Matt Larsen - Lists wrote: > Our mail server locked up on Wednesday afternoon with some hard drive > weirdness. >snip snip > except for > the 26 hours that the mail server was not accepting email. > > However, our Barracuda box was accepting the messages and storing them. > After a phone call to Barracuda Networks, their tech support logged into > the box and it is spending the next six hours resending all 30,000+ > messages. Oh yeah! > > For anyone who is on the fence, jump over to the Barracuda side. My > Barracuda box has been a genuine lifesaver, beyond just stopping spam.
I'm really glad this worked out for you. However, this particular incident could have been avoided by using backup MX boxes. Further, 26 hours isn't a long outage. It's a big inconvient outage, and no doubt would upset folks pretty badly. But no email server worth it's salt should dump it's spool and put out an undeliverable bounce in that amount of time. I am strongly of the opinion that nearly all of the woes we currently associate with email stem from having poorly administered email servers and email systems. (No, I am not implying this is the case with you, quite the opposite, you obviously care about your service, to take the time, money and effort to use such a service). Adding yet still another blind layer of bureaucratic 'not my faults' to the chain does little in the long run other than just add another patch (and expensive one) to a very leaky infrastructure. I'm glad you are happy with your solution. But personally, I see this kinda fix as much more part of the problem "Oh? No, we use some *-insert outsourced plausible deniability clause here- service, so it isn't our problem" rather than part of a solution. Note *= usual suspect, postini, symantic email gateway, -- WISPA Wireless List: firstname.lastname@example.org Subscribe/Unsubscribe: http://lists.wispa.org/mailman/listinfo/wireless Archives: http://lists.wispa.org/pipermail/wireless/