Konstantin Lebedev

I'm poking around further after vacuum ``` weed -v 4 fix ./disk2 2>&1 |grep -c key 100 ``` before vacuum ``` weed -v 4 fix ./sw_fix/disk1 2>&1 |grep -c key...

We return to synchronization and try to correct the consequences of an asynchronous vacuum of volumes. When there may be more actually deleted files. ``` weed -v 4 fix ./sw_fix/disk1...

@chrislusf Hello, I think I need help. At this point I don't know how to get the old value of the deleted key. ``` I1011 16:41:21.346748 fix.go:53 key 3060820471930646086 offset...

> > 没有合并进去😭 > > 有什么问题么，这个后面有计划么 https://github.com/seaweedfs/seaweedfs/issues/4519#issuecomment-1727373429

It seems it's time to connect OpenTracing :)

I looked at my metrics on version 3.28  There is no difference in response time for S3 and filer. You need to look for a problem in your infrastructure

> Can you provide any glue to the issue. Still slow chunkAssign requests are seen on 1 of 3 filer at a time This seems like a reason to take...

@ginn13 There is a hypothesis that one of the fillers is not connected to the leader of the master or before him with a connection, something is wrong. Try to...

> I've patched 3.55 with your PR to just add this Warn log. Should I update to latest version instead? no, just need to print the error.

> @kmlebedev I get a bunch of errors like > > ``` > master_grpc_server_volume.go:173 PickForWrite error: no writable volumes available for collection:ma-bucket replication:010 ttl: > ``` > > And it's...