Hardware can be weird, Intel 10G-T X540-AT2 edition

Every so often I get a pointed reminder that hardware can be veryweird. As I mentioned on Twitter today , we've beenhaving one of those incidents recently. The story starts with thehardware for our new fileservers and iSCSI backends , which is built around SuperMicroX9SRH-7TF motherboards .These have an onboard Intel X540-AT2 chipset that provides two 10G-Tports. The SuperMicro motherboard and BIOS lights up these portsno later than when you power the machine


A peculiarity: I'm almost never logged in to websites

Over time, I've come to understand that normal people are almost alwayslogged in to a whole host of big websites out there, even when they'renot actually using them. Facebook, Twitter, Flickr, LinkedIn, Google,and so on and so forth, all of these pervasive behemoths that manypeople sooner or later wind up with accounts on (undoubtedly plus manyother sites).

As you may have guessed, I don't do this. I especially don't dothis with behemoths


Why LinkedIn's 'you must join to unsubscribe' is evil

Recently I got a ' would like to add you to their professionalnetwork' email message from LinkedIn (from what I'm certain is aspammer). I'm not a LinkedIn user, so in an excess of optimism Iwent to the 'unsubscribe' link in the email. And, well, let me quotemy own Tweet summarizing things:

@thatcks :I see. To get LinkedIn to stop emailing me connection invitations, Ihave to actually join LinkedIn. That makes those emails


Another piece of my environment: running commands on multiple machines

It's my belief that every sysadmin who has a middling number ofmachines (more than one or two and less than a large fleet) sooneror later winds up with a set of tools to let them run commands oneach of those machines or on subsets of those machines. I am noexception, and over the course of my career I have adopted, used,and built several iterations of this.

(People with large fleets of machines generally never run commandson them by hand but have some sort of


Why our new SAN environment is separate from our old SAN environment

Our move from our old ZFS fileservers to our new ZFS fileservers is a totalmove; we're replacing both the fileservers and the backends and migratingall of the data to newly created ZFS pools (primarily because of theZFS 4K sector disk issue ). However, thereare at least two ways to do this overall move.

The first way is to build the new environment out so that it iscompletely separated from the old environment and both runindependently. The new fileservers get new virtual fileserver namesand migrating


Our second generation ZFS fileservers and their setup

We are finally in the process of really migrating to the secondgeneration of our ZFS fileserver setup , soit seems like a good time to write up all of the elements in oneplace. Our fundamental architecture remains unchanged. Thatarchitecture is NFS servers that export filesystems from ZFS poolsto our client machines (which are mostly Ubuntu). The ZFS pools aremade up of mirrored pairs, where each side of a mirror comes froma separate iSCSI backend. The fileservers and iSCSI backends areinterconnected over two separate 'networks


The benchmarking problems with potentially too-smart SSDs

We've reached the point in building out our new fileservers andiSCSI backends where we're building the one SSD-based fileserverand its backends. Naturally we want to see what sort of IO performancewe get on SSDs, partly to make sure that everything is okay, so Ifired up my standard basic testing tool for sequential IO . It gave me some numbers, the numberslooked good (in fact pretty excellent), and then I unfortunatelystarted thinking about the fact that we're doing this with


The temptation to rebuild my office machine with its data in ZFS on Linux

A while back, I added a scratch drive to my office workstation tohave more space for relatively unimportant data, primarily ISOs forvirtual machine installs and low priority virtual machine imagesand extra disks (which I used to test, eg, virtual fileservers andiSCSI backends). About a month ago I gave into temptation and rebuiltthat disk space as a ZFS pool using ZFS on Linux , fundamentally because I've come toreally like ZFS in general and I reallywanted to try out ZFS on Linux for something real


Why I like ZFS in general

Over time I've come around to really liking ZFS, not just in thecontext of our fileservers and our experiences with them but in general. There are two reasons for this.

The first is that when left to myself the data storage model Igravitate to is a changeable collection of filesystems withoutpermanently fixed sizes that are layered on top of a chunk of mirrored storage . I believe I've been doing thissince before I ran into ZFS because it's just the right and simpleway: I don


My view on FreeBSD versus Linux, primarily on the desktop

Today I wound up saying on Twitter:

@thatcks : @devbeard I have a pile of reasons I'm not enthused about FreeBSD, especiallyas a desktop with X and so on. So that's not really an option.

I got asked about this on Twitter and since my views do not in anyway fit into 140 characters, it's time for an entry.

I can split my views up into three broad categories: pragmatic,technical, and broadly cultural and social. The pragmatic