Service overview

RevDB manages your MySQL database infrastructure, provides remote DBA service and support, including, but not limited to:

Operating the entire database infrastructure end-to-end

Monitoring and alert handling 24/7, with an SLA as low as 30 minutes.

Provides highly available database backend services with automatic failure detection and remediation process.

Managing any unexpected events that affects or threatens production
Monitors performance and on demand consulting

Keeping database software stack versions up to date*

* Subject to SLA agreement

Service details

From the moment we start the onboarding, RevDB will provision all the future MySQL instances. Provisioning will include setting up and configuring MySQL and ProxySQL instances, updating and maintaining packages to the latest versions, monitoring performance, controlling the instances’ footprint. This includes complete all around service of the EC2 instance, periodic rotation and security patches as well.

This comes at no additional cost.

This provisioning process is done through RevDB infrastructure as a code solution. Once a request is made through our support system, the new cluster will be created automatically. As a result of our CI/CD mechanism, every new cluster is configured automatically. From the moment a cluster comes online, it is automatically enrolled in backups, monitoring, cost tracking and all other components we rely on.

This comes at no additional cost.

With the setup of the instances comes our own monitoring solution. For this purpose, we utilize a dedicated Datadog endpoint to the customer. Performance and availability metrics will be sent to Datadog (does not include any customer specific details), where monitor rules will help us to get alerted when a service is not available and also monitor utilization, issues and performance. When the system detects a failure, our oncall team will be alerted and committed to start working on the issue within provided SLA. The interface is available to the customer for their own use.

This comes at no additional cost.

On top of availability and server performance, we will also very closely monitor the MySQL Server performance. This includes slow or long running queries, memory usage, connection counts, replication delays etc.

This comes at no additional cost.
Alerting comes from two different sources mentioned above. A lagging replica is at least as big of a problem as a host not answering calls. In either of these cases, our oncall will be alerted through a dedicated PagerDuty service. We provide SLA on response time to any alerts that’s impacting production services.

This comes at no additional cost, time used for emergencies, counts towards support time.
No matter how excellent a team is, failures happen. In case of any issues, we have a commitment to provide an exhausting analysis and report on the failure. Backed by logs, monitoring metrics and fact based analysis.

This comes at no additional cost.

As a part of the solution, we offer a source of truth of the state of any and all clusters. In case an instance is unreachable, this service will be notified of the replacement as soon as it’s ready (failover process). When the process is completed, the applications using our service discovery can continue their operation based on the information here. This is optional, but highly recommended.

This comes at no additional cost.

With detailed and precise monitoring, we also include HA failure detection and handling. Meaning that our configuration detects any failure automatically, notifies the oncall person and attempts to process on its own. This succeeds in most cases and the service usually recovers before anyone notices under 30 seconds. However, even when automatic failure handling is unable to complete the failover, our oncall personnel will be there to recover the service as soon as possible. Which is going to be followed by a careful and detailed post mortem analysis.

This comes at no additional cost. Time used for emergencies counts towards pre-allocated support hours.

Every service and company has their own needs when it comes to disaster recovery. RevDB systems are built by default, to run from multiple availability zones. However, even the best and most reliable cloud providers have entire regions offline occasionally. That doesn’t mean your business must stop operating. RevDB can support any kind of DR requirements you might require, whether it’s a hot standby or cold site.

This comes at no additional cost.
One of the largest challenges of running database clusters is to manage the database schemas and run table alters. Most services (including RDS), do not support fully online table schema changes. Not mentioning integration testing, DBA reviews and best practices. We believe that’s unacceptable and therefore offer all of these with no downtime.

This comes at no additional cost.
All and every database shard and cluster have backups with point in time recovery capability. The backups are continuously tested to make sure the content is valid.

This comes at no additional cost.
For a complex system, user management can be difficult. In a way too many cases we have seen, as a result of the complexity, many companies don’t manage their database users properly.

Meaning, they often doesn’t enforce or maintain:
● Need to know privileges
● User and password rotation
● Encrypted password storage
● Enforced expiration
● Audit trails and approval process

With RevDB as your DBA team, we take care of user management. In case the customer uses our service discovery, this is completely transparent with no outages or effects on production.

This comes at no additional cost.

Every solution we use, every configuration, and every tool has detailed documentation on how it operates and how it is being operated and utilized. This is available to our customers. However, the know-how components are available on demand only and strictly attached to an NDA.

This comes at no additional cost. 


© 2020 Revenants CIE LLC.

US toll-free: +1-877-REVDB4U

International: +1-669-777-6044

Redwood City, CA 94061
PO Box 610126