From owner-freebsd-hackers Fri Jun 15 21:54: 9 2001 Delivered-To: freebsd-hackers@freebsd.org Received: from earth.backplane.com (earth-nat-cw.backplane.com [208.161.114.67]) by hub.freebsd.org (Postfix) with ESMTP id E975437B403 for ; Fri, 15 Jun 2001 21:53:55 -0700 (PDT) (envelope-from dillon@earth.backplane.com) Received: (from dillon@localhost) by earth.backplane.com (8.11.3/8.11.2) id f5G4rtl57912; Fri, 15 Jun 2001 21:53:55 -0700 (PDT) (envelope-from dillon) Date: Fri, 15 Jun 2001 21:53:55 -0700 (PDT) From: Matt Dillon Message-Id: <200106160453.f5G4rtl57912@earth.backplane.com> To: Dan Phoenix Cc: freebsd-hackers@FreeBSD.ORG Subject: Re: high cpu spikes References: Sender: owner-freebsd-hackers@FreeBSD.ORG Precedence: bulk List-ID: List-Archive: (Web Archive) List-Help: (List Instructions) List-Subscribe: List-Unsubscribe: X-Loop: FreeBSD.ORG : : :nfs_getpages: error 70 :vm_fault: pager read error, pid 72424 (httpd) : : :I get occasional cpu spikes for like 10-20 sec occasionally :wondering what this is from.....i am thinking nfs timeout maybe.... :i am ready got maxusers set to 500. : :-- :Dan error 70 is 'stale NFS file handle', which means that the web server running on the NFS client tried to access a page from a file that has been ripped out from under it (probably by the NFS server itself since client deletions usually use the NFS rename hack). The only safe solution is to figure out what is deleting or overwriting the file and make it rename the file to something else instead, then have a cron job that goes in every so often and deletes the renamed files over X amount of time old. -Matt To Unsubscribe: send mail to majordomo@FreeBSD.org with "unsubscribe freebsd-hackers" in the body of the message