From owner-freebsd-current Fri Feb 5 12:44:04 1999 Return-Path: Received: (from majordom@localhost) by hub.freebsd.org (8.8.8/8.8.8) id MAA27893 for freebsd-current-outgoing; Fri, 5 Feb 1999 12:44:04 -0800 (PST) (envelope-from owner-freebsd-current@FreeBSD.ORG) Received: from apollo.backplane.com (apollo.backplane.com [209.157.86.2]) by hub.freebsd.org (8.8.8/8.8.8) with ESMTP id MAA27884 for ; Fri, 5 Feb 1999 12:44:02 -0800 (PST) (envelope-from dillon@apollo.backplane.com) Received: (from dillon@localhost) by apollo.backplane.com (8.9.2/8.9.1) id MAA98966; Fri, 5 Feb 1999 12:22:44 -0800 (PST) (envelope-from dillon) Date: Fri, 5 Feb 1999 12:22:44 -0800 (PST) From: Matthew Dillon Message-Id: <199902052022.MAA98966@apollo.backplane.com> To: current@FreeBSD.ORG Subject: More on Seeing NFS saturation 'loop' when installworld'ing to NFS / and /usr Sender: owner-freebsd-current@FreeBSD.ORG Precedence: bulk X-Loop: FreeBSD.ORG Here's more. I seem to be getting stale NFS handle replies on the NFS-V3 commits and writes. I am guessing that the problem is due, somehow, to install's writing or renaming over of the original file, probably after stat'ing the original file. doing a mount -u -o acdirmin=0,acdirmax=0,acregmin=0,acregmax=0 ... on my / and /usr NFS mounts does not fix the problem. I am not ripping these files up from under the client .. the client is the only machine messing with these files. -Matt : This is very odd. This is the approximate backtrace that I get : when I throw my test machine into DDB: : : --- interrupt, ... : nfs_* routines.... : cluster_wbuild : vfs_bio_awrite : flushdirtybuffers : bdwrite : nfs_write : vn_write : write : syscall : : What is happening is that I am doing a 'make installworld' on my : test machine with / and /usr NFS V3 mounted R+W. : : The install goes well, but four times so far an 'install' command has : gotten 'stuck' in 'R'un state and the network has gone into saturation. : It seems to be repreating the same NFS I/O over the network over and : over again as far as I can tell. It is very odd. : : The network stays in saturation ( at 8 MBytes/sec ) until I kill -STOP :... : nfs_* routines.... : cluster_wbuild : vfs_bio_awrite : flushdirtybuffers : bdwrite ( I think ) : ffs_update : : It is very odd. I don't suppose very many people try to make install :... 12:11:51.065802 209.157.86.12.287646691 > 209.157.86.2.2049: 132 commit fh 0,7/2006718018 65536 bytes @ 0 12:11:51.066276 209.157.86.12.287646692 > 209.157.86.2.2049: 1472 write fh 0,7/2006718018 8192 bytes @ 0 (frag 54615:1480@0+) 12:11:51.066417 209.157.86.12 > 209.157.86.2: (frag 54615:1480@1480+) 12:11:51.066558 209.157.86.12 > 209.157.86.2: (frag 54615:1480@2960+) 12:11:51.066698 209.157.86.12 > 209.157.86.2: (frag 54615:1480@4440+) 12:11:51.066839 209.157.86.12 > 209.157.86.2: (frag 54615:1480@5920+) 12:11:51.066930 209.157.86.12 > 209.157.86.2: (frag 54615:940@7400) 12:11:51.067842 209.157.86.2.2049 > 209.157.86.12.287646691: reply ok 36 commit ERROR: Stale NFS file handle 12:11:51.067938 209.157.86.2.2049 > 209.157.86.12.287646692: reply ok 36 write ERROR: Stale NFS file handle 12:11:51.068038 209.157.86.12.287646693 > 209.157.86.2.2049: 132 commit fh 0,7/2006718018 16384 bytes @ 65536 12:11:51.068185 209.157.86.2.2049 > 209.157.86.12.287646693: reply ok 36 commit ERROR: Stale NFS file handle 12:11:51.068512 209.157.86.12.287646694 > 209.157.86.2.2049: 1472 write fh 0,7/2006718018 8192 bytes @ 0 (frag 54617:1480@0+) 12:11:51.068654 209.157.86.12 > 209.157.86.2: (frag 54617:1480@1480+) 12:11:51.068796 209.157.86.12 > 209.157.86.2: (frag 54617:1480@2960+) 12:11:51.068932 209.157.86.12 > 209.157.86.2: (frag 54617:1480@4440+) 12:11:51.069072 209.157.86.12 > 209.157.86.2: (frag 54617:1480@5920+) 12:11:51.069161 209.157.86.12 > 209.157.86.2: (frag 54617:940@7400) 12:11:51.069188 209.157.86.12.287646695 > 209.157.86.2.2049: 132 commit fh 0,7/2006718018 57344 bytes @ 0 12:11:51.069374 209.157.86.12.287646696 > 209.157.86.2.2049: 1472 write fh 0,7/2006718018 8192 bytes @ 65536 (frag 54619:1480@0+) 12:11:51.069512 209.157.86.12 > 209.157.86.2: (frag 54619:1480@1480+) 12:11:51.069652 209.157.86.12 > 209.157.86.2: (frag 54619:1480@2960+) 12:11:51.069688 209.157.86.2.2049 > 209.157.86.12.287646694: reply ok 36 write ERROR: Stale NFS file handle 12:11:51.069793 209.157.86.12 > 209.157.86.2: (frag 54619:1480@4440+) 12:11:51.069832 209.157.86.2.2049 > 209.157.86.12.287646695: reply ok 36 commit ERROR: Stale NFS file handle 12:11:51.069940 209.157.86.12 > 209.157.86.2: (frag 54619:1480@5920+) 12:11:51.070030 209.157.86.12 > 209.157.86.2: (frag 54619:940@7400) 12:11:51.070215 209.157.86.12.287646697 > 209.157.86.2.2049: 132 commit fh 0,7/2006718018 24576 bytes @ 0 12:11:51.070603 209.157.86.12.287646698 > 209.157.86.2.2049: 1472 write fh 0,7/2006718018 8192 bytes @ 0 (frag 54621:1480@0+) 12:11:51.070743 209.157.86.12 > 209.157.86.2: (frag 54621:1480@1480+) 12:11:51.070777 209.157.86.2.2049 > 209.157.86.12.287646696: reply ok 36 write ERROR: Stale NFS file handle 12:11:51.070889 209.157.86.12 > 209.157.86.2: (frag 54621:1480@2960+) 12:11:51.070922 209.157.86.2.2049 > 209.157.86.12.287646697: reply ok 36 commit ERROR: Stale NFS file handle 12:11:51.071034 209.157.86.12 > 209.157.86.2: (frag 54621:1480@4440+) 12:11:51.071176 209.157.86.12 > 209.157.86.2: (frag 54621:1480@5920+) 12:11:51.071267 209.157.86.12 > 209.157.86.2: (frag 54621:940@7400) 12:11:51.071456 209.157.86.12.287646699 > 209.157.86.2.2049: 132 commit fh 0,7/2006718018 49152 bytes @ 0 12:11:51.071729 209.157.86.12.287646700 > 209.157.86.2.2049: 1472 write fh 0,7/2006718018 8192 bytes @ 0 (frag 54623:1480@0+) 12:11:51.071871 209.157.86.12 > 209.157.86.2: (frag 54623:1480@1480+) To Unsubscribe: send mail to majordomo@FreeBSD.org with "unsubscribe freebsd-current" in the body of the message