Date: Thu, 21 Mar 2002 11:56:35 -0800 From: Jeff Behl <jeff@expertcity.com> To: freebsd-questions@freebsd.org Subject: NFS woes: "got bad cookie vp" Message-ID: <3C9A3AF3.4090602@expertcity.com>
next in thread | raw e-mail | index | archive | help
FreeBSD hormel.snv 4.5-RELEASE-p2 FreeBSD 4.5-RELEASE-p2 #0: Fri Mar 8 15:11:46 PST 2002 root@hormel.snv:/usr/src/sys/compile/hormel i386 This machine is used as the relay to send out to our user mailing lists, one of which is 300k+ long. I have exim setup to use our netapp (via nfs mount) as the spool directory simply because I figured it would be a lot faster than using the local disk when the bulk-mailer starts sending 300k+ emails to it as fast as it can (the idea is to accept the mail as quickly as possible to free up the sending machine). Things work great (it accepted around 80k+ emails in an hour) for the first hour, but after that performance steadily declines as more queue runner processes start sending the mail out. It looks like nfs problem were the cause of the decline as actual cpu usage wasn't that high. The problem doesn't seem to be in writing out the temporary files to the netapp, but rather having multiple queue runner processes going through and accessing/deleting individual files (emails) for delivery. Eventually, things slow to a crawl with the following being constantly displayed: Mar 21 11:52:29 hormel /kernel: got bad cookie vp 0xd7e27140 bp 0xcc600270 Mar 21 11:52:34 hormel /kernel: got bad cookie vp 0xd7e27140 bp 0xcc51f780 Mar 21 11:52:34 hormel /kernel: got bad cookie vp 0xd7e27140 bp 0xcc51f780 Mar 21 11:52:52 hormel /kernel: got bad cookie vp 0xd7e48900 bp 0xcc540b38 Mar 21 11:52:52 hormel /kernel: got bad cookie vp 0xd7e48900 bp 0xcc540b38 Is there anything I can do? thanks. jeff To Unsubscribe: send mail to majordomo@FreeBSD.org with "unsubscribe freebsd-questions" in the body of the message
Want to link to this message? Use this URL: <https://mail-archive.FreeBSD.org/cgi/mid.cgi?3C9A3AF3.4090602>