Support and Escalation

Introduction

Your Echo applications are now up and running in production. Here are some answers to some of the questions you may have regarding managing and updating your Echo-related applications in the weeks and months to come. Although it is Echo's priority to minimize the impact of applications running on production, we continuously improve our cloud service and client components. We are also prone to un-scheduled service disruptions to which we want everyone to be as prepared as possible.

Echo releases process

Echo releases both infrastructure, API and client updates on a daily basis during the Echo available release windows which are currently Wednesday, Thursday, and Friday between 12am and 3am PST. Echo is committed to release backward compatible code. 

Exceptional releases can be made at anytime based on urgency. Those exceptional releases are first authorized by senior management.

Code release with no perceptible impact on users

Regular code releases are made several times a week during the release windows. 

Information about each release is posted on the Echo Developer Forum as they occur. Because these releases have no perceptible impact on the user experience, they are not announced ahead of time. The engineering release team first announces the items to be released and then announce what was actually released. Should post-release testing reveal any issues, the release team will immediately roll-back to the previous version.

Here is an example of release announcements on the Echo developer forum. 

@EchoReleases + SMS alerts 

Echo releases are also Tweeted at @EchoReleases. Your OPS, Dev, Community Support and any other relevant teams can subscribe to that Twitter account to get SMS directly on your cell phones.

To receive @EchoReleases by SMS, go to @EchoReleases, and click the Mobile subscription icon and follow the instructions. You should be able to set the following options.

Code release with perceptible impact on users

In some rare instances, some improvements require a release that will require an action from customers. Those releases are announced ahead of time (depending on the impact up to a month). Our largest customers are also informed directly.

The announcements include a detailed description of the upcoming release, its tentative date and time, the nature of the release, the action required by the customer, and the benefits.

Release notes

The aggregated release notes can be found there.

Echo outage procedures

Echo is continuously investing in engineering efforts to reach a high level of performance and up-time. Nevertheless, ongoing and rapid growth combined with innovation sometimes leads to un-predicted service degradation and in some instances to a global system outage.

Our OPS team monitors several internal dashboards 24/7 in addition to receiving automated alerts. The development teams are alerted as soon as a sub-system shows risks of failure before impacting end users.

No system is perfect, however, and in some occasions, the system will experience outages. When this happens, we have a strict outage internal procedure that ensures communication, investigation and resolution within the shortest amount of time possible.

@EchoStatus + SMS alerts 

All service level changes are immediately tweeted @EchoStatus. We recommend that your OPS, Dev, Community Support and any other relevant teams subscribe to that Twitter account to get SMS directly on your cell phones. To receive @EchoStatus by SMS, go to @EchoStatus, and click the Mobile subscription icon and follow the instructions. You should be able to set the following options.

The Echo team will provide follow-up updates through the same channel, including notice when the system is back to normal.

Echo Developer forum alerts

All service level changes are immediately communicated through the Echo Developer Forum.

Outage logs

The Echo development and OPS teams use several internal monitoring and charting tools. We are planning to expose some public-facing monitoring tools in the future. 

In the short term, we are sharing an outage log here.

Monitoring Echo API

Some of our customers are monitoring the performance and up-time of their Echo based applications using off-the-shelf monitoring solutions such as HP SiteScope. Those tools allow them to break-down and track the time of each processes involved to deliver the end-experience. This includes, for example, the time of client side load times, Echo API submission, Echo processing and Echo API query response time, network latency, etc.

Echo offers a generic monitoring dashboard located at http://aboutecho.com/status. This dashboard reports every 5 minutes on the performance and availability of Echo StreamServer item API.

Escalation process

Day-to-day support

Most of the day-to-day development questions are discussed on the Echo Developer Forum. The turn around is usually within a few hours during weekdays, with good coverage during the night.

Should you have a more private/sensitive question, send them to solutions@aboutecho.com. Urgent questions are usually identified within a few minutes and escalated internally.

Emergencies

In case of emergency (reporting an outage for example), use emergency@aboutecho.com. This email address has a large distribution including via cell phone text alerts. More about this emergency email address can be found here. This is the most effective emergency escalation process, 24/7.

Echo also offers a 24/7 phone emergency number.

Cost of support

Basic responses including Echo-related issues are included with your Echo license.

You can also purchase additional services such as active engineering support to help implement your Echo-based applications, or the development of Echo plugins to enable custom features on your site. Echo can also provide core business and strategic expertise and product recommendations based on your requirements.  Please contact sales@aboutecho.com to discuss your needs.