You signed in with another tab or window. Reload to refresh your session.You signed out in another tab or window. Reload to refresh your session.You switched accounts on another tab or window. Reload to refresh your session.Dismiss alert
Copy file name to clipboardExpand all lines: content/docs/advanced/blackbox.mdx
+14Lines changed: 14 additions & 0 deletions
Display the source diff
Display the rich diff
Original file line number
Diff line number
Diff line change
@@ -6,6 +6,20 @@ weight: 30
6
6
7
7
# BlackBox Flight Recorder
8
8
9
+
<Callouttype="info"title="Do I Need This?">
10
+
11
+
**For most users: Not initially.** BlackBox is for when you need to understand crashes and failures.
12
+
13
+
**When this page IS useful:**
14
+
- Robot crashes and you need to know what led up to it
15
+
- Running long-term deployments and need audit logs
16
+
- Investigating intermittent failures
17
+
- Compliance requirements for event logging
18
+
19
+
**Quick answer:** Add BlackBox once your robot runs well. It's like dashcam footage: unnecessary until something goes wrong.
20
+
21
+
</Callout>
22
+
9
23
The BlackBox flight recorder provides continuous event logging in a circular buffer for post-mortem analysis. Like an aircraft flight recorder, it captures all significant events leading up to failures for debugging and incident investigation.
Copy file name to clipboardExpand all lines: content/docs/advanced/checkpoint.mdx
+13Lines changed: 13 additions & 0 deletions
Display the source diff
Display the rich diff
Original file line number
Diff line number
Diff line change
@@ -6,6 +6,19 @@ weight: 46
6
6
7
7
# Checkpoint System
8
8
9
+
<Callouttype="info"title="Do I Need This?">
10
+
11
+
**For most users: Not initially.** Checkpointing is for long-running systems that need recovery.
12
+
13
+
**When this page IS useful:**
14
+
- Robot runs for hours/days and must recover from crashes
15
+
- Training ML models and need to save progress
16
+
- Complex state that's expensive to rebuild from scratch
17
+
18
+
**Quick answer:** If restarting your robot is quick and easy, skip this. Add checkpointing when downtime matters.
19
+
20
+
</Callout>
21
+
9
22
The Checkpoint System provides periodic state snapshots that can be used to recover from crashes or rollback to known-good states. This is essential for long-running robotics applications that need fault tolerance.
Copy file name to clipboardExpand all lines: content/docs/advanced/circuit-breaker.mdx
+13Lines changed: 13 additions & 0 deletions
Display the source diff
Display the rich diff
Original file line number
Diff line number
Diff line change
@@ -6,6 +6,19 @@ weight: 32
6
6
7
7
# Circuit Breaker
8
8
9
+
<Callouttype="info"title="Do I Need This?">
10
+
11
+
**For most users: Not initially.** Circuit breakers add resilience but also complexity.
12
+
13
+
**When this page IS useful:**
14
+
- A sensor or node occasionally fails/hangs
15
+
- You need graceful degradation when hardware is unreliable
16
+
- Building long-running systems that must stay up
17
+
18
+
**Quick answer:** Start without it. Add circuit breakers when you have specific reliability problems to solve.
19
+
20
+
</Callout>
21
+
9
22
The circuit breaker pattern prevents cascading failures by temporarily disabling failing nodes. When a node fails repeatedly, the circuit "opens" to stop calling it, giving it time to recover.
0 commit comments