From mboxrd@z Thu Jan 1 00:00:00 1970 Return-Path: Received: from fieldses.org ([174.143.236.118]:60700 "EHLO fieldses.org" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S1751377Ab0CIWel (ORCPT ); Tue, 9 Mar 2010 17:34:41 -0500 Date: Tue, 9 Mar 2010 17:36:06 -0500 To: iisaman@citi.umich.edu Cc: linux-nfs@vger.kernel.org Subject: pynfs Message-ID: <20100309223606.GA29261@fieldses.org> Content-Type: text/plain; charset=us-ascii From: "J. Bruce Fields" Sender: linux-nfs-owner@vger.kernel.org List-ID: MIME-Version: 1.0 Currently some of the pynfs tests handle delay errors by sleeping up to 20 seconds. This can cause spurious failures in some cases if your lease time happens to be less than 20 seconds. An ideal fix might be to query the lease time and use it as an upper bound for those delays. Alternatively, we could set the maximum very small, and/or just document a minimum acceptable server lease time. But it's tempting to set the lease time to something very short when running the pynfs tests, since it can make a huge difference to their run-time. Just noting the problem for future reference.... --b.