Date: Fri, 8 Feb 2013 06:30:01 GMT From: Andriy Gapon <avg@FreeBSD.org> To: freebsd-bugs@FreeBSD.org Subject: Re: kern/175950: Possible deadlock in zfs after long uptime Message-ID: <201302080630.r186U1ec006100@freefall.freebsd.org>
index | next in thread | raw e-mail
The following reply was made to PR kern/175950; it has been noted by GNATS. From: Andriy Gapon <avg@FreeBSD.org> To: bug-followup@FreeBSD.org, pascal.guitierrez@gmail.com Cc: Subject: Re: kern/175950: Possible deadlock in zfs after long uptime Date: Fri, 08 Feb 2013 08:26:03 +0200 Thank you for the report. You seem to have run into a problem that is triggered by either low available physical memory or low KVA. r242859 on stable/8 branch may reduce chances of the problem. Unfortunately, there is no complete resolution at the moment. If you are interested, some technical details could be found here: http://article.gmane.org/gmane.os.freebsd.stable/84981 Could you please also provide output of the following sysctls (for completeness sake): vm.stats vm.kmem_size vm.kmem_map_size vm.kmem_map_free P.S. The bad thread in your report: 1963 100308 nfsd nfsd: service mi_switch+0x176 sleepq_wait+0x42 _sleep+0x317 arc_lowmem+0x77 kmem_malloc+0xc1 uma_large_malloc+0x4a malloc+0xd7 arc_get_data_buf+0xb5 arc_read_nolock+0x1ec arc_read+0x93 dbuf_read+0x452 dmu_tx_check_ioerr+0x9a dmu_tx_count_write+0x29c dmu_tx_hold_write+0x4a zfs_freebsd_write+0x372 VOP_WRITE_APV+0xb2 nfsrv_write+0x969 nfssvc_program+0x1a6 -- Andriy Gaponhome | help
Want to link to this message? Use this
URL: <https://mail-archive.FreeBSD.org/cgi/mid.cgi?201302080630.r186U1ec006100>