Is Scalr stable enough for production usage?
Hi
From what I've seen, this service seems excellent, very impressive. I just wonder how people are finding this for stability as I only started using it yesterday and so far haven't been able to build any roles due to a bug. Because of this it's made me a bit cautious as if this happened at a critical time, if it took a day for us to be able to build a server it may be a big issue
Thanks for your input
Comments are currently closed for this discussion. You can start a new one.
Support Staff 2 Posted by Nick Toursky on 08 Dec, 2011 12:58 PM
Hi Ben,
I've moved this discussion to 'questions' category as I believe this is more of a question for our community. Correct me if I'm wrong.
Regards,
Nick
3 Posted by Andreas on 09 Dec, 2011 01:51 PM
Hi Ben,
Our 2 month experience has so far been great and scalr itself has been very robust.
Remember that scalr is just a management layer working over your cloud provider. A lot of a system's stability depends on the actual instance's configuration (e.g. we use monit to build automated self healing scenarios for various services failing etc)
Any issues so far have been related to custom roles/scripts or us doing some "dirty" manual action not supported via the UI.
We have the emergency service and the support team has been very helpful in guiding us how to resolve such situations with their knowledge of what should be happening behind the scenes.
Scalr for us has been an accelerator for moving to the cloud. It is a great framework with the server templates, preconfigured backups, scaling algorithms etc it allowed us to quickly apply good practices to our setup. Now we are moving to the next level by making use of scalr command line tools to automate lots of IT operational tasks. I believe scalr is also working on monitoring solution as well.
hope this helps
4 Posted by Ben Sebborn on 09 Dec, 2011 02:03 PM
Thanks for this
I tested the system using EC2 as Rackspace wasn't working well, and it seemed to work very very well
Unfortunately, every task we've tried to do with Rackspace (build a role, build a server farm) has failed so far and we've had to wait whilst bugs were fixed.
I don't mind this at the moment but it's quite worrying that we could put this into production and a server fails to come up!
5 Posted by Andreas on 09 Dec, 2011 02:15 PM
Yeah I forgot to mention we only have experience over amazon's cloud.
Support Staff 6 Posted by Sebastian Stadil on 10 Dec, 2011 08:39 AM
Amazon represents about 95% of our managed servers, so we prioritize it.
Julien Rialan closed this discussion on 22 Feb, 2012 06:30 PM.