Conversation
| if total_checkpoints_pruned > 0 { | ||
| if let Some(tracker) = progress_tracker { |
There was a problem hiding this comment.
I think it latest rust you can write something like if let Some(x) = var && var2 > 0 {...}
| checkpoint_pruned_seq_number.saturating_sub(prev_ckpt_pruned); | ||
|
|
||
| if exec_delta > 0 || object_prune_delta > 0 || checkpoint_prune_delta > 0 { | ||
| let exec_time_delta = |
There was a problem hiding this comment.
do we need to subtract those times? can we take the value and swap with 0 and always print what we've taken from the atomic? that way we don't ever risk overflows or anything.
roman1e2f5p8s
left a comment
roman1e2f5p8s
left a comment
There was a problem hiding this comment.
could we also mention in the PR description (maybe even in release notes) that:
- log level for some events changed from
infotodebug - some public APIs in
iota-coregot an extra optional parameter
muXxer
force-pushed
the
feat/node-reduce-log-spamming
branch
from
e8ca563 to
b92aa4e
Compare
I added in the description for the log messages that were changed. The public API's in |
6 participants
Description of change
This PR massively reduces the log spamming in the
iota-nodeon"info"level.It introduces a periodic status message that is printed every second, that reports about progress of synced and pruned checkpoints, instead of prints per every single received, executed and pruned checkpoint.
It also removes the log spam if peers don't have the range of checkpoints we want to sync, which was repeatedly 400 lines every 10s.
The
lowestavailable checkpoint is now calculated based on the information from the connected peers on the same chain. Before it was based on all connected peers, which was wrong.We reduced the log level from
infotodebugfor"scheduling pruning for checkpoint","executing checkpoint"and"received synced checkpoint". The Log message"Finished syncing and executing checkpoint"was removed and replace by theCheckpointProgressTrackermessage.How the change has been tested
Release Notes