Downtime and Emails/Notifications versus Forum Threads and Twitter

[Randy]

Staff Edit:This was split off of another thread, to keep that thread concise and to the point. The original thread can be seen here.

 

--------

 

I had to join this forum just to say this:

 

We should be notified if our sites are going down by email automatically. I shouldn't have to join a forum I found out about from your twitter account to know when my site is going down randomly.

Edited November 4, 2011 by MikeDVB
Staff Edit

[Michael D.]

I had to join this forum just to say this:

 

We should be notified if our sites are going down by email automatically. I shouldn't have to join a forum I found out about from your twitter account to know when my site is going down randomly.

We do send emails (for example, the most recent issue):

http://www.screen-shot.net/2011-10-20_1346.png

Here is a link to that thread: [Emergency] Fresco RAM Replacement

Here is a link to the tweet: http://twitter.com/#!/MDDHosting/status/127068529676656640

 

The forums are just a more centralized way to notify users and allows users to choose to be informed about less-than-critical issues should they decide to do so. We use multiple mechanisms to inform our customers of issues.

 

I do, however, appreciate your feedback.

[Randy]

I never recieved a email about it. Is this due to my site not being on that server? Just want to make sure.

[Michael D.]

I never recieved a email about it. Is this due to my site not being on that server? Just want to make sure.

That would be correct, you would only be notified if your service with us was affected.

[die2mrw007]

i never did received any such email even though my site is affected.

Its so strange. Visiting the forum just to get with maintenance or issue updates is nearly not possible or inconvenient to users.

Please provide an instant email notification of critical issues and maintenance work.

[Scott]

Its so strange. Visiting the forum just to get with maintenance or issue updates is nearly not possible or inconvenient to users.

 

As Michael demonstrated above, you can easily subscribe to the this forum for non-critical updates and updates that do not effect your service specifically.

 

Please provide an instant email notification of critical issues and maintenance work.

 

Also as Michael has pointed out, you should be emailed automatically for critical issues and maintenance. If you are not getting these emails, please check your spam folder and/or white list all @mddhosting.com email addresses. You should also check your profile in our billing and support system to ensure that we have the correct email to send these notices to.

 

Lastly, if your primary email address is hosted on our services, you wouldn't get some notifications, such as notifications for downtime... because the service, which includes this email address, is down. For this reason, we strongly suggest using an off-service email account, or at least provide one as an alternate contact.

[Michael D.]

The DDoS attack was affecting our entire network, it's really not efficient to send out thousands upon thousands of emails when we can make a singular post in a centralized location and direct anybody reporting issues or wanting more information to that centralized location. We worked quickly to mitigate the issues so there was only minutes of slowness and then a few minutes of reported downtime today which is quite incredible considering the size and type of attack.

 

The total downtime for today on our network was 21m 55s as reported by Pingdom but a vast majority of that time you could connect (although very slowly) causing Pingdom to simply time-out and report a failure.

 

Now don't get me wrong, I'm not trying to make any sort of excuses - there was an attack on our network and we worked to resolve it as quickly as possible. That 22 minutes was spread over 8 individual outages throughout the day with the average length being 2m 44s (directly from Pingdom) meaning that each and every time there was an issue we were on it and resolved it quickly.

 

At the end of the day, we had no way of knowing when the attack was going to move/change so no way to warn our customers. It could have stayed on the original IP and never caused more damage once that one was null-routed or it could have shifted numerous times as it did.

 

We don't hide our uptime history or reports, you can see them freely on our server status page. This thread was designed to simply inform new users to the forums that they could watch this forum section for updates and not to discuss internal company policies and procedures so I'm splitting these posts off into a new thread and then I'm going to lock the original thread so it stays concise and to the point.

[Michael D.]