The link of the time interval comes from the online comic strip BonoboConspiracy , whichneatly summarizes the reality of XML on the web in today's strip .

The following is the ZFS-based design we would like to use for our newfileserver environment , presentedfor your entertainment and whatever use you can get from it.

The basic thing we give people are 'storage pools', which are made upfrom one or more standard sized 'bricks' of storage. Each storage poolcontains one or more filesystems, and is owned by a group (or a singleperson).

(Here I am using 'filesystem' to mean 'distinct mount point' or'different

I suspect that one consequence of allowing packages to ask usersquestions at install time, as Debian does ,is that it allows divisive packaging issues to stay unresolved becausepeople can always say 'fine, we'll offer both options and let the userchoose'. No one actually has to compromise or give in, so everyone sitsaround quietly nursing the issue until (if you are lucky) it all getsforgotten about.

(Of course, then you can have a new flamewar over what the defaultchoice should be

We're still in the process of designing our next generation offileserver architecture, so I've decided to write down what we want outof the design.

• good long term storage management ,as we are likely to keep the architecture for at least five years andprobably more.
• we'd like be able to do highly available storage that has no singlepoint of failure, because we have some filesystems that are crucialto our entire infrastructure.

• we should be able to add special purpose capacity without hugeexpense.

At least for personal code, I've decided that my standard for what Iconsider the clearest idiom for something is usually what first comesto mind when I try to write code to do the something. My idea here isthat if a particular approach is what came to mind first when I tried tosolve the problem, it is quite likely to be obvious to me what the codeis trying to do when I reread it in the future because clearly thisparticular approach is what is on the top of my mind

This week, we:

• got 10,691 messages from 231 different IP addresses.
• handled 17,807 sessions from 1,456 different IP addresses.
• received 301,407 connections from at least 76,444 different IPaddresses.
• hit a highwater of 11 connections being checked at once.

Connection volume is up slightly from last week , but at this point 10,000 connections a weekmore or less is probably

Recent events have made me morecurious than ever about just how big the Slashdot effect is. If you get linked toby one of the popular sites like Slashdot, Digg, or reddit, how muchload are you actually going to get hit with?

(Correspondingly, if you are trying to make sure a web application canstand up to a Slashdotting, how much load do you need to be able tohandle?)

Unfortunately, good current data is hard to come by; most of the reportson the details

This morning, I got to see how well DWiki could stand up to high load.While getting attacked by something that has all the signs of a spambotgone berserk doesn't exactly make me happy, I was cheered to see that DWiki stood up to the situation.

(And there's a little bit of me that's dancing around in triumphthat all of the work that I've done to prepare for this finallypaid off and actually worked out for real.)

Over the incident DWiki averaged

Solaris Volume Manager (which I still call DiskSuite) keeps informationabout the state of its logical volumes in what it calls a 'metadevicestate database' (a metadb for short). You normally keep a number ofreplicas of this state database, scattered around the physical devicesthat DiskSuite is managing for you.When you are using metasets , all of the metadbreplicas have to be on disks in the metaset. This is a logicalconsequence of the DiskSuite tools needing to update the metadata toreflect which machine owns a