I've been running Marathon in production (real production) to power more than 100 applications for the past six months. I chose it because it seemed like the most stable thing at the time; however, quickly found it was not production ready. While many of the original issues I encountered in 0.7.x were resolved with the 0.8.x release, 0.8.x brought new issues such as stuck deployments, etc. Additionally, I have found the upgrade path to be obtrusive and frankly scary. I am actively moving away from Marathon because of these issues.
Marathon does not make using Docker or building microservices simple. There are many important pieces that Marathon does not provide. Sure your operations team can tie in Mesos-DNS / Bamboo / Consul / whatever else, but it's going to take time, requires a specialized team, and leaves you feeling nervous about what happens if everything crashes in the middle of the night. Even when tying in these third party tools, it is likely you will have to make significant code updates to utilize features such as service-discovery / SRV records. You will inevitably end up with a hobbled-together system that needs serious support from your operations team.
I am fairly frustrated as a whole with Mesosphere, and expected more from a company who raised so much capital.
I like the idea they present in the post: "We want to let users build and deploy applications whose components span entire datacenters, but with the simplicity of programming for a single machine." Unfortunately in my experience that isn't a reality. DCOS / Marathon do not provide the necessary components to really make this feasible. Sure your operations team can tie in Mesos-DNS / Bamboo / Consul / whatever else, but that completely negates the statement, while also requiring a specialized team. Even when tying in these third party tools, it is likely you will have to make significant code updates to utilize features such as service-discovery / SRV records.
It seems like Mesosphere's real innovation is a wrapper around Mesos with a pretty UI sitting on top. Many operations teams will still require standing up their own environment to install necessary packages for compliance, etc. At that point what makes the DCOS install easier than Mesos? By creating this wrapper around Mesos, it seems like the issue they present: "needing a team of distributed systems experts", is simply changed to needing a team of DCOS / Marathon experts. Having a team of DCOS experts will not replace your operations team.
In my opinion most of the selling points mentioned in the article are really Mesos benefits, rather than anything DCOS itself provides. The image in the article with features of the Enterprise Edition only really provides 3 benefits over Mesos itself:
* Platinum 24x7x365 support (which is hopefully better than what I've experienced thus far with Marathon)
* Emergency Patches
* Professional Training
I am not sure why teams would pay for an annual per-node Enterprise Edition of DCOS when they are just as capable of deploying Mesos themselves. Either way they are ending up with a hobbled-together system needing serious support from an operations team.
Marathon does not make using Docker or building microservices simple. There are many important pieces that Marathon does not provide. Sure your operations team can tie in Mesos-DNS / Bamboo / Consul / whatever else, but it's going to take time, requires a specialized team, and leaves you feeling nervous about what happens if everything crashes in the middle of the night. Even when tying in these third party tools, it is likely you will have to make significant code updates to utilize features such as service-discovery / SRV records. You will inevitably end up with a hobbled-together system that needs serious support from your operations team.
I am fairly frustrated as a whole with Mesosphere, and expected more from a company who raised so much capital.