Date: Tue, 28 Feb 2012 04:49:38 GMT From: "Eugene M. Zheganin" <eugene@zhegan.in> To: freebsd-gnats-submit@FreeBSD.org Subject: misc/165521: livelock on 1 Gig of RAM with zfs when 310.locate is run Message-ID: <201202280449.q1S4ncG2057766@red.freebsd.org> Resent-Message-ID: <201202280450.q1S4o87i088374@freefall.freebsd.org>
next in thread | raw e-mail | index | archive | help
>Number: 165521 >Category: misc >Synopsis: livelock on 1 Gig of RAM with zfs when 310.locate is run >Confidential: no >Severity: serious >Priority: low >Responsible: freebsd-bugs >State: open >Quarter: >Keywords: >Date-Required: >Class: sw-bug >Submitter-Id: current-users >Arrival-Date: Tue Feb 28 04:50:08 UTC 2012 >Closed-Date: >Last-Modified: >Originator: Eugene M. Zheganin >Release: 8.2-RELEASE >Organization: RealService LLC >Environment: FreeBSD moscow-omega 8.2-RELEASE FreeBSD 8.2-RELEASE #2: Wed Jan 25 11:30:58 YEKT 2012 emz@moscow-omega:/usr/obj/usr/src/sys/MOSCOW amd64 >Description: Every saturday this server hangs around 4:15 in the morning. 4:15 in the morning is the time when periodic weekly is run. After some investigations it looks like 310.locate is the critical script. This is reproduceable. Simply launching this script also makes server hang. Other servers with similar hardware configuraton and use also hang. loader.conf: zfs_load="YES" vfs.root.mountfrom="zfs:zfsroot" ng_iface_load="YES" ng_ether_load="YES" vm.kmem_size="330M" vm.kmem_size_max="330M" vfs.zfs.arc_max="30M" Console is responsive, but the machine doesn't allow to log in and doesn't respond to network. >How-To-Repeat: Get a FreeBSD, use zfs for system, use 1 Gig of RAM, run 310.locate from weekly set of periodic scripts. >Fix: Turn of periodic runs. Add more RAM (problem disappears on 4 Gigs of RAM with the same config set). >Release-Note: >Audit-Trail: >Unformatted:
Want to link to this message? Use this URL: <https://mail-archive.FreeBSD.org/cgi/mid.cgi?201202280449.q1S4ncG2057766>