On Thu, 13 Jan 2011, Timo Sirainen wrote:
How do they fail? Supposedly once a cell has reached its erase-limit it should become read-only. Maybe the failures had nothing to do with wearing?
Hi Timo. I start seeing I/O errors on read or write. I must admit I don't have definitive proof that it is a wear level problem but there isn't any other obvious cause either. ie, no known heat or other problems have occured on the systems.
It's left me a bit gun-shy of deploying SSD more widely. I would certainly love to be able to trust it as an alternative to disk.
Yes cells certainly should go read-only when they run out of write-cycles, which would be a far less painful way for them to fail :)
One of the systems to fail was a firewall running off SSD. A Linux based firewall can lose entire filesystems and keep running[1] so I first noticed the problem when the backups started to fail.
[1] Although you can't change the firewall ruleset without userspace tools.
Cheers,
Rpb
-- Email: robert@timetraveller.org Linux counter ID #16440 IRC: Solver (OFTC & Freenode) Web: http://www.practicalsysadmin.com Contributing member of Software in the Public Interest (http://spi-inc.org/) Open Source: The revolution that silently changed the world