note the smartctl report is
"failure prediction"
on two HDs not failed.
I used the scterc option (see Phil's recommend so the smart is cut-off. I 'll check his full ist of options.
smartctl -iA -l scterc /dev/sdXn
Tom
On 09/01/2017 07:24 PM, Andreas Klauer wrote:
On Fri, Sep 01, 2017 at 03:15:41PM -0500, Thomas C. Bishop wrote:
I messed up my raid5 array .
That's an understatement...
I know a two of HDs are "failure
prediction" and one is out..
RAID 5 with three failed drives, chances of survival are very low.
You should never let things get this far. Timeouts? Doesn't matter!
You either have no disk monitoring at all or never acted on it.
ddrescue the broken drives to new ones first.
Then always use overlays for recovery experiments.
https://raid.wiki.kernel.org/index.php/Recovering_a_failed_software_RAID#Making_the_harddisks_read-only_using_an_overlay_file
Experiments for example could be:
*) --assemble --force
*) --assemble --update=force-no-bbl
*) --create --metadata=1.0 --chunk=128 with one 'missing' drive
Again, use overlays for everything.
Bad Block Log : 512 entries available at offset -8 sectors - bad
blocks present.
You have bbl entries on more than one drive, use --examine-badblocks
to see if they are identical. You have to clear those or md will
either not work at all or always give read errors even after replacing
the drives. Bad block list issues were previously discussed on the list,
you might find it when searching for "no-bbl".
=== START OF READ SMART DATA SECTION ===
SMART Health Status: OK
Never trust this unconditionally. It's a false friend.
Always look at the detailed output with reallocated etc. sectors.
Run selftests regularly, detect disk errors early, replace drives immediately.
Good luck
Andreas Klauer
--
***********************************
Thomas C. Bishop
Hazel Stewart Garner Associate Professor
Chemistry & Physics
Tel: 318-257-5209
Fax: 318-257-3823
www.latech.edu/~bishop
***********************************
--
To unsubscribe from this list: send the line "unsubscribe linux-raid" in
the body of a message to majordomo@xxxxxxxxxxxxxxx
More majordomo info at http://vger.kernel.org/majordomo-info.html
