Heine Ferreira wrote:
> Are there any best practices for avoiding database corruption?
First and foremost, do not turn off fsync or full_page_writes in your
configuration. After that the most common causes for database
corruption I've seen are bad RAM (ECC RAM is a requirement, not an
option for a serious database server in my book), failing hard drives
(use RAID and SMART monitoring, and have a sane hardware replacement
policy), or buggy device drivers (pay attention to releases which fix
bugs or security vulnerabilities which could affect you). It's
getting rare to see this now, but it used to be common for some
drives to lie to you about having written data when it was actually
just in cache -- a good BBU RAID controller will go out of its way to
keep the drives honest.
-Kevin
