We greatly appreciate the feedback from our Free Monitoring Tool users regarding the issue of Server Alerts not being delivered. We are currently working on it and it should be fixed shortly.
This is just one of the features included in the Free Monitoring Tool and we encourage our customers to continue using the product. Please continue to let us know your feedback or any improvements you would like to see.
Stay tuned to our blog and Forums for more updates.
UPDATE 5/20/2009 @ 8:59pm: Server Alerts are now functioning again, we apologize to everyone for this interruption in service. We were working on upgrade the database and code and the most recent code roll out did cause a problem. We have resolved this problem at this time.
UPDATE 5/22/2009: We are aware that the Free Monitoring Tool is issuing some false positive alerts at this time. We are working on correcting this as quickly as possible.
UPDATE 5/22/2009: The false positive problem has been brought under control. There were two issues, both related to a software update. The first was that previously, we were only alerting if your email server failed to accept our TCP connection on port 25. Sending us a 400 SMTP error would not generate an alert. The code change caused servers that were either issuing 40x errors OR terminating our SMTP connections abruptly to be flagged as down. The numerous UP/down alerts were caused by servers that were running anti-spam software and intermittently refusing our SMTP sessions or issuing 400 deferrals.
Remember, to reduce the chance of false positives from our monitoring service, you should whitelist 22.214.171.124/24 on your email servers or gateways that we are monitoring.
We have restored the behavior of ONLY alerting you that your server is down if it fails to accept a TCP connection on port 25. Second, we have increased our timeouts, giving slow or busy servers more time to respond before flagging them as down.
We appreciate your patience as we have worked to resolve this and sincerely regret any trouble this has caused you.
I have received numerous updates throughout the day indicating that our mail server has been up and down. According to other monitoring services which I trust, there has been no interruption. Is there maybe something wrong with your monitoring service???
Thanks for your comment, I have sent this along to my Production Engineer so they can make sure everything is working properly.
Tier 2 Analyst
Thank you very much for bringing this to our attention. I will be cracking open the code to see why this is happening, but I also wanted to make sure that you are aware that you have some defensive software which might be causing at least part of this problem.
Below is the monitor history for your server around the 3:29 down alert. The 220 code is a good normal banner. The 421 code shows that some piece of hardware or software is actively blocking our mail. I fear that the “Not allowed in this state” could possibly be a complete denial of a connection which would be a more aggressive defense against our IP address than the 421.
As I said I will be looking at the code to make sure that we don’t have a bug which is causing the connection to fail from our side, but if you could possibly add our IP address range (126.96.36.199/28) to an approved list, it would help things greatly.
I will keep you posted on my progress and if you would like to have me check the monitor after you add us to any whitelist you can, I can let you know if it has made a difference from our side.
2009-05-21 15:42:48.000 421 4.0.0 Intrusion prevention active for [64.XXX.XXX.XXX]
2009-05-21 15:29:56.000 Not allowed in this state. Connect first.
2009-05-21 15:22:08.000 Not allowed in this state. Connect first.
2009-05-21 15:09:33.000 220 ns1.XXXX.com ESMTP; Thu, 21 May 2009 16:09:30 -0400
Thank you again for your feedback.
Product Development Engineer
Wendy or Peter –
What is the current status of this problem? I’m still getting several “Server Down” alerts per day for a server that, as far as I can tell, is runing normally. I never used to have this problem with this server.
More- I can go to the “Performance monitor” on my MXToolBox account and view the transaction times for the last couple of weeks – there’s no value shown that’s greater than 1.5s. If the monitor is really timing out on a connection attempt to my server, why don’t I see some huge transaction times here?
And- I’m getting “Server Down” messages several times a day, but I rarely see a “Server Up” message. I thought it only emailed me on a state change – how can I get several “Server Down” notifications in a row without an intervening “Server up”?
I tried to look up your account by name but could not find you in the system…would you mind emailing me your company name, IP or domain so we can access your account?
mail.jfcl.com (188.8.131.52) is the server, firstname.lastname@example.org is the account.
FWIW – I don’t believe I’ve received any alerts since Friday, although there were a number that came in around the end of last week.
BTW, I discovered an unrelated bug in the web page. Log in and go to “Configure Alerts” page. Where it says “Configure Alerts, Server Down” uncheck the box for “pager or mobile phone” and then click “Save Preferences.” Now visit another page and then go back to the Configure Alerts page – “mobile phone” is checked again! It’s seems impossible to turn this off – the only way I could get MXToolbox to stop SMSing my phone was to clear my mobile phone email address instead.
I would like to thank you for your help in regards to this matter. I really love how quickly you responded. This really has helped to put me at ease knowing that there is something I can do to protect myself.