Mailtrust down.

Portal Home > Knowledgebase > Articles Database > Mailtrust down.

Posted by sliqua, 10-07-2008, 08:44 PM
Appears that Mailtrust's IMAP, Webmail, and main website are all down. Not sure what's going on over there at Rackspace tonight. [alex@falcon ~]$ curl -I http://webmail.mailtrust.com HTTP/1.1 500 Internal Server Error Content-Type: text/html Content-Length: 369 Date: Wed, 08 Oct 2008 00:44:28 GMT Server: lighttpd/1.4.18
Posted by gRF, 10-07-2008, 09:16 PM
I was just about to report the same, but you already did it. I tried even calling their customer support (tel: 866.392.3336, option 1), but couldn't get live person to talk to. It's been about an hour since the email is down (POP, IMAP, webmail).
Posted by sliqua, 10-07-2008, 09:24 PM
I got a call back after leaving a voicemail - they are going to give me a call when an ETA is available. I will update this thread accordingly. At this point, according to my monitoring of their gear: secure.emailsrvr.com 56 minutes 40 seconds ago Secure IMAP Email, Secure POP Email, etc. webmail.mailtrust.com 1 hour 7 minutes ago Secure HTTP
Posted by tanfwc, 10-07-2008, 09:40 PM
At this point of testing, mx1 is online and mx2 is down. Their DNS services is responding as well, glad no email loss will do
Posted by sunpost, 10-07-2008, 09:56 PM
Noteworthy Service Interruption Posted on 10/07/2008 8:42 pm Dear Mailtrust Customer, Mailtrust would like to notify you of a situation that is affecting some of our customers on our Noteworthy environment. Currently, we are experiencing issues with our customers ability to log into their Noteworthy and Webmail.us accounts via webmail.mailtrust.com. Our engineers are working diligently to resolve this issue. We will update your support ticket as soon as we have more information to provide. Please be advised this is not affecting our Hosted Exchange customers. Mailtrust understands that uptime is the most important aspect of your e-mail service. We truly appreciate your patience and understanding in this matter. Regards, Mailtrust Support Update Posted on 10/07/2008 9:29 pm This issue will also affect some of our Noteworthy subscribers who connect to their accounts via POP3 and IMAP4 on their email clients, such as Outlook and Thunderbird. Currently we are unable to provide an estimated time when this service interruption will be resolved.
Posted by coolspot, 10-07-2008, 10:03 PM
Thanks for the info everyone... I was wondering why my webmail was unavailable. Doesn't mailtrust promote 100% uptime? 100% Uptime Guarantee Email is the most important business communication tool, and Mailtrust™ knows that when email is down, business is down. Our industry-leading 100% uptime guarantee is a reflection of both our understanding of email's central importance to businesses and the high availability email system engineered by our experts. It looks like the site is back up now. Last edited by coolspot; 10-07-2008 at 10:07 PM.
Posted by sunpost, 10-07-2008, 10:07 PM
my POP3 is now working.
Posted by tanfwc, 10-07-2008, 10:10 PM
IMAP working with no email loss
Posted by sliqua, 10-07-2008, 10:48 PM
Began at: 20:16:00 EDT, total duration 1 hour 51 minutes... Ouch. I wonder if any of us will get an SLA credit After doing quite a bit of research into this- it appears that only their IAD location- which hosts part of their IMAP/POP, SMTP, and Webmail- was down. They do geographic load balancing for their DNS records, and the two other IPs they gave off in San Antonio seemed to be working fine throughout this entire outage (for IMAP over SSL at least). It would've been nice to see them actually fail-over between datacenters in the event an entire site (IAD- Dulles, VA) becomes unavailable. Just my two cents.
Posted by tanfwc, 10-07-2008, 11:30 PM
Why not ask them whether they provide SLA ? Guess their failover DNS did not work this time, maybe you can report to them as well so they can improve to the next level
Posted by gRF, 10-08-2008, 02:22 AM
The explanation I got was that there was problem with the firewall. So, failover DNS won't help much in that case, am I right?
Posted by sliqua, 10-08-2008, 06:56 AM
Failover DNS would work in this case for sure. Their DNS servers are announcing three different IP addresses-- only one of which was down (at their IAD location). Thus, 2/3 of their customers around the world experienced no issues last night. (Besides webmail and mailtrust.com not loading) If Mailtrust had removed the bad IP from their load balancing pool in DNS, so the other two locations handled all incoming requests, no one would've experienced an outage. If their system is truly clustered, it seems to me that they would've taken this action after a few minutes of downtime. But to let it stay down for almost two hours in unacceptable. This makes me wonder if their other sites have enough capacity to handle their user load in the event a site goes completely down (as was the case last night). I can't even begin to imagine this happening in the middle of a business day, it would've been catastrophic. The performance last night was definitely not what I would expect from Rackspace. NOTE: I've been in contact with a manager at Mailtrust, who has assured me that I will get a technical explanation about what happened. In addition, he will let me know what steps are being put into action to make sure this won't happen again.