View Single Post
Old 22nd August 2011
jggimi's Avatar
jggimi jggimi is offline
More noise than signal
 
Join Date: May 2008
Location: USA
Posts: 7,977
Default

1) Review your /var/log/messages* files -- this is your kernel message history -- and search for error messages with timestamps that match the time of this most recent crash or hang.

2) Confirm you are able to write to, then read from, every sector of the drive, and that your two tables of known bad blocks have returned to zero. As I stated multiple times in this thread, a bad sector will remain unreadable until it is replaced by a spare sector, which only occurs when a known bad sector is written. You do not need to use OpenBSD for this; you can use Ubuntu, where you seem to have more success. I'll continue to recommend badblocks.

3) An interesting test is the SMART extended offline self-test. It had been run on your drive at least once, because you have "offline uncorrectable" sectors. In this test, the drive electronics reads every sector on the drive. The value of this test are multiple: a) it finds bad sectors that are not in use by your OS, mitigating the risk of future data loss from writing to undiscovered bad sectors; b) it puts bad sectors on a list that are automatically replaced with spares when written; c) it is a test by the electronics on the drive, eliminating bus cables, connectors, or software drivers producing false positives; d) it can be run while using the drive with your OS, though drive performance will be impacted, and the elapsed time for the test will be extended. You can request one with the -t long option of smartctl; you then use smartctl with the -a option to compare results after the test has completed.
Reply With Quote