That varies with your definition of RAID. With a decent hardware RAID controller that works properly, the RAID controller can have megabytes or gigabytes of data which has not been written to the physical spinning disk when power is abruptly turned off. And none of it will be lost.
Then you have a range of getting ever cheaper hardware RAID where varying amounts of data get lost, either from the controller or the embedded drive cache. Then you have "fake RAID", also known as software raid, where all data not physically written is lost. In other words you get what you pay for. > -----Original Message----- > From: sqlite-users-bounces at mailinglists.sqlite.org [mailto:sqlite-users- > bounces at mailinglists.sqlite.org] On Behalf Of Simon Slavin > Sent: Saturday, 30 January, 2016 13:00 > To: SQLite mailing list > Subject: Re: [sqlite] Bug: Successfully committed transaction rolled back > after power failure > > > On 30 Jan 2016, at 7:56pm, James K. Lowden <jklowden at schemamania.org> > wrote: > > > Given that the fsync has returned successfully, I don't know of any > > hardware that then will take 1000 ms to complete the write. That's the > > basis for my "subsecond interval" assumption. > > Writing to a RAID which has other write commands queued. > > Simon. > _______________________________________________ > sqlite-users mailing list > sqlite-users at mailinglists.sqlite.org > http://mailinglists.sqlite.org/cgi-bin/mailman/listinfo/sqlite-users
