Date: Sun, 10 Feb 2002 22:47:14 +0200 From: Valentin Nechayev <netch@iv.nn.kiev.ua> To: Matthew Dillon <dillon@apollo.backplane.com> Cc: Mike Silbersack <silby@silby.com>, "David W. Chapman Jr." <dwcjr@inethouston.net>, stable@FreeBSD.ORG Subject: Re: cvs commit: src/sys/ufs/ffs ffs_softdep.c Message-ID: <20020210224714.A269@iv.nn.kiev.ua> In-Reply-To: <200202101911.g1AJBOk30917@apollo.backplane.com>; from dillon@apollo.backplane.com on Sun, Feb 10, 2002 at 11:11:24AM -0800 References: <20020210113348.P28078-100000@patrocles.silby.com> <200202101911.g1AJBOk30917@apollo.backplane.com>
next in thread | previous in thread | raw e-mail | index | archive | help
Sun, Feb 10, 2002 at 11:11:24, dillon (Matthew Dillon) wrote about "Re: cvs commit: src/sys/ufs/ffs ffs_softdep.c": > On the second issue - long disk activity times after a 'syncing disks... > done'. This can occur if there are a lot of dirty VM pages in the > system belonging to files that need to be written out. This can take an > especially long time if a large file was created with ftruncate() > and written via mmap() instead of write(), because the file can become > severely fragmented. Three cases when this happened: two were deleting of too deep tree (first with this activity >1 minute, second with ~20 secs), one was immediately after exiting mutt with huge folder (cvs-all@ inbox;)) in maildir format, with ~50 deleted letters; this one was ~10 secs. Cel-800, 256M, IC35L040AVER07-0 at ATA100 with tags, 4.5-release. Moment which can be important: vfs.vmiodirenable=1. I didn't test it yet with vfs.vmiodirenable=0, it will be tried now. /netch To Unsubscribe: send mail to majordomo@FreeBSD.org with "unsubscribe freebsd-stable" in the body of the message
Want to link to this message? Use this URL: <https://mail-archive.FreeBSD.org/cgi/mid.cgi?20020210224714.A269>