The collected works of badapple

I've been testing out a new [home, not work] fileserver that has been put together from spare parts and after a couple of weeks of constant battering it's passed burn-in and is running nicely. It's nothing special - just a basic x86 with an Adaptec 2400 raid card and 4 x 250Gb disks in raid 5. The XFS filesystem was built with a basic 'mkfs.xfs /dev/whatever' command and although performance is already better than the LVM/ext3 system it's replacing I'd like to optimize it before this box is promoted from testing to production and I start migrating all my data to it.

The man pages, SGI site, google, etc have got loads of information which I've been wading through: sunit/su, swidth/sw, log and inode options, etc for the filesystem and the raid card obviously supports various different stripe width, block sizes, etc as well. To save me spending the rest of my life rebuilding arrays and doing I/O benchmarks any advice would be really appreciated - I already spend far too much time doing stuff to computers rather than with them.

The machine runs debian and is a general purpose file server which makes things awkward - it will handle everything from rotating rsync backups of other machines to storing mp3s and getting 200Gb+ forensic disk images dumped on it so I can't really optimize in favour of small or large files but it doesn't have to do anything clever like real-time. None the less, I'm sure that I can get better performance than just using the default options.

So, can any XFS experts point me in the right direction? What values should I be setting the physical raid stripe/block and XFS filesystem variables to? Should I align the values on the hardware and filesystem exactly - I'm more used to ext2/3 filesystems and '-R stride= x '. Although I'm already happy with the default configuration it would be nice to squeeze whatever extra overhead I can out of it before I build it once and leave it to run for a couple of years.

_________________
hardware/software agnostic sadmin
Okay - decidedly not the answer I was expecting! There is a lot of wisdom there though, I suppose, and it's not like I really have the time to spend doing non-essential stuff.

I did specify what it will be doing though - pretty much everything from fast sequential writes of small files to copying huge images. It won't be hosting any databases (PA-RISC and Power machines do that).

Maybe I'll start on replacing the XBow in my Octane with the 1.4 version that just arrived instead :]

_________________
hardware/software agnostic sadmin
Although only the O2 is currently supported, there are some recent commits for IP30 and IP27 and IP35 is being worked on too. Backwards support for IP22 doesn't seem to be that much of a priority looking at the thread, which is a shame - I'm happy with Irix on the faster machines but OpenBSD is increasingly useful on many older architectures as an alternative to linux. It's already effectively superseded linux on sparc32 now that nearly everyone has dropped support, and I'd really like it as an option on my older MIPS stuff, like the Indy currently running Debian under my desk. OpenBSD 4.3 is looking very promising.

_________________
hardware/software agnostic sadmin
Hi all, first post for a long time but I still visit to read frequently and fire up my SGIs whenever I get a chance (which is not as often as I'd like anymore, sadly).

However, having some time free over the bank holiday weekend I thought I'd reacquaint myself with good ol' Irix and update some of my stuff, particularly the Nekoware packages on the Octane and Indigo2. Somehow I forgot my Indigo2 wasn't a r10k and cheerfully slapped a load of standard mips4 goodies on it - feel free to laugh. doh!

Obviously, that didn't end well so I carefully removed all the offending items and stripped it back to a pretty basic 6.5.22 install before adding a small selection of essential mips3 programs. Probably unsurprisingly, the poor fellow isn't really behaving itself so well anymore - most stuff has been fixable but the show-stopper is neko_sshd, which I've reconfigured a couple of times now but is still borked.

Remote login attempts silently fail and trying to ssh to localhost just gets me "Bad Request Code". I've googled, grepped and searched here but am really thinking I've probably doomed myself to a fresh start and total reinstall (the Indigo2 was my 'learner machine' for years so to be fair, it was pretty messy anyway unlike my Octane, which is lovely and neat). That wouldn't be a terrible thing anyway, and I already have my stack of overlays to hand and ready to roll.

Any suggestions? Other than "haha you idiot!"

_________________
hardware/software agnostic sadmin