We have been hunting an serious OmniOS problem for some time . Todaywe finally have enough data that I feel I can say somethingdefinitive:

An OmniOS NFS server will lock up under (some) sustainedwrite loads if the write volume is higher than its disks cansustain.

I believe that this issue is not specific to OmniOS; it's likelyIllumos in general, and was probably inherited from OpenSolaris andSolaris 10. We've reproduced a similar lockup on our old fileservers , running Solaris 10 update

I've mentioned in passing that I keep my browser'shistory database basically forever, and I've also kind of mentioned that it drives me up the wall when websites make visited links and unvisited links look the same. These twothings are closely related.

Put simply, the visited versus unvisited distinction between linksis a visible, visual representation of your current state of dealingwith a (good) site. A visited link tells you 'yep, I've been there,no need to visit

I wrote yesterday about how I keep my browser history forever , because it represents the memory of what I'veread. A corollary of this is that it bugs me if things I've read don'tshow up as visited URLs. For example, if all of the blog entries and soon here at Wandering Thoughts were to turn unvisited tomorrow,that'd make me twitch every time I read something here and saw a bluelink that should instead be visited purple.

(One of the reasons for

If you run a mail server on the Internet and look at your logs, you mayhave noticed a lot of connections from machines that EHLO with thename ylmf-pc . There are many pages about this on the web, and thegeneral consensus is that this is some sort of long standing brute forceSMTP authentication guessing botnet or piece of software. Whatever itis, it's quite annoying and may also be unevenly distributed in action.

( I've mentioned them before in passing .)

I can'

It all started when John Regehr askeda good question in a tweet :

serious but undoubtedly stupid question: why does writing to filedescriptor 0 in Linux and OS X work?

My first impulse was to say 'lazy code that starts with a generalread/write file descriptor and doesn't bother to make it read onlywhen the fd becomes a new process's standard input', but I decidedto check the manual pages first. Much to my surprise it turns outthat in Unix there is no portable

Once upon a time, tcpwrappers were a big thing in (Unix) host security. Plenty of programs supportedthe original TCP Wrapper library by Wietse Venema, and people wrotetheir own takes on the idea. But nowadays, tcpwrappers is clearlyon the way out. It doesn't seem to be used very much any more inpractice, fewer and fewer programs support it at all, and of theremaining ones that (still) do, some of them are removing supportfor it . This isn't exclusive

Yesterday I wrote about some twelveyear old Python code Ihave that's still running unmodified, and commented about both howit's a bit sad that Python 2 has changed so little that this codedoesn't even emit warnings and that this is because Python has movedon to Python 3. This raises the obvious question: am I going tomove portnanny (thiscode) to Python 3? My current answer is that I'm not planning to,because portnanny is basically end of life and abandoned

I've been thinking for a while about the interesting and perhapssomewhat remarkable fact that I'm still running a twelve year oldPython program (although not as intensely as I used to be ). When I say twelve year old, I don't mean thatthe program was first written twelve years ago and is still running;I mean that the code hasn't been touched since 2004. For bonus points, it uses a compiledmodule and the source of the module hasn't

I use the DKMS-based approach for my ZFS on Linux install , fundamentally because using DKMS makesupgrading kernels painless and convenient. It's worked well for along time, but recently some DKMS commands, particularly ' dkmsstatus ', started erroring out with the odd message:

Error! Could not locate dkms.conf file.File:  does not exist.

Since everything seemed to still work I shrugged my shoulders andbasically ignored it. I don't know DKMS myself; as far as I've