Date: Fri, 12 Oct 2001 15:46:44 -0500 (CDT) From: Mike Silbersack <silby@silby.com> To: Matt Dillon <dillon@earth.backplane.com> Cc: <cvs-committers@FreeBSD.ORG>, <cvs-all@FreeBSD.ORG> Subject: Re: cvs commit: src/sys/vm vnode_pager.c Message-ID: <20011012154433.W29945-100000@achilles.silby.com> In-Reply-To: <200110122026.f9CKQMS35969@earth.backplane.com>
next in thread | previous in thread | raw e-mail | index | archive | help
On Fri, 12 Oct 2001, Matt Dillon wrote: > :> Finally fix the VM bug where a file whos EOF occurs in the middle of a page > :> would sometimes prevent a dirty page from being cleaned, even when synced, > :> resulting in the dirty page being re-flushed to disk every 30-60 seconds or > :> so, forever. The problem is that when the filesystem flushes a page to > : > :How commonly did this occur? > : > :Mike "Silby" Silbersack > > In recent years the use of mmap() has increased hugely. For example, > 'cp', 'install', and 'tail' now use it, as does 'samba' and 'apache'. > I only know of one or two 'severe' cases. 99% of installations will > not be effected because they simply do not access a large enough number > of files or do not access files in a way that hits the bug for it to > become a problem. For example, my home system has been up 69 days > and not a single page is in this state. One of our production severs > with an uptime of 4 days has one page in this state. Another production > server up 185 days has 0 pages in this state. Ah. Ok, I was just wondering if this was a contributing factor to any of the various heavy load -> crash/freeze problems people occasionally report. Sounds like it probably isn't. Nonetheless, cool that it's fixed. :) Mike "Silby" Silbersack To Unsubscribe: send mail to majordomo@FreeBSD.org with "unsubscribe cvs-all" in the body of the message
Want to link to this message? Use this URL: <https://mail-archive.FreeBSD.org/cgi/mid.cgi?20011012154433.W29945-100000>