From mboxrd@z Thu Jan 1 00:00:00 1970 From: Josh Durgin Subject: Re: Hang when mapping a long name rbd image Date: Fri, 16 Dec 2011 18:40:01 -0800 Message-ID: <4EEC0101.3030905@dreamhost.com> References: <8512670932FB654F81AF0FEF1BE6D49DA2ECA6@WHQBEMAIL1.whq.wistron> Mime-Version: 1.0 Content-Type: text/plain; charset=ISO-8859-1; format=flowed Content-Transfer-Encoding: 7bit Return-path: Received: from mail.hq.newdream.net ([66.33.206.127]:53099 "EHLO mail.hq.newdream.net" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S1752814Ab1LQCkC (ORCPT ); Fri, 16 Dec 2011 21:40:02 -0500 In-Reply-To: <8512670932FB654F81AF0FEF1BE6D49DA2ECA6@WHQBEMAIL1.whq.wistron> Sender: ceph-devel-owner@vger.kernel.org List-ID: To: Eric_YH_Chen@wistron.com Cc: ceph-devel@vger.kernel.org, Chris_YT_Huang@wistron.com On 12/16/2011 06:21 PM, Eric_YH_Chen@wistron.com wrote: > Hi, all: > > My ceph version is > ceph version 0.39 (commit:321ecdaba2ceeddb0789d8f4b7180a8ea5785d83) > > When I try to map a long name rbd image to device, it would hang for > long time. > > For example: > sudo rbd map iqn.2012-01.com.sample:storage.ttttttt --secret > /etc/ceph/secretfile sudo rbd map iqn.2012-01.com.sample:storage.abcdef > --secret /etc/ceph/secretfile The limit was increased in this commit: commit 224736d9113ab4a7cf3f05c05377492bd99b4b02 Author: Stratos Psomadakis Date: Thu Nov 10 15:45:37 2011 +0200 libceph: Allocate larger oid buffer in request msgs ceph_osd_request struct allocates a 40-byte buffer for object names. RBD image names can be up to 96 chars long (100 with the .rbd suffix), which results in the object name for the image being truncated, and a subsequent map failure. Increase the oid buffer in request messages, in order to avoid the truncation. There is still an outstanding issue with the limits in the kernel not being enforced by userspace tools though (http://tracker.newdream.net/issues/1701). > > It would not hang at every long name image, however, it only happens > when the image name is very long. > > Is it a known issue? > > Thanks!