Return-Path: Received: from fieldses.org ([173.255.197.46]:51852 "EHLO fieldses.org" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S934132AbdIYNVq (ORCPT ); Mon, 25 Sep 2017 09:21:46 -0400 From: "J. Bruce Fields" To: linux-nfs@vger.kernel.org Cc: "J. Bruce Fields" Subject: [PATCH 0/2] tweak knfsd session slot table sizing Date: Mon, 25 Sep 2017 09:21:42 -0400 Message-Id: <1506345704-9486-1-git-send-email-bfields@redhat.com> Sender: linux-nfs-owner@vger.kernel.org List-ID: From: "J. Bruce Fields" Current client and server defaults can limit NFSv4.1+ mounts to as few as 4 per gigabytes of server RAM, and people have been running into the limit on upgrading the protocol version from 4.0 to 4.1. Bump the limit. Also return smaller ca_maxrequests as the limit approaches instead of waiting till we have to fail CREATE_SESSION completely. A few years ago Trond also posted some patches to dynamically resize existing sessions as limits approach--I'll revisit those at some point too. The numbers chosen here are honestly still a little arbitrary, and we may tweak them again if we see more problems. --b. J. Bruce Fields (2): nfsd: increase DRC cache limit nfsd: give out fewer session slots as limit approaches fs/nfsd/nfs4state.c | 5 +++++ fs/nfsd/nfssvc.c | 2 +- 2 files changed, 6 insertions(+), 1 deletion(-) -- 2.13.5