So we’re a Mimecast customer and today they had a few problems with their email services in the UK going offline for a large part of the working day. No big deal as such, other than that they make big claims for their own availability and have been quite ready to point out the failings of others in the past – and the responsible people at those others are probably reacting to today’s events by pouring champagne down their throats as fast as they can get the bottles open… well maybe not at Google.
To my mind it’s not whether or not a company or individual will make mistakes or have a bad day – that’s a given – but how they handle it that matters. To that end, Mimecast could have handled things better to start with but they have gone on to redeem themselves in my eyes.
We noticed mail flow in and out of our systems stopping at about 10.45 am today and initially didn’t do too much about it. We initially started looking at our own systems and didn’t notice any problems. We actually had a few problems with our email system earlier in the week and we were taking some steps to fix those issues and the fixes took the brunt of our initial suspicion.
Once we’d eliminated that, our eyes turned to our rapidly filling “outgoing email” queue and from there to Mimecast.
When we first started looking for a hint to the problem we tried logging onto our administrative console at Mimecast, which of course failed. Given the serious nature of their issues today that’s again, fair enough. It happens.
However, when I went to the main website, I was a little surprised not to see any incident note. Equally there was nothing under the support heading on the site. While this has now been corrected with a couple of good posts on the subject (1 , 2), there was nothing on the site blog either. All the action was on Facebook and Twitter. While most of the nation was following End it Like Beckham on twitter, a fair few sysadmin types were following #mimecast instead.
This is a problem. Now while I’m all blogged and facebooked and tweeted to the gills myself, there are a lot of people including a lot of senior manager types in the world who still don’t understand social media. As far as these people are concerned, if it isn’t on the front page of your website or under the support heading then you’re doing nothing about it. These people are about as likely to check twitter or facebook or suchlike for updates as I am to start campaigning for electricity to be abolished and all “computer people” to be put up against a wall and shot.
Some of them might have found the updates on the blog eventually but as these went up quite late, this is something that Mimecast needs to handle better in my opinion. Twitter was where the real action was, with various people reacting with varying degrees of calmness to outright stupidity and competitors were circling like vultures, though some at least had a bit more class than others.
To Mimecast’s credit, they were still following the twitter feed, and one of the managers took the time to call me (and I suspect other customers too) who expressed concern. This call was much appreciated at my end because it allowed me to feed back some more info and more positive contact to my bosses. While they trust as a senior sysadmin to manage our email provision I’m also expected to sound like I know what’s going on during an outage, and the call from Mimecast helped greatly. More information earlier on would have been even better mind you.
I don’t expect we’ll be changing providers because of this problem or anything like that. I wonder about the people who are threatening to do so – what will they do if their next provider has problems? As I said earlier (and this is a lesson I hope Mimecast and their “100% uptime” claims will learn from too) everyone has problems and the real test is not how good you are at delaying the inevitable but how well you handle things when it does finally catch up with you.Todays incidents have proven that the cloud is no magic bullet. This might put some people off moving to the cloud (though if I were a cloud vendor I’d be asking myself if I wanted the hassle of customers who really believed that anyway) but if Mimecast and other cloud/SaaS vendors can show how they learn from problems like this then they might be able to turn it to their advantage. Being able to accept and deal with my point above about the inevitablity of problems is a sign of maturity in all segments of the IT industry.
I think the more fundamental issue here is that Mimecast positions itself as a continuity provider to deliver guaranteed availability when your own mail routing fails and here they are the root cause of the issue.
Customers may not move, but they will enact their contract SLA terms and find out if they have what they expect as mitigation to the issues. I suspect many business managers will be querying IT as to what they get back from Mimecast for this and with a ton of Mimecast customers in the legal sector that could pose for some interesting compensation and legal discussions.