Allow AsyncWrapper to flush in batches instead of one giant flush
Nobody has claimed this yet.
- Dominant language
- C#
- Stars
- 6.5k
- Forks
- 1.4k
- Avg merge
- 9h 17m
- Merged PRs (30d)
- 24
Description
Have an application that writes several GByte data using NLog as part of startup-rutine.
After startup-rutine has completed, and just about to go online, then it performs a blocking garbage collection.
It could be nice if one could flush NLog buffers before starting the garbage collection. The problem is that the async-target might have a lot of data pending at this point.
If one calls NLog LogManager.Flush() then it will take the entire async-queue-buffer and write in one operation.
Lets say 200.000 items are still pending and each of them is 1000 bytes. Then it becomes close to 2 gigabyte which is the upper limit for a MemoryStream. Calling NLog LogManager.Flush() will then either lead to application crashing with out-of-memory exception, or discarding data because of stream-is-too-long.
It could be nice if AsyncWrapper had a MaxBatchSize (nullable) that could be configured instead of BatchSize. And it would also have effect during flush-operations. Instead of writing 200.000 in one operation, then it would do 200 write-operations of MaxBatchSize=1000
Contributor guide
First steps
- Read the whole issue, then the project's contributing guide.
- Comment on the issue to say you are picking it up — it saves two people doing the same work.
- Fork the repository and make your change on a branch.
- Open a pull request that references the issue number.
Research direction
Start by locating AsyncWrapper and the LogManager.Flush() path, then trace how pending queue items are passed to the target during a flush. Verify the behavior with a large pending queue and confirm that a configured maximum batch size limits each flush operation without discarding data.
Written by the indexing model from the issue text.
Assessment
- Tech stack
- csharp
- Domain
- observability
- Issue type
- Feature
- Difficulty
- 3/5
- Estimated time
- 1-2 days
- Activity status
- Stale
- Clarity
- Mostly clear
- Newbie friendliness
- 42/100