Applying low distraction design to alerting systems
Writing yesterday's entry hasleft me with some thoughts on creating low-distraction alerting andmonitoring systems. Obviously this should only include informativemonitoring , but once you've got that you stillneed to present the information on what alerts are active in a goodway. And because you want sysadmins to check your alerts page relativelyfrequently, you want it to be low distraction in the same way that emailchecks should be.
A low distraction system needs to show you enough information for youto make at least a
How my mail notifier avoids interrupting me
For years, I've used a program called xlbiff to handle notifying me about newmail, and I have it set to check quite frequently. On the surface, thisought to be a really disruptive thing; new mail notification is widelyheld to be one of the best ways to yank your attention away from whatyou're currently doing. But in practice xlbiff doesn't act this wayfor me; it's far less interrupting than it sounds while still lettingme deal very rapidly with important things
Don't make your 'I am processing' animation too complex
This is a lesson from long ago, one that I learned from the veryfirst Amiga word processor. This word processor's mouse busy-cursoranimation was of a bird leisurely flapping its wings; because theoriginal Amiga was kind of slow, people wound up seeing this animationfairly often. Several years after the Amiga was released, I wound uptalking to one of the people who had designed this (she was my HCI teacher) and she asked me whether us actual Amiga users had liked thebusy cursor;
ZFS and multi-pool shared spares
One of the features of ZFS's spares handling is that if you havemultiple pools, you can share spare disks between them. This lets youhave a single global pool of spares that are used by whichever poolneeds them, or more complex schemes if you really want them (you mightin, say, a SAN environment).
Our experience is that you want to be really cautious around multi-poolshared spares, because they're buggy and they don't necessarily workvery well in practice in
One reason why I prefer browser windows to browser tabs
I recently read about a study of tabbed browsing ,and it inspired some reflective thoughts. I am well out on the 'manythings open at once' curve of browser users, but unlike almost everyonein the study, I do almost all of it in separate windows instead of intabs. I've touched on this before , butI feel like taking another shot at one specific aspect of it, which ishow one deals with inactive browser sessions.
(By 'browser session' here I mean one window
The problem of testing firewall rules changes
In an earlier entry , I mentioned thatfirewalls are a classical case of difficult testing where differencesbetween your test and your production environments can be vitallyimportant. Let's elaborate on that.
Suppose that you have some firewall rules changes that you want tomake. As a good developer-style sysadmin, you are not going to just dumpthem on your production firewall; instead you have a test firewall thatyou push rules to first for testing. But here's the question: how isyour test firewall'
The practicalities of non-GPL'd Linux kernel modules
A commotion has broken out on the ZFS mailing list; once again, someonehas announced an open-source project to port the CDDL-licensed ZFS codeto the Linux kernel, turning it into a kernel module. Predictably, anumber of people are pointing out that this mixing of the CDDL and theGPLv2 is (probably) not legally viable, and equally predictably otherpeople are both saying that it is and putting forth ways around theproblem. But all the legal argument is missing the point; regardless of
A thought on feed readers versus the social web
It's become common for bloggers to announce their new entries byposting little announcements to places like Twitter and Facebook;the ultimate version of this is the Planet Sysadmin twitter feed . Even beyond how common this is, myvague Referer-watching here on WanderingThoughts suggests that readersactually use this.
(Although a certain amount of the hits are from robots; for some reason,harvesting URLs from Twitter streams seems popular.)
In fact, anecdotally a fair number of people have switched from feedreaders to finding out about
A (surprising) missing Unix tool program
Every so often, I run across surprising omissions in the Unix toolchestof small utilities and the like. Today's is a program to print some orall of the various passwd fields for a user, by extended analogy to id (which will print user and group information in useful ways); youwould use it to extract data like the shell or the home directory for aparticular account.
(Perhaps I should instead call this a lack in the 'GNU Unix toolchest',since everyone else seems to have
iSCSI Enterprise Target 1.4.20.1 and disk write caches (and ZFS)
First, a note about IO modes. IET has two ways of doing IO to whatever actual storage is behind the iSCSItargets it advertises, called 'fileio' and 'blockio' respectively. Fileiomore or less does regular filesystem IO, as if you had opened the backingstorage at user level and were doing read() and write() to it. Blockiodoes low-level direct block IO to the backing storage. All of the followingis specific to blockio.
There are two levels of possible write caching