HCoop / hcoop / debian / openafs.git / blame
| Commit | Line | Data |
|---|---|---|
| 805e021f CE |
1 | Copyright 2000, International Business Machines Corporation and others. |
| 2 | All Rights Reserved. | |
| 3 | ||
| 4 | This software has been released under the terms of the IBM Public | |
| 5 | License. For details, see the LICENSE file in the top-level source | |
| 6 | directory or online at http://www.openafs.org/dl/license10.html | |
| 7 | ||
| 8 | Suggested test case for the backup system | |
| 9 | ||
| 10 | 1) | |
| 11 | Set variabel sizes on a batch of tapes to be used. Read | |
| 12 | labels on them to verify the name and sizes. | |
| 13 | ||
| 14 | Dump a large enough volume set so that some volumes | |
| 15 | fragment across tapes and more than one tape is used. | |
| 16 | ||
| 17 | Do restore of volume dumped in the beginning, at the | |
| 18 | end and inbetween on the tape. | |
| 19 | ||
| 20 | Enumerate the tapes used to check the volumes on them | |
| 21 | and to determine info for previous step. | |
| 22 | ||
| 23 | Repeat the above with incrementals as well ie do above | |
| 24 | steps and repeat them doing incremental dumps of the same | |
| 25 | volume set. Then do a full restore on those volumes. | |
| 26 | ||
| 27 | 2) | |
| 28 | Do the mix suggested in 1) but introduce variable delays | |
| 29 | in changing tapes and aborting jobs in differing stages | |
| 30 | of dump and restore. | |
| 31 | ||
| 32 | Also try putting in the wrong tapes. | |
| 33 | ||
| 34 | 3) | |
| 35 | Do a restore after zapping a volume but keeping the vldb | |
| 36 | entry, but without the .ext flag. | |
| 37 | ||
| 38 | 4) To test the diskrestore command. Create some test volumes | |
| 39 | on the selected partition. Dump these volume on tape | |
| 40 | (which creates the db entry as a side effect). Zap these | |
| 41 | volumes using vos zap. issue the diskrestore command with .ext | |
| 42 | switch. | |
| 43 | ||
| 44 |