[FLINK-18662] Provide more detailed metrics why unaligned checkpoint is taking long time

• Type: Improvement
• Status: Closed
• Priority: Critical
• Resolution: Fixed
• Affects Version/s: 1.11.1
• Fix Version/s: 1.12.0
• Component/s: Runtime / Metrics, Runtime / Network, Runtime / Web Frontend
• Labels: pull-request-available

Release Note

This is modifing the {{checkpointAlignmentTime}} metric for the at-least-once checkopinting and unaligned exactly-once checkpointing. Previously it was
always 0, now it's defined as the duration between processing first and the last checkpoint barrier.

• Assignee: Piotr Nowojski
• Reporter: Piotr Nowojski
• Created: 21/Jul/20 10:31
• Updated: 27/Jan/21 12:25
• Resolved: 13/Oct/20 15:17

Description

With unaligned checkpoint there can happen situation as in the attached screenshot.

Task reports long end to end checkpoint time (~2h50min), ~0s sync time, ~2h50min async time, ~0s start delay. It means that task received first checkpoint barrier from one of the channels very quickly (~0s), sync part was quick, but we do not know why async part was taking so long. It could be because of three things:

1. long operator state IO writes
2. long spilling of in-flight data
3. long time to receive the final checkpoint barrier from the last lagging channel

First and second are probably indistinguishable and the difference between them doesn't matter much for analyzing. However the last one is quite different. It might be independent of the IO, and we are missing this information.

Maybe we could report it as "alignment duration" and while we are at it, we could also report amount of spilled in-flight data for unaligned checkpoints as "alignment buffered"?

Ideally we should report it as new metrics, but that leaves a question how to display it in the UI, with limited space available. Maybe it could be reported as:

Alignment Buffered	Alignment Duration
0 B (632 MB)	0ms (2h 49m 32s)

Where the values in the parenthesis would come from unaligned checkpoints.

Issue Links

• contains: FLINK-17572 Remove checkpoint alignment buffered metric from webui
• is related to: FLINK-18656 Start Delay metric is always zero for unaligned checkpoints and source tasks
• is related to: FLINK-17468 Provide more detailed metrics why asynchronous part of checkpoint is taking long time
• links to: GitHub Pull Request #13562