Konstantin Lebedev

@chrislusf This happened after the commit https://github.com/seaweedfs/seaweedfs/commit/d8c574a5ef1a811f9a0d447097d9edfcc0c1d84c

> > [@chrislusf](https://github.com/chrislusf) This happened after the commit > > [d8c574a](https://github.com/seaweedfs/seaweedfs/commit/d8c574a5ef1a811f9a0d447097d9edfcc0c1d84c) > > This commit is between 3.87 ～ 3.88？ No, I made a cherry pick in 3.85, this is...

@chrislusf We can rename the fsync option and how to do it on the wiki for description ? Since it overlaps with the system call name. > fsync is a...

> One concern is that if one file is partially updated, for example, in a file that is split into 2 chunks, and only one chunk is updated, than the...

@chrislusf @FQHSLycopene I took a closer look at the proposed PR and suggest implementing this logic exclusively on the S3 service side, as it currently does with deleting empty folders,...

> S3 objects are always creation anyway. No need to use modification time for TTL? For chunks in volumes you don't need it, but you do need the delete function...

weed fix ``` I0507 08:07:18.866442 volume_read.go:270 ==> new entry offset 437448303264 I0507 08:07:18.866446 volume_read.go:277 new entry needle size:0 rest:16 I0507 08:07:18.866449 fix.go:56 key 0 offset 437448303264 size 0 disk_size 32...

I apologize, but it seems the problem is in my changes and the delay due to the vacation of the next PR https://github.com/seaweedfs/seaweedfs/pull/5884.

@faurik > But intermediate dir is missing Maybe it's worth turning it on? `-allowEmptyFolder=false`

Use cmd for build binary `CGO_ENABLED=0 go build`