On this page 6 sections
1. General principle#
In short: Do not use the broker for anything illegal, harmful, or that damages it for other people.
Use AyaiMQ lawfully, responsibly, and in a way that does not harm the service, other customers, or third parties.
2. Prohibited activities#
In short: The list you would expect. Nothing surprising in it.
You must not use AyaiMQ to:
- Break any law, or infringe anyone's rights including intellectual property or privacy rights.
- Send malware, or content that is unlawful, defamatory, or harmful.
- Gain unauthorised access to any system, account, cluster, or data, or attempt to bypass authentication or permission controls.
- Interfere with or disrupt the service, for example through denial of service attacks, flooding, or abusive message volumes.
- Send unsolicited bulk messages or use the broker to relay spam.
- Resell or provide the service to third parties except as expressly permitted.
3. Security testing#
In short: Ask first. Report anything you find and we will work with you on it.
Do not conduct penetration testing or vulnerability scanning against AyaiMQ without our prior written consent. If you believe you have found a vulnerability, report it responsibly to support@logicboks.com.
4. Fair use and the limits on your plan#
In short: Messages are unlimited and never charged. Traffic has a fair use figure so that promise can stay true, and reaching it starts a conversation rather than an invoice.
Your plan sets how many clusters you may have, how many credentials each may carry, how many clients may be connected at once, and how quickly new connections may be made. Those limits are enforced by the platform, not by us watching, and the console shows you where you stand against each of them.
The connect rate limit deserves a sentence of its own, because it is the one people meet by accident. It exists so that one misconfigured device reconnecting in a loop cannot take the broker down for everybody else. A device reconnecting on the same client id does not count against it.
Separately from your plan, we may apply reasonable limits to protect the health of the platform. We will act reasonably and tell you where practical.
There is no limit on the number of messages and there is never a charge for them. Publish as much as your devices need to. That is not a beta concession and it is not going to change: it is the whole point.
What each plan does carry is a fair use figure for TRAFFIC, shown on the plan itself, and it is there so that the sentence above can stay literally true rather than being quietly walked back later. It is set far above what any real estate moves - a thousand devices reporting every second uses a small fraction of the Building allowance - so you should never meet it by using the service normally. In practice the accounts that approach it are running a device stuck in a loop, sending the same value thousands of times a second to nobody.
If you do reach it, nothing is charged and nothing is cut off. We get in touch, tell you which client it is, and help you work out whether it is a fault or whether you have outgrown a shared broker and want a machine of your own.
5. Your responsibility for devices and content#
In short: You own the credentials you issue and what your devices publish.
You are responsible for the devices you connect, the credentials you issue, the permissions you set, and the content you publish. Scope credentials to the least access they need and rotate them if they may be compromised.
6. Enforcement, and what we will actually do#
In short: We limit the one connection causing the problem, tell you which it was, and help you fix it. Suspending an account is the last step, not the first.
If you breach this policy we may rate limit or disconnect a client, suspend a credential, suspend or limit an account, or terminate it. Where there is a risk of harm to the platform or to another customer we may act immediately.
In practice almost every case is a misconfigured device rather than anybody doing anything wrong, and we treat it that way. The order we work in is: limit the specific connection that is causing the problem, tell you what we did and which client it was, and help you fix it. Suspending an account is the last step and not the first.
We will always tell a person what happened. A connection that stops working with no explanation is worse for you than the original problem, because you cannot fix what nobody told you about.
If you think we have got it wrong, reply to whatever we sent you or write to support@logicboks.com and a person will read it.
Questions about this document? Write to support@logicboks.com.