From owner-freebsd-bugs@FreeBSD.ORG Fri Jan 2 19:58:30 2015 Return-Path: Delivered-To: freebsd-bugs@FreeBSD.org Received: from mx1.freebsd.org (mx1.freebsd.org [IPv6:2001:1900:2254:206a::19:1]) (using TLSv1.2 with cipher AECDH-AES256-SHA (256/256 bits)) (No client certificate requested) by hub.freebsd.org (Postfix) with ESMTPS id 1A08CB4C for ; Fri, 2 Jan 2015 19:58:30 +0000 (UTC) Received: from kenobi.freebsd.org (kenobi.freebsd.org [IPv6:2001:1900:2254:206a::16:76]) (using TLSv1.2 with cipher ECDHE-RSA-AES256-GCM-SHA384 (256/256 bits)) (Client did not present a certificate) by mx1.freebsd.org (Postfix) with ESMTPS id 01FEE293B for ; Fri, 2 Jan 2015 19:58:30 +0000 (UTC) Received: from bugs.freebsd.org ([127.0.1.118]) by kenobi.freebsd.org (8.14.9/8.14.9) with ESMTP id t02JwT8W041528 for ; Fri, 2 Jan 2015 19:58:29 GMT (envelope-from bugzilla-noreply@freebsd.org) From: bugzilla-noreply@freebsd.org To: freebsd-bugs@FreeBSD.org Subject: [Bug 194513] zfs recv hangs in state kmem arena Date: Fri, 02 Jan 2015 19:58:28 +0000 X-Bugzilla-Reason: AssignedTo X-Bugzilla-Type: changed X-Bugzilla-Watch-Reason: None X-Bugzilla-Product: Base System X-Bugzilla-Component: kern X-Bugzilla-Version: 10.0-STABLE X-Bugzilla-Keywords: X-Bugzilla-Severity: Affects Some People X-Bugzilla-Who: alc@FreeBSD.org X-Bugzilla-Status: In Progress X-Bugzilla-Priority: --- X-Bugzilla-Assigned-To: freebsd-bugs@FreeBSD.org X-Bugzilla-Target-Milestone: --- X-Bugzilla-Flags: X-Bugzilla-Changed-Fields: Message-ID: In-Reply-To: References: Content-Type: text/plain; charset="UTF-8" Content-Transfer-Encoding: 7bit X-Bugzilla-URL: https://bugs.freebsd.org/bugzilla/ Auto-Submitted: auto-generated MIME-Version: 1.0 X-BeenThere: freebsd-bugs@freebsd.org X-Mailman-Version: 2.1.18-1 Precedence: list List-Id: Bug reports List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , X-List-Received-Date: Fri, 02 Jan 2015 19:58:30 -0000 https://bugs.freebsd.org/bugzilla/show_bug.cgi?id=194513 --- Comment #8 from Alan Cox --- I suspect that there are two distinct issues at play here. I believe that Andriy has correctly identified a synchronization error in kmem_reclaim(). However, the size of the failed allocations, 10 MB, is also interesting. While vmem's allocation strategy is better at reducing external fragmentation than the simplistic "first-fit" strategy that we previously used, it is not immune to external fragmentation. Thus, kmem arena may not have 10 MB of contiguous virtual address space even though there is plenty of free space. I suggest that you try increasing the kmem arena size to 1.25 or even 1.5 times the physical memory size and report back on the effect. -- You are receiving this mail because: You are the assignee for the bug.