yoshiyuki.kanno

Suggested by @vstax on #756. It seems useful to admins in case leo_gateway need to be got up/down elastically according to the load.

Found through the discussion on https://github.com/leo-project/leofs/issues/722#issuecomment-301007144. It would be useful for anyone using LeoFS with the multipart upload to avoid unintentional disk uses. Reference: http://docs.aws.amazon.com/AmazonS3/latest/dev/mpuoverview.html#mpu-abort-incomplete-mpu-lifecycle-config

[leo_object_storage][optimize] Calc MD5 during waiting for disk I/O

2

### Problem here is a discussion. https://groups.google.com/d/msg/leoproject_leofs/QsxY2wNWhYg/SnnyPZAbCQAJ ### TL;DR - the current: md5 after reading the whole block in one process - pros - code readability - less GC -...

[leo_gateway][cowboy] Upgrade Ninenines/Cowboy to 2.4.0

1

The following procedures will be needed to achieve the goal. ### TODO - [ ] Refactor the code in `leo_gateway` based on [Cowboy v1.0 To v2.0](https://ninenines.eu/docs/en/cowboy/2.4/guide/migrating_from_1.0/) - [ ] Refactor...

[leo_gateway] PUT/POST a large file causes higher load and memory consumption

17

### Background shared by @vstax The POST/PUT operations for large objects (which will be stored as multipart objects) create higher load on leo_gateway compared to (client-assisted) multipart uploads in S3...

[benchmark] Compare the performance between hipe-enabled and disabled

1

We've release the official LeoFS packages with hipe disabled as we faced the erlang runtime with hipe disabled has been more performant than the one with hipe enabled in the...

[leo_object_storage] Pluggable AVS

2

The spec of the current AVS aims to realize - High read/write throughput - Durable against power outages - Able to recover the index (metadata) from AVS - Free from...

Now leo_mq consumes items one by one by simply iterating all items through eleveldb's iterator, however let's say a LeoFS running with N=3, R=W=D=2 and one node in the cluster...

When creating a new file via NFS, NFS client always issues `create` and at this point leofs create a new empty file so that the size of AVS header take...

This can happen when a MDC replication succeeded but its corresponding local replication in the destination cluster failed and queue files got lost for some reason. http://leo-project.net/leofs/docs/admin_guide/admin_guide_4.html#recover-cluster-command help recover objects...