From f61fef3b25b3801ba545af82db0466d2bf6712b0 Mon Sep 17 00:00:00 2001 From: orthoxerox Date: Sun, 11 May 2025 13:24:05 +0300 Subject: [PATCH] Documenting PR #6775 --- Hadoop-Compatible-File-System.md | 5 +++++ 1 file changed, 5 insertions(+) diff --git a/Hadoop-Compatible-File-System.md b/Hadoop-Compatible-File-System.md index 0af4eb3..76e47ea 100644 --- a/Hadoop-Compatible-File-System.md +++ b/Hadoop-Compatible-File-System.md @@ -78,6 +78,7 @@ Both reads and writes are working fine. 1. `fs.AbstractFileSystem.seaweedfs.impl`: Add the SeaweedFS implementation of Hadoop AbstractFileSystem to delegates to the existing SeaweedFS FileSystem and is only necessary for use with Hadoop 3.x. 1. `fs.seaweed.buffer.size`: Optionally change the default buffer size 4194304 to a larger number. It will be the default chunk size. 1. `fs.seaweed.volume.server.access`: `[direct|publicUrl|filerProxy]` Optionally access volume servers via their publicUrl settings, or use filer as a proxy. This is useful when volume servers are inside a cluster and not directly accessible. + 1. `fs.seaweed.filer.cn`: if you have enabled gRPC security on your filer and the certificate doesn't match the hostname of the filer, set this to the CN of your filer's TLS certificate (see [[[[Security-Configuration]], `grpc.filer.cert`). Seaweed doesn't use host name verification anywhere, but Java libraries for gRPC do, so to sidestep this you need to pass them the expected CN instead. ``` @@ -102,6 +103,10 @@ Both reads and writes are working fine. direct + + fs.seaweed.filer.cn + filer01 + ``` * Deploy the SeaweedFS HDFS client jar