System restarts in the Lobster Cloud are managed by the Lobster Cloud Operations team. You cannot restart your system yourself. This page explains how to request a restart, what to expect, and important scheduling considerations.
Requesting a restart
To request a system restart, submit a support ticket to support@lobster.de with the following information.
Information | Description |
|---|---|
System | Which system should be restarted (production, test, DEV), and what is the System ID (example: 0001t). |
Preferred schedule | Your preferred date and time for the restart. Allow sufficient time for the team to respond. |
Reason | Brief description of why the restart is needed (for example, after downloading a software update). |
How restarts are performed
Aspect | Details |
|---|---|
Performed by | Lobster |
During service hours | Restarts are carried out during standard service hours (Monday to Friday, 08:00 to 17:00 CET). |
After-hours restarts | It can be planned using automated processes. For work outside business hours, a lead time of at least one working day must be allowed. |
Downtime | A standard restart typically takes a few minutes. |
Automatic restarts
The monitoring system can trigger automatic restarts in certain situations:
Trigger | Action |
|---|---|
Lobster service failure (service down or crashed) | Automatic restart attempted. |
Out-of-memory error | Automatic service restart. |
These automatic restarts happen without a support ticket and are part of the standard monitoring response.
Scheduling recommendations
Recommendation | Details |
|---|---|
Avoid Fridays, weekends, and holidays. | In rare cases, automated restarts may not complete successfully outside service hours. If an issue occurs, resolution may be delayed until the next service day. |
Plan restarts during service hours. | This ensures that a cloud administrator is available to monitor the restart and respond to any issues. |
Combine restarts with updates. | If you are managing your software updates, request the restart at the same time as submitting the update ticket. Download the update approximately 30 minutes before the planned restart. |
HA system restarts
For High Availability environments, restarts require special handling.
Info
For information on system availability during software updates and planned maintenance, see Maintenance Schedule
Aspect | Details |
|---|---|
Restart sequence | DMZ servers are restarted sequentially (one after another). After all DMZ servers are completed, all internal nodes (Node Controller and Working Nodes) are stopped, then restarted in order: Node Controller first, then Working Nodes. |
Downtime | Approximately 10 to 15 minutes for standard restarts, 15 to 30 minutes for restarts that include patching. |
Mandatory procedure | This sequence is required to prevent database corruption and inconsistent data. |