2009-01-06
See /root/messages.sdf-failing. Or possibly just a raid glitch.
2009-01-18
Test
- Connect drives
- Assemble array
- dd if=/dev/md3 bs=4k of=/dev/null
Reference
- Drives direct SATA connected
- Drives powered by enclosure PSU
Configuration
| A | B | C | D | E |
|---|---|---|---|---|
| 6 | 5 | 1 |
| F | 2 |
|---|---|
| G | 3 |
| H | 4 |
Test backplane
Test Slot D
| A | B | C | D | E |
|---|---|---|---|---|
| 6 | 5 | 4 | 1 |
| F | 2 |
|---|---|
| G | 3 |
| H |
- Not Stable
Test Slot B
| A | B | C | D | E |
|---|---|---|---|---|
| 6 | 4 | 5 | 1 |
| F | 2 |
|---|---|
| G | 3 |
| H |
- Stable
Test Slot B & D
| A | B | C | D | E |
|---|---|---|---|---|
| 6 | 4 | 5 | 3 | 1 |
| F | 2 |
|---|---|
| G | |
| H |
- Not Stable
Swap disks 3 and 4
| A | B | C | D | E |
|---|---|---|---|---|
| 6 | 3 | 5 | 4 | 1 |
| F | 2 |
|---|---|
| G | |
| H |
- Not Stable
Swap power supply cables
- Removed and cleaned 5-bay IcyDock
- Swap power supply cables
- Old: Cables 1->A, 2->BC, 3->DE.
- New: 1->BC, 2->DE, 3->A
- Stable
| A | B | C | D | E |
|---|---|---|---|---|
| 6 | 4 | 5 | 1 |
| F | 2 |
|---|---|
| G | 3 |
| H |
Slot D again
| A | B | C | D | E |
|---|---|---|---|---|
| 6 | 5 | 4 | 1 |
| F | 2 |
|---|---|
| G | 3 |
| H |
- Stable
Slots B & D again
| A | B | C | D | E |
|---|---|---|---|---|
| 6 | 3 | 5 | 4 | 1 |
| F | 2 |
|---|---|
| G | |
| H |
- Stable
Connect PMP bridges
- Power up, leave disks direct-connected
- Connect bottom bridge
| A | B | C | D | E |
|---|---|---|---|---|
| 6 | 3 | 5 | 4 | 1 |
| F | 2 |
|---|---|
| G | |
| H |
-
Stable
- Connect upper bridge
- Same configuration
- Stable
PMP Connect Slot F
- Connect F to bottom bridge
- Connect bottom bridge through eSATA
| A | B | C | D | E |
|---|---|---|---|---|
| 6 | 3 | 5 | 4 | 1 |
| F | 2 |
|---|---|
| G | |
| H |
Reconnect Everything
- Reconnected all drives through PMP bridges
- PMP bridges to SiI3132
- Reboot seemed normal
| A | B | C | D | E |
|---|---|---|---|---|
| 6 | 5 | 1 |
| F | 2 |
|---|---|
| G | 3 |
| H | 4 |
- All seems normal
2009-06-29
- Occurred at 20h47
- double disk failure
- sdc? failed first
- alarm in enclosure, ejected failed disk (should have removed with mdadm?)
- made array read-only
- system lock-up shortly thereafter?
- See HasturRaidRecovery