Skyscrapers Support Process¶
Priorities¶
We handle service requests via 3 main categories: Production down, urgent, and non-urgent. Below is the expected response time for each category, what channel to use for it and when to use it:
Production down | Urgent | Normal priority | |
---|---|---|---|
Scope | Only for requests type “Problem/Troubleshooting” about issues affecting a lot of end-users. | For requests that are blocking work or are affecting end-users | For all normal requests that can be handled in a regular flow |
Maximum response time | < 30 minutes 24/7 | 1 business hour | Next Business day (unless otherwise agreed) |
Channel to use | Escalation number | Log Github issue and @help us through Slack |
Log Github issue and we will get back to you |
Request types¶
Classifying requests allows to better collect the information beforehand, leading to a higher quality response.
Type | What? |
---|---|
Add/Remove User | A user needs to have access or a user's access needs to be revoked. |
Change Request | Something needs a change (add, modify, remove) on the existing customer environment. This can be a configuration change or technology. |
Guidance & usage support | Practical questions, guidance and explanations on architecture, technologies, reference solution, tools, etc |
On-call | Messages received during on-call |
Problem/Troubleshooting | Something is broken or not behaving as expected. Troubleshooting and investigation is needed. |
Other | Anything else |
Communication channels¶
We maintain multiple channels to interact with us. Each has its strengths, depending on the type of request and priority.
Channels | |
---|---|
GitHub Issues |
|
Slack |
|
Dedicated 24/7 phone number | For very urgent requests related to major production events requiring immediate attention, regardless of the time of day. |
Remote status meetings | Determine the status of the project and to see in which way we can help you achieve your goals. |
Escalation process when production is down or hampered¶
- If production is hampered and/or down, please use the automated number which is available in the main README file. Clearly state your name and describe the problem.
- Our 24/7 on-call engineer is automatically alerted.
- The engineer will acknowledge the problem and take the first steps towards a solution.
- Closing:
- The on-call engineer will notify you when everything is resolved
- In case of major downtime caused by the hosting platform, we will produce a post-mortem document (within ~3 days)