3 ms·
Imagine the issues on something like AWS, Google Cloud, Digital Ocean, Rackspace, etc. Imagine if a company sets a limit. One that allows for burst and growth.
by rovr138 5y ago
Imagine the issues on something like AWS, Google Cloud, Digital Ocean, Rackspace, etc.
Imagine if a company sets a limit. One that allows for burst and growth. Then they forget to update it, and then one day, all their servers, data and everything else is gone.
Even if things are offline, they accrue costs, so the only thing to do when a limit is reached is destroy everything.
What’s the balance?
- heavyset_go 5y ago> Then they forget to update it, and then one day, all their servers, data and everything else is gone. Even if things are offline, they accrue costs, so the only thing to do when a limit is reached is destroy everything. That's not the only thing you can do. It takes quite a while from non-payment for OVH to take down their clients' services and delete their data, for example. They give customers a month long grace period before taking their service offline, and then there's another waiting period before they delete the data. At any point clients can resume service with payment.
- ahtihn 5y agoIt should be trivial for the provider to compute the cost of offline resources and account for it when triggering the limit to give a few days to react before destroying everything.
- joshmanders 5y agoI'm implementing this in my cloud hosting platform (https://primcloud.com https://primcloud.com) and how we plan to do it is email triggers when certain thresholds are met, and if no action is taken, give a grace period before taking stuff offline. Yes they will be charged for that grace period, but setting a limit of $200 and being charged $250 sure beats getting a random $1,000 charge.