Insofar as anyone has admitted, yes. Since the problem appeared, I did update the NIC driver, since in my reading on the problem mentioned that it may be a communications problem caused by a bad NIC or driver, though I had not noticed a problem in some multi-gigabyte downloads of Microsoft DVD ISO files recently. I also had tested using a few long ping periods to various IP addresses, and received no timeouts.
\\Steve// From: Micheal Espinola Jr [mailto:michealespin...@gmail.com] Sent: Tuesday, April 13, 2010 7:30 PM To: MS-Exchange Admin Issues Subject: Re: 4.4.2 Connection Dropped Is that the only [recent] thing done prior to the issue? -- ME2 On Tue, Apr 13, 2010 at 4:26 PM, Steve Szabo <steve...@gmail.com> wrote: We just use the Exchange anti-spam filter and Spamhaus's Zen list. We have Vipre Enterprise 4.0, and the Exchange directories are excluded from scanning by it. The upgrade to 4.0 was done about 2 weeks ago-and did not appear to do any harm to Exchange at the time. \\Steve// From: Micheal Espinola Jr [mailto:michealespin...@gmail.com] Sent: Tuesday, April 13, 2010 6:51 PM To: MS-Exchange Admin Issues Subject: Re: 4.4.2 Connection Dropped Ouch, ouch and ouch. Forgive me if I missed it, but what kind of anti-spam and/or anti-virus do you have installed on/along-side-of Exchange? -- ME2 On Tue, Apr 13, 2010 at 12:47 PM, Steve Szabo <steve...@gmail.com> wrote: Micheal, I ended up restarting the server when the attempt to restart the Transport service locked the server. I am coming to the conclusion that there is a problem with the server itself or the Exchange install that is the root cause of this problem. This is not the first time the server has locked when stopping and starting Exchange services since this problem manifested itself. Also, on server restarts, several Exchange services may not start, the most frequent are the Anti-Spam Update, File Distribution, and POP3 services, followed closely be the Active Directory Topology service. We are now discussing whether we should put dome money into building another server and migrating Exchange to it to see if that will resolve the problem. About 5 minutes ago, I dropped an e-mail into the queue that will be delivered, or, at least, Exchange has been able to deliver, to a domain with which we have had no delivery problems with yet. But, it should have gone out nearly immediately, yet has not. I can't say if this is consistent with previous behavior--the long stay in the queue, since I've never needed to pay much attention to the queue before. \\Steve// On Tue, Apr 13, 2010 at 1:40 PM, Micheal Espinola Jr <michealespin...@gmail.com> wrote: You could restart services instead of the server. -- ME2 On Tue, Apr 13, 2010 at 7:32 AM, Steve Szabo <steve...@gmail.com> wrote: We have a very busy day if we have 50 e-mails going out, so I do not think that would be an issue. As for a username/password, I suspect that you only get one if you are using them for e-mail service as we do not have one, and never received one. I did not receive one with the business FiOS account I have at home either. (And, I do have all the documentation for both accounts.) As for a further follow-up for those following this, it appears that the PTR record ha been restored, I checked several DNS servers and all returned it, but mail is still hanging in the queues with the same "dropped connection" error message. I would not think I had to, but should I restart the server? \\Steve// On Mon, Apr 12, 2010 at 8:56 PM, <greg.swe...@actsconsulting.net> wrote: I should also mention that we tend to see them throttle the number of emails you can send daily. They say they don't do it, but when all the mail queues up and overnight it all goes out, and then hey look at that..5pm they are queuing again at noon after a big email and then they go out overnight.. Greg From: greg.swe...@actsconsulting.net [mailto:greg.swe...@actsconsulting.net] Sent: Monday, April 12, 2010 7:43 PM To: MS-Exchange Admin Issues Subject: RE: 4.4.2 Connection Dropped You can configure Exchange to do an authenticated SMTP connection through your Smarthost using your FIOS username and pw, using basic authentication. We have done it and it works, but I would not make it permanent. If you are like most of our clients, they never hang onto their Verizon.net info and they have to go calling tech support to get the pw reset. Greg From: Campbell, Rob [mailto:rob_campb...@centraltechnology.net] Sent: Monday, April 12, 2010 2:12 PM To: MS-Exchange Admin Issues Subject: RE: 4.4.2 Connection Dropped You might ask them if they'll let you use one of their servers as a smarthost until it gets sorted out. From: Steve Szabo [mailto:steve...@gmail.com] Sent: Monday, April 12, 2010 1:09 PM To: MS-Exchange Admin Issues Subject: Re: 4.4.2 Connection Dropped When I was on the phone with them, that issue really did not come up. I think the problem is more that the rDNS record disappeared. \\Steve// On Mon, Apr 12, 2010 at 1:52 PM, Campbell, Rob <rob_campb...@centraltechnology.net> wrote: Any chance Verizon mistakenly published that Class C you're in as being a dynamic / DHCP range? From: Steve Szabo [mailto:steve...@gmail.com] Sent: Monday, April 12, 2010 12:47 PM To: MS-Exchange Admin Issues Subject: 4.4.2 Connection Dropped We are having a serious problem sending mail out from our Exchange server. We have Exchange 2007, fully patched, running on Server 2003 R2, again, fully patched. The problem started Thursday when it was noted that some e-mail was not going out, but sitting in the queues. The message give was ""421 4.4.2 Connection dropped.' Attempted failover to alternate host, but that did not succeed. Either there are no alternate hosts, or delivery failed to all alternate hosts." A check showed that we were listed on an RBL, NOMOREFUNN which apparently blocks whole ranges of IP addresses--checking showed that the whole class C network we are part of was blocked by them. I applied for a removal of the block, but it never happened, and now that RBL is timing out when I try to check it using MXToolbox, but I can reach it manually, and, despite the requests, it still shows us as being on the list. Their complaint is "washdc.fios.verizon.net. Dynamic IP or generic rDNS". It does appear that there is a generic rDNS right now, which there did not used to be. I have talked to Verizon about getting this fixed, but I do not think this is the source of the problem, since whether the mail goes through is a hit or miss affair. I do not believe that NOMOREFUNN is a widely used RBL, and do not think it is a source of the problem, however, it may be that the rDNS is. Am I going the right way with this, or am I barking up the wrong tree and fixing the rDNS will not fix the problems we are having?