fredia commented on code in PR #22772:
URL: https://github.com/apache/flink/pull/22772#discussion_r1268077347
##########
docs/content/docs/ops/metrics.md:
##########
@@ -1343,6 +1343,11 @@ Note that for failed checkpoints, metrics are updated on
a best efforts basis an
<td>The time in nanoseconds that elapsed between the creation of the
last checkpoint and the time when the checkpointing process has started by this
Task. This delay shows how long it takes for the first checkpoint barrier to
reach the task. A high value indicates back-pressure. If only a specific task
has a long start delay, the most likely reason is data skew.</td>
<td>Gauge</td>
</tr>
+ <tr>
+ <td>checkpointRestoreTime</td>
+ <td>The time in milliseconds that one task spends on
restoring/initialization, return 0 when the task is not in
initialization/running status.</td>
+ <td>Counter</td>
Review Comment:
I renamed `checkpointRestoreTime` to `initializationTime` in subtask level,
and moveed it to [`IO`
section](https://nightlies.apache.org/flink/flink-docs-master/docs/ops/metrics/#io).
For `Availability` section, I think it's a metric that describes job level
status, we can reuse `initializingTime/initializingTotalTime`.
--
This is an automated message from the Apache Git Service.
To respond to the message, please log on to GitHub and use the
URL above to go to the specific comment.
To unsubscribe, e-mail: issues-unsubscr...@flink.apache.org
For queries about this service, please contact Infrastructure at:
us...@infra.apache.org
