From mboxrd@z Thu Jan 1 00:00:00 1970 Date: Thu, 11 Oct 2018 10:55:04 -0500 From: David Teigland Message-ID: <20181011155504.GA17418@redhat.com> References: <20181010190857.GB10633@redhat.com> MIME-Version: 1.0 Content-Disposition: inline In-Reply-To: Subject: Re: [linux-lvm] [lvmlockd] lvm command hung with sanlock log "ballot 3 abort1 larger lver in bk..." Reply-To: LVM general discussion and development List-Id: LVM general discussion and development List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , List-Id: Content-Type: text/plain; charset="us-ascii" Content-Transfer-Encoding: 7bit To: Damon Wang Cc: linux-lvm@redhat.com On Thu, Oct 11, 2018 at 09:03:01PM +0800, Damon Wang wrote: > This is because I regenerate a host id when the host join a new > lockspace OK, I guess there's something about your setup that makes it difficult to assign static host_id's. > 2018-10-09 20:49:16 4854717 [29802]: r2320 write_host_block host_id > 19 flags 1 gen 1 dblock 29802:510:140245418403952:140245440585933:140245418403840:4:RELEASED. Perfect, that shows exactly where the bug is. There's a case where it has missed saving the latest dblock values, so random values are being copied back to the dblock instead. I'll push out a patch once I've had a chance to test it. > does the "paxos_release skip write last lver" is abnormal? Not very unusual when you have lock contention (multiple hosts trying to acquire the same lock concurrently.) Thanks, Dave