Fix log message format bugs #118354
Fix log message format bugs #118354
Conversation
joegallo
added
>bug
:Data Management/Ingest Node
|
Pinging @elastic/es-data-management (Team:Data Management) |
|
Hi @joegallo, I've created a changelog YAML for you. |
|
Related to #104782, in that with the changing of the seasons, I sometimes get the urge to |
joegallo
added
the
auto-backport
original-brownbear
left a comment
original-brownbear
left a comment
There was a problem hiding this comment.
LGTM one cleanup suggestion though :)
| // A normally running shard snapshot should be in stage INIT or STARTED. And we know it's not in PAUSING or ABORTED because | ||
| // the ensureNotAborted() call above did not throw. The remaining options don't make sense, if they ever happen. | ||
| logger.error( | ||
| () -> Strings.format( |
There was a problem hiding this comment.
This one is kinda nonsense? :) Why use a lambda and String.format here when you're not logging an exception?
I know it was like that before but maybe just remove the format call and lambda? :)
| return (String) asMap.keySet().toArray()[0]; | ||
| } | ||
| logger.warn("--> No matching rollup name for path [%s]", endpoint); | ||
| logger.warn("--> No matching rollup name for path [{}}]", endpoint); |
There was a problem hiding this comment.
Just dropping by, typo here {}} -> {}?
There was a problem hiding this comment.
YIKES! Great catch, Sam!
💔 Backport failed
You can use sqren/backport to manually backport by running |
5 participants
The codebase is very trappy on this, which is unfortunate, but at present we have no way of preventing this from happening (at least in the broadest sense, I think some simple special cases are checked automatically) besides being really careful all the time.
I just happened to notice that this snuck in via #117575, so I'm chasing that PR (hence the wide list of backports). I'm also including @original-brownbear as a reviewer because of the snapshot code.