterly, on the first backup of the quarter, as an offsite backup. Thus at any time there is a weekly backup made starting Friday night for the last four weeks, and a quarterly backup for the last four quarters. | We have four save packs that we rotate for /nfs/home. We set aside one of these savepacks quarterly, on the first backup of the quarter, as an offsite backup. Thus at any time there is a weekly backup made starting Friday night for the last four weeks, and a quarterly backup for the last four quarters. | ||
At time of writing, this policy is only implemented on [[Cluster 2]]. We are currently implementing it on [[Cluster 1]] and [[Cluster 0]]. We will update this page when those two systems adhere to the stated policy. In the meantime, ask us if you have any doubts. | |||
[[Category:Internal]] | [[Category:Internal]] | ||
[[Category:Sysadmin]] | [[Category:Sysadmin]] | ||
Revision as of 18:44, 18 March 2014
Backups are run as follows:
- Weekly starting Friday night: /nfs/home/
- Quarterly (or on demand) on the first day of the quarter: /nfs/db, /nfs/store
- Never: /nfs/work
- No other files are backed up routinely. If you have files elsewhere, we recommend the use of Crashplan to copy files to /nfs/home/<userid> whence they will be backed up.
We have four save packs that we rotate for /nfs/home. We set aside one of these savepacks quarterly, on the first backup of the quarter, as an offsite backup. Thus at any time there is a weekly backup made starting Friday night for the last four weeks, and a quarterly backup for the last four quarters.
At time of writing, this policy is only implemented on Cluster 2. We are currently implementing it on Cluster 1 and Cluster 0. We will update this page when those two systems adhere to the stated policy. In the meantime, ask us if you have any doubts.