From owner-freebsd-fs@freebsd.org Thu Nov 19 12:30:24 2015 Return-Path: Delivered-To: freebsd-fs@mailman.ysv.freebsd.org Received: from mx1.freebsd.org (mx1.freebsd.org [IPv6:2001:1900:2254:206a::19:1]) by mailman.ysv.freebsd.org (Postfix) with ESMTP id 61D01A33999 for ; Thu, 19 Nov 2015 12:30:24 +0000 (UTC) (envelope-from supportme@ukr.net) Received: from mbob.nabble.com (mbob.nabble.com [162.253.133.15]) by mx1.freebsd.org (Postfix) with ESMTP id 50FD21D2A for ; Thu, 19 Nov 2015 12:30:23 +0000 (UTC) (envelope-from supportme@ukr.net) Received: from msam.nabble.com (unknown [162.253.133.85]) by mbob.nabble.com (Postfix) with ESMTP id 2223B1A6BDBB for ; Thu, 19 Nov 2015 04:17:48 -0800 (PST) Date: Thu, 19 Nov 2015 05:30:16 -0700 (MST) From: Dmitriy Makarov To: freebsd-fs@freebsd.org Message-ID: <1447936216089-6054111.post@n5.nabble.com> In-Reply-To: <564D9930.1080509@bytecamp.net> References: <564D9930.1080509@bytecamp.net> Subject: Re: filesystem deadlock, process in vodead state MIME-Version: 1.0 Content-Type: text/plain; charset=us-ascii Content-Transfer-Encoding: 7bit X-BeenThere: freebsd-fs@freebsd.org X-Mailman-Version: 2.1.20 Precedence: list List-Id: Filesystems List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , X-List-Received-Date: Thu, 19 Nov 2015 12:30:24 -0000 Hi, we has exactly the same issue on our two boxes, running CURRENT. But in our case stuck is on read one exactly file: ls $file - works fine, cat $file - hang. The file is on ZFS pool, stripe of mirrors too. The first box: FreeBSD 11.0-CURRENT #0 r290331 (uptime 15 days) [box92:~]$ ps auxwww|grep cat root 21344 0,0 0,0 8256 2012 6- D+J 13:52 0:00,00 cat .main.lock [box92:~]# procstat -kk 21344 PID TID COMM TDNAME KSTACK 21344 103410 cat - mi_switch+0xd8 sleepq_wait+0x3a _sleep+0x2df vnode_create_vobject+0x131 zfs_freebsd_open+0xf6 VOP_OPEN_APV+0xa0 vn_open_vnode+0x26a vn_open_cred+0x33e kern_openat+0x25f amd64_syscall+0x50b Xfast_syscall+0xfb [box92:~]# procstat -t 21344 PID TID COMM TDNAME CPU PRI STATE WCHAN 21344 103410 cat - -1 120 sleep vodead [box92:~]# procstat -S 21344 PID TID COMM TDNAME CPU CSID CPU MASK 21344 103410 cat - -1 6 0-11 [box92:~]# procstat -H 21344 PID PPID PGID SID TSID THR LOGIN WCHAN EMUL COMM 21344 84508 21344 9350 9350 1 endo vodead FreeBSD ELF64 cat [box92:~]# procstat -f 21344 PID COMM FD T V FLAGS REF OFFSET PRO NAME 21344 cat text v r r------- - - - /var/jail64/jail1/bin/cat 21344 cat cwd v d r------- - - - /var/jail64/jail1/var/mail/path 21344 cat root v d r------- - - - /var/jail64/jail1 21344 cat jail v d r------- - - - /var/jail64/jail1 21344 cat 0 v x rwa----- 18 635287 - - 21344 cat 1 v x rwa----- 18 635287 - - 21344 cat 2 v x rwa----- 18 635287 - - The second box running FreeBSD 11.0-CURRENT #0 r287745 (uptime 51 days) I'm not familiar with debugging tools, but ready to help if somebody interesting. -- View this message in context: http://freebsd.1045724.n5.nabble.com/filesystem-deadlock-process-in-vodead-state-tp6054087p6054111.html Sent from the freebsd-fs mailing list archive at Nabble.com.