Date: Mon, 3 May 2004 14:25:33 -0400 From: John Baldwin <jhb@FreeBSD.org> To: freebsd-current@FreeBSD.org Cc: Andrew Gallatin <gallatin@cs.duke.edu> Subject: Re: partial dumps (was Re: Change default dumpdir to /usr/crash?) Message-ID: <200405031425.33399.jhb@FreeBSD.org> In-Reply-To: <16534.21617.310294.982202@grasshopper.cs.duke.edu> References: <200404301403.50634.past@noc.ntua.gr> <20040430211948.GC85783@dragon.nuxi.com> <16534.21617.310294.982202@grasshopper.cs.duke.edu>
next in thread | previous in thread | raw e-mail | index | archive | help
On Monday 03 May 2004 10:17 am, Andrew Gallatin wrote: > > On Fri, Apr 30, 2004 at 02:30:40PM +0200, Thomas Quinot wrote: > > > The proper fix would probably be to change the default partitioning > > > scheme, not to move the crash dumps. I think one property we try to > > > guarantee is that /usr be mountable read-only through NFS for a > > > cluster of workstations, whereas /var is always mounted read-write, > > > for its purpose is to contain files whose contents *vary* over time. > > Another good idea (perhaps in combination with a larger /var) is to > accept and port to -current the Duke "partial dump" patches. These > patches allow the user to optionally dump just the kernel virtual > address space. This results in dumps that are generally less than > 100MB, rather than multiple gigs. > > In nearly all cases, only the kernel address space is needed to > interpret a dump. From what I've seen, this is what Solaris, AIX, and > Tru64 do by default. > > Porting to -current will be non-trivial because of the dump changes > between 4.x and 5.x. If I was to do this, is there any chance that > it could get into the tree? This sounds like an excellent idea! -- John Baldwin <jhb@FreeBSD.org> <>< http://www.FreeBSD.org/~jhb/ "Power Users Use the Power to Serve" = http://www.FreeBSD.org
Want to link to this message? Use this URL: <https://mail-archive.FreeBSD.org/cgi/mid.cgi?200405031425.33399.jhb>