Date: Sat, 4 Sep 2004 00:42:17 -0300 (ADT) From: "Marc G. Fournier" <scrappy@hub.org> To: Giorgos Keramidas <keramida@linux.gr> Cc: freebsd-current@freebsd.org Subject: Re: what is fsck's "slowdown"? Message-ID: <20040904004104.C812@ganymede.hub.org> In-Reply-To: <20040903215054.GD1199@gothmog.gr> References: <20040903175434.A812@ganymede.hub.org> <20040903211427.GB1199@gothmog.gr><20040903215054.GD1199@gothmog.gr>
next in thread | previous in thread | raw e-mail | index | archive | help
On Sat, 4 Sep 2004, Giorgos Keramidas wrote: > On 2004-09-03 17:35, Chris Laverdure <dashevil@sympatico.ca> wrote: >> On Fri, 2004-09-03 at 21:14, Giorgos Keramidas wrote: >>> (Regarding "parallelization" of fsck by spawning many instances of >>> fsck for parts of the same partition...) >>> >>> My intuition says that if metadata of the first part of the disk references >>> data residing on the second part synchronization and locking would probably >>> be a bit difficult; actually very difficult. >> >> My intuition tells me that it would be a much better solution to run >> multiple fsck's concurrently. What harm could there be in fscking (num >> of processors) partitions at the same time? > > AFAIK, this is exactly what "background fsck" does in 5.X :-) fsck -p in 4.x does this also .. but, when there is only one large file system, and 4 or 5 smaller ones, those 4 or 5 smaller ones don't take long t do ... in my case, that one large one just took 12hrs to complete, on a system where that one fsck was the only thing running :( I don't believe that moving to 5.x's bkgd fsck will speed that up any, and, in fact, would probably slow it down since it will be completing with other processes ... ---- Marc G. Fournier Hub.Org Networking Services (http://www.hub.org) Email: scrappy@hub.org Yahoo!: yscrappy ICQ: 7615664
Want to link to this message? Use this URL: <https://mail-archive.FreeBSD.org/cgi/mid.cgi?20040904004104.C812>