‘elekktretterr’ posted a note asking if anyone intended to add RAIDFrame, a software RAID utility, to DragonFly. It’s already been brought to FreeBSD and NetBSD.
jgarcia, on #dragonflybsd (EFNet), posted a link to this interesting IBM developerWorks story talking about the many types and small variation of Unix standards. Note to self: read developerWorks more often.
Miguel Filipe posted a link to a paper (slides) on speeding up the Linux network stack. The answers: It’s been discussed, and while the approach could sometimes affect speed, it doesn’t solve an actual problem, and introduces far more complexity (and therefore bugs) than it’s worth.
Sascha Wildner posted that due to spam on wiki.dragonflybsd.org, he’s only allowing known people to modify the wiki, which should only affect you if you’re a dirty, bottom-feeding, no-good spammer.
BSDTalk has an interview up with Matthew Dillon, where he talks about his goals with the DragonFly BSD Project, and makes some good points about application availability.
There’s a good pile of other useful interviews at the BSDTalk site; I did not know of any of this before. Browsing through the past interviews, I see mention of FreeBSD-based FreeNAS, a free network attached storage solution, which is also new to me. (Reminds me of the now-defunct DataHive servers…)
Liam J. Foy’s BSD Portal, which aggregates a large number of BSD headlines, has been moved to a new location: http://liamjfoy.freeshell.org/
A recently discovered tip: if you want to build world somewhere else, you need to set the right environment variables.
Simon ‘corecode’ Schubert has committed kdgb, which comes mostly from FreeBSD. This replaces ‘gdb -k
‘.
Matthew Dillon has issued a warning: HEAD (the bleeding edge code) is currently very stable. Update now, for it’s going to become pretty unstable soon. The base of the cache-coherency management system will be coming in, which he calls “probably the single most complex piece of code that is planned for DragonFly.”
Slashdot linked to a IBM Developerworks story about SCTP, a recent addition to the Linux 2.6 kernel. Yeah, we’ve got that too.
Jeremy C. Reed has gone through quite an ordeal getting FreeBSD, NetBSD and DragonFly to all boot on his laptop. Check this thread to read the process, to the happy conclusion.
The current Release version of DragonFly has bumped to 1.4.2, which includes a whole slew of recent bugfixes and the like. If you’re running 1.4.1, now’s a good time to update.
Terry Tree has been working on porting the FreeBSD hybrid scheduler; he’d like some input as the merging has been difficult.
DragonFly inherited a recently-found nfs bug from FreeBSD; however, it has been fixed.
#dragonflybsd denizen jgarcia passed along a link to this Linux.com article on “Viewing Word files at the command line“. The article says Linux, but there’s nothing really Linux-specific, as it covers various Word alternatives that are all available in pkgsrc.
Matthew Dillon’s commited a large quantity of bug fixes back to the 1.4 release tag today, and if no problems arise, 1.4.2 will be released.
William Grim is proposing to use DragonFly in his master’s thesis, where he writes a framework for user-space device drivers. Matthew Dillon and Emiel Kollof had some interesting feedback.
For those who like their console a little more roomy, vidcontrol is a way to fit smaller/different text onto your console screen immediately on startup. Attention to >font and LCD use is a good idea.
Curious about where to place rc scripts from pkgsrc? Joerg Sonneberger says where to stick them.
There’s a new mailing list – pkgsrc-users@netbsd.org – specifically for people using pkgsrc packages. tech-pkg@netbsd.org is now for packagers. To subscribe, send ‘subscribe pkgsrc-users’ to majordomo@netbsd.org.
Pkgsrc questions should generally go to this new list, though DragonFly-specific questions should be asked on users@dragonflybsd.org first. (Unless, of course, the package doesn’t build yet on DragonFly.)
Interestingly, the number of actual broken pkgsrc packages is down to only 10% of the entire collection. Much credit is due to Joerg Sonnenberger, Jeremy C. Reed, and others, for knocking this quantity down.