Is swap on ZFS supposed to work?

2010-04-27 Thread Andrew Gallatin
I have a machine where I enabled swap on ZFS somewhat accidentally (really wanted to dump to ZFS, but that didn't pan out) and then forgot about it. A month later, under a heavy load (make -j8 buildworld) it hung with: swap_pager: indefinite wait buffer: bufobj: 0, blkno: 289218, size: 4096 swap

Re: wi0 and mtu setting [bad idea]

2003-01-06 Thread Andrew Gallatin
M. Warner Losh writes: > > Well, this converstation needs to take place on the net@ list. In > -current, at least, if_ethersubr.c prevents 'ethernet' drivers from > setting the mtu higher. There's some experimental patches that make It doesn't prevent anything. It just sets the mtu to a

Re: AW: mbuf clusters behavior (NMBCLUSTERS)

2002-07-16 Thread Andrew Gallatin
Alexander Maret writes: > > -Ursprungliche Nachricht- > > Von: [EMAIL PROTECTED] [mailto:[EMAIL PROTECTED]] > > > > When I copy a 500M file (like iso image) from the workstation to the > > server, the server starts to emit: > > > > Jul 12 09:28:54 nile /kernel: All mbuf clusters

RE: number mbufs / cluster

2002-07-01 Thread Andrew Gallatin
Don Bowman writes: > > Doug White wrote: > > > On Sat, 29 Jun 2002, Don Bowman wrote: > > > > > > > > I have an application where there is lots of connections, > > > but packet sizes aren't that large. I'm running into the > > > problem where I need an enormous amount of clusters, >

Re: 4.6-RC system hangs (fxp0, smp, sym)

2002-05-21 Thread Andrew Gallatin
Pete French writes: > > > [ I'd say this is the time for developers to speak up ;-) ] > > Assuming we have giiven them enough information - sadly random freezeups > with no errormessages and no real repeatabaility are very hard to debug. It is obvious that the disk controller is no long

Re: Moused broken on alpha

2002-04-18 Thread Andrew Gallatin
Dermot McNally writes: > Folks, > > Moused refuses to build on alpha because the symbol MOUSE_PROTO_JOGDIAL > (defined in the Intel version of mouse.h) isn't defined. Defining it > fixes the problem, but is this the correct fix? Nobody else seems to > have been bitten by this one... The M

Re: NeoMagic 256Z: pcm0: play interrupt timeout, channel dead

2001-09-02 Thread Andrew Gallatin
Gregory Neil Shapiro writes: > imp> Does it work if you disable pcic in your kernel? I hope the answer is > imp> "no change" > > Since I have the same sound driver and it is having the same problem, I am > "happy" (though that is the wrong word) to report that removing pcic from > the ker

Re: NeoMagic 256Z: pcm0: play interrupt timeout, channel dead

2001-09-02 Thread Andrew Gallatin
Warner Losh writes: > In message <[EMAIL PROTECTED]> Andrew Gallatin >writes: > : And I've appended a dmesg. > : > : Thanks for any help.. > > Does it work if you disable pcic in your kernel? I hope the answer is > "no change" > >

Re: Gnome with 4.4 PRERELEASE

2001-08-05 Thread Andrew Gallatin
Jake Bishop writes: > Hello, > I installed 4.3 RELEASE and cvsuped to 4.4 PRERELEASE using the > stable-supfile.I am build Gnome from the port x11/gnome. When I ran make > install it stopped with an error about gnome-guile-0.20.There is not > alot of info about the actual error though. If y

Re: SMP changes and breaking kld object module compatibility

2000-04-24 Thread Andrew Gallatin
g a module flag day, I'm thinking that it might be a good time to MFC Jonathan Lemon's checksum offloading code. Doing this would require changing MSIZE to 256, which in turn would require recompiling any module using mbufs (all NICs, network filesystems, etc). Cheers, Drew --