  battleop
join:2005-09-28 00000 | 10% of that is from Barracuda boxes.
Nothing annoys me more than to get bounce back messages from Barracuda boxes. They read the spam and bounce messages back to the forged reply to address. |
|
  Matt Take me down to the paradise city Premium join:2003-07-20 Jamestown, NC
·North State Commun..
| said by battleop :Nothing annoys me more than to get bounce back messages from Barracuda boxes. They read the spam and bounce messages back to the forged reply to address. Don't blame Barracuda, blame the admin who configure them. That option has to be ENABLED by the admin. -- Pretty Fly for a White Guy |
|
 ElJay
join:2004-03-17 | They should force the admin to remove one of his eyes to enable that option. I had a huge backscatter problem last summer and those poorly configured Barracuda boxes were a good slice of it. |
|
 six9
join:2001-12-03 Atlanta, GA
·Comcast
| Yeah, any Barracuda admin that would enable that "feature" is just contributing to the problem.
I put one in at my last job and it was blocking like 95% of the email as spam or quarantine. I had that option turned off, no need spamming back to addresses that don't exist. And most of them did not exist as they were from residential zombie networks. |
|
 B Premium,MVM join:2000-10-28
| The buyer of an antispam appliance has already admitted they can't handle spam on their own and want to simply throw money at the problem and make it go away. The Barracuda "admin" shouldn't even have an option to screw up and end up generating more spam byproduct.
Therefore it's irresponsible of Barracuda to continue to offer a feature that is never a good idea in today's (or the foreseeable future's) spam climate. If and when it becomes a reasonable feature they can activate it automatically.
Not only do losers purchase these stupid appliances, but apparently the companies who make them are about as wise as the likes of Symantec vis a vis antivirus "research".
-- B -- In a realm outside causality and function |
|
  Matt Take me down to the paradise city Premium join:2003-07-20 Jamestown, NC
·North State Commun..
| said by B :The buyer of an antispam appliance has already admitted they can't handle spam on their own and want to simply throw money at the problem and make it go away. The Barracuda "admin" shouldn't even have an option to screw up and end up generating more spam byproduct. Therefore it's irresponsible of Barracuda to continue to offer a feature that is never a good idea in today's (or the foreseeable future's) spam climate. If and when it becomes a reasonable feature they can activate it automatically. Not only do losers purchase these stupid appliances, but apparently the companies who make them are about as wise as the likes of Symantec vis a vis antivirus "research". -- B And how would you suggest these "losers" handle spam? -- Pretty Fly for a White Guy |
|
 B Premium,MVM join:2000-10-28
2 edits | Seriously? Any number of free and/or open source solutions that are at least equal in performance.
Start with everybody's old reliable, SpamAssassin. Enable the Bayesian filtering.
If you like something a little different, try DSPAM.
Or try what I use, ASSP -- it's a completely multiplatform (Perl) proxy and rejects spam at the SMTP level, BEFORE it reaches your mail store. It doesn't require a dedicated server, and it includes antimalware (ClamAV) scanning.
There are plenty of other software solutions (paid and free) for an "admin" to consider. Again, buying a box (throwing money at the problem) is perfectly viable too, it's just a bit lazy and not necessarily as effective. It certainly makes sense for organizations who decide to commit employee resources to other things. That said, I have an older version of ASSP running that I basically NEVER have to tweak; and yet it remains very very effective.
-- B -- In a realm outside causality and function |
|
  Matt Take me down to the paradise city Premium join:2003-07-20 Jamestown, NC
·North State Commun..
| said by B :Seriously? Any number of free and/or open source solutions that are at least equal in performance. Start with everybody's old reliable, SpamAssassin. Enable the Bayesian filtering. If you like something a little different, try DSPAM. Or try what I use, ASSP -- it's a completely multiplatform (Perl) proxy and rejects spam at the SMTP level, BEFORE it reaches your mail store. It doesn't require a dedicated server, and it includes antimalware (ClamAV) scanning. There are plenty of other software solutions (paid and free) for an "admin" to consider. Again, buying a box (throwing money at the problem) is perfectly viable too, it's just a bit lazy and not necessarily as effective. It certainly makes sense for organizations who decide to commit employee resources to other things. That said, I have an older version of ASSP running that I basically NEVER have to tweak; and yet it remains very very effective. -- B That's what you pay for. We have an OpenSource spam filter as well and while it is effective, it's no where near as effective or as user friendly as the Barracuda devices. That's really what you get with the spam filtering network appliances which can be a godsend for smaller shops like ours. -- Pretty Fly for a White Guy |
|
 B Premium,MVM join:2000-10-28
| That's good. They fit your needs and you can afford them.
If you (or your employers) chose to spend time for a real "bake off" I'd be quite curious as to how and more importantly why the Barracudas are so much more effective for you than your open source filter is.
I've never seen any statistics that show the commercial antispam products are any better than the free ones; on the other hand I haven't really looked much. 
The DSPAM author has written a bit about the fallacy of relying on commercial spam filtering. »www.zdziarski.com/papers/justifying.html is one such piece but there are shorter ones.
I guess my own biggest worry about a black box appliance (other than trust, and hardware failure) is that it may become useless if you stop paying for updates (?), whereas my good Bayesian filter should continue to perform nearly perfectly, nearly untended.
-- B -- In a realm outside causality and function |
|