On 18/02/14 15:49, Mark Martinec wrote: > One server in each continent might be acceptable, but hasn't > been tried.
Yeah, in fact I can separate into Europe and US (responsible for different domains), so two Redis makes more sense. > No corruption can happen due to network problems. Cases where some > but not all tokens are learned, or tokens learned but 'seen' entry > not added are non-problematic if it doesn't happen too often. > Token updates usually fit within a single IP packet, so in most > cases either all of the transaction gets committed or none, > even in case of network problems. My definition of "corruption" is when it causes the app to be unhappy - not necessarily that the data is corrupt :-) Obviously a TCP packet either arrives and passes checksum or it's thrown away. Your last sentence answers my question - network outages shouldn't cause the Bayes data to become useless to SA - good > > A full network breakdown (or server down) would cause SpamAssassin > to log warnings for each mail message, but will move on anyway, > just without Bayes checks. Yep - that's fine. I think it'll be worth a shot :-) Thanks! -- Cheers Jason Haar Information Security Manager, Trimble Navigation Ltd. Phone: +1 408 481 8171 PGP Fingerprint: 7A2E 0407 C9A6 CAF6 2B9F 8422 C063 5EBB FE1D 66D1
