On Wed, 5 Dec 2018 05:21:30 +0000 Simon Slavin <[email protected]> wrote:
> On 5 Dec 2018, at 5:16am, Ryan Schmidt <[email protected]> wrote: > > > https://kb.vmware.com/s/article/1008542 > > > > "VMware ESX acknowledges a write or read to a guest operating > > system only after that write or read is acknowledged by the > > hardware controller to ESX. Applications running inside virtual > > machines on ESX are afforded the same crash consistency guarantees > > as applications running on physical machines or physical disk > > controllers." > > Interesting. That paragraph is a well-written piece of text > explaining the opposite of what I thought. Maybe things have changed > in the past decade. VMware may well be doing the best it can on unreliable hardware. I believe it's common knowledge that consumer-grade hard drives lie when acknowledging writes: the acknowlegement is sent when the data are received into the device's write buffer, not after being written to disk. It's good for benchmarks. No one benchmarks data corruptions. 'Twas ever thus: If you want a reliable database, use a reliable disk. --jkl _______________________________________________ sqlite-users mailing list [email protected] http://mailinglists.sqlite.org/cgi-bin/mailman/listinfo/sqlite-users
