Hi Ted, On Tue, 2009-06-09 at 13:01 -0400, Ted Unangst wrote: > On Tue, Jun 9, 2009 at 2:03 PM, Bill Maas <b...@stsx.org> wrote: > > I posted a message earlier about a kernel panic occurring when I > > accessed a file on some of my ext3 fses. I've also been having trouble > > with r/w extfs entries in fstab. At boot time I'm dropped to a shell > > because fsck thinks the fs is unclean, even when "the other side" says > > it's clean. > > ext3 is marked dirty because the journal hasn't been played back. You > have to convert it to ext2 in linux before mounting in openbsd. >
That makes sense, I guess. And it does keep the unclean fs messages away - not the bad ref count panic however. The docs could be a bit more explicit about the lack of support for ext3 journaling. And in reply to the various "why would you want to do that?"'s I encountered while searching for the issue: very witty, but ext2 happens to be a widely supported fs, which makes it a good candidate for shared data on multiboot systems (FAT16/32? - can't be serious...!). Moreover, so far OpenBSD has proven to have excellent support for ext2, apart from that single issue. FFS support from Linux on the other hand, is C.R.A.P. Thanks, Bill
