Do you have a system like this yourself, Shingoshi? Just wondering.

What benefits does the Hammer filesystem offer over native linux filesystems? I looked on wikipedia and on the website, but they list a bunch of facts, but I don't know what they mean, is it faster? Safer? Use less energy?

I'm also curious about this. Have you tried out your suggestion?

vinegaroon:
1.) I have had a liquid-cooled quad-socket system with 1TB RAID1. I am now building another system to replace it. It'll be a phase-change chilled liquid-cooled build engine.

Romanus81:
2.) Hammer is intended to support the cluster infrastructure being built by DragonflyBSD. It is designed to scale into size dataset sizes that are almost astronomical. It offers new forms of redundancy across networks. But the truth is, if you don't understand any of this, there's really no point in pursuing it.

hitest:
3.) I already clearly stated I DON'T have a system to do this on. Currently, I'm spending all of my money on building my new system. The topic of Hammer was originally raised by another linuxquestions.org user. It was Tallship who suggested that another user try Hammer (http://www.linuxquestions.org/questi...83#post3626283). We considered another approach to doing this. I simply raised the suggestion of an alternative to facilitate this. But I also know that your point is that if I haven't done this myself, I shouldn't be making suggestions for anyone else to try it. That's irrelevant! My suggestion is for those who had already thought about doing this on their own. And you're obviously not one of those people. So I have nothing else to say about this to anyone who thinks they have the right to be critical or disapprove of the idea.

Shingoshi

I'm not being overly critical, Shingoshi. You are perhaps being a bit over-sensitive. The system suggestion probably has merit, but, we don't know if it works.
I'm asking a fair question that is relevant. It would be good for new users to know that the idea actually works before they alter a working system. In other words you build the system, explain how you did it, then call for testing and bug reports.
I can understand that you are saving your money for a new system. Maybe you could test the idea in a VM? If you don't want to run the system that is fine. I hope the system idea works out for you.

Haha... you think?!

I wish ALL of YOU good luck.

1.) The first time this topic was raised, Fred87 (Slamd64) came in with criticism of the very idea (http://www.linuxquestions.org/questi...61#post3628061). After that all discussion on this ended. This thread is intended to eliminate the original objections.
2.) How much more clear can I be. This is not a solution for the inexperienced. If you're a new user, you likely wouldn't even have the understanding, much less the skill to consider this.
3.) DragonflyBSD can best be described as an expert system. Novices DON'T use or install enterprise Linux systems.
4.) People who have run pkgsrc on Slackware DON'T need to be tutored through the steps to do this.
5.) You can't test Hammer in a VM. The R/W capabilities simply wouldn't work. This was suggested for the sake of having a FULLY-FUNCTIONAL Hammer filesystem, which presently DOES NOT exist on Linux.
6.) So this thread is simply for those who already have an interest in this. Why others feel the need to interject themselves here makes no sense. But I'm sure this won't be the end of it.

Shingoshi

My point was I don't see how can you say this will be no problem without having done it.

Cool... I guess, seems irrelevant to my question though.

Hi,

How are you going to eliminate objections? I feel you won't gain any ground until you prove that this will work as you hypothetically present. Fred had a valid argument;

You say that you are going to attempt modularity with the kernel. How? I reread your threads and posts on this subject and all I find is bloviating. Wording something doesn't make it happen. Matthew Dillon has no reason to try to implement on a GNU/Linux kernel. Compare the two, sure if Mr. Dillon would want to cross over from BSD to GNU/Linux it could possibly be implement within after major inclusion from the BSD kernel (which I don't think will happen). But me thinks that Dillon is not going to tarnish his rep in the BSD community by working on a Filesystem for GNU/Linux or even association. Then add in the differences in licenses your attempt will be futile not just restricting.

So you think exclusion of valid arguments to not do it are not worthy because you say so? Yes, your right there, it won't be the end of it. Much in the same way you tend to reply to others threads in that fashion. Until your proof that the use of the filesystem will work in the way you state to do it then those same people that you seem to belittle will continue to inject.

To attempt to do the inclusion will not work the way you present it ;

Do it then report back your success. I'm sure that the failure(s) would be easier to present back here.

Hint: The argument is removed by not running a Linux kernel!
1.) I've already installed and run NetBSD pkgsrc on Slackware. Did that without any trouble. Reversing the procedure is no different. You simply need to select which of the original DragonflyBSD packages aren't needed on the new system which will be using the Slackware packaging tools, and remove them. And you don't have to replace all of them at once. But starting out with a minimal install of Dragonfly would be the best thing to do.

2.) And my point about hardware, is that I have the disk space to do this. The big problem for me here is that typically I've found that the BSD's require formatting which tends to dominate other procedures. And since the big payoff here is having Hammer, it doesn't make any sense doing it any other way. You simply cannot run Hammer with WRITE capabilities on anything other than the DragonflyBSD kernel. So what's your point? You certainly aren't offering any suggestions on how to go about this yourself.

And let's be real about this. Probably less than 5% of Slackware users run any sort of RAID system. But then that's true of most Linux users in general. And the same thing is true about their filesystems. Most users predominantly use ext3. They just can't see the point to doing anything differently. So what's the point in the continued discussion? I doubt very seriously that you had any desire whatsoever to do this yourself. If you had, you would have gone ahead and done it by now. Or at the very least said "cool, I think I'll try that! Someone else has read this thread and already decided to do just that. That was the only purpose of writing this thread. Not so I could debate with people who have no interest in doing anything than what they're familiar with.

The thing that is more likely the case is a desire to discredit or disprove any benefit from departing in any manner from what is the official Slackware system. Slackware users typically aren't even respectful of someone's desire to use slapt-get or src2pkg. Same mantra, same results. For a system that's supposed to be flexible enough to do whatever you want, it amazes me how often people complain to the point of sounding threatened or indignant, that someone else would choose another path. Instead what some of the adventurous are constantly faced with is this sort of communal mindset. Similar arguments would have been raised here if I had suggested to anyone to use NetBSD pkgsrc. The same "just can't see the point" argument would be raised as well.

So who cares! You don't think it will work. Who cares? You don't think it would be easy. Who cares? It's too much work. Who cares? Have I made my point yet? Who cares!!

No one needed to lead me by the hand through the steps of how to install NetBSD pkgsrc. Didn't need any help doing it. Didn't need any help installing and running Gentoo-AMD64 on my Slamd64 system in chroot. Didn't need any help figuring out that I could do a direct upgrade from Slackware to Slamd64. The same was later true for upgrading directly to Slackware64. I just went ahead and did it. And you're complaining that it might not work. That's ok. You probably haven't done half of the stuff that I've done. I'm confident in my abilities. I do something because I set my mind to doing it.

And frankly, I shouldn't have had to write any of this. For me, it's just a waste of time. I think I've made all the points I need to here. And there won't be anymore. The rest of you can keep coming back complaining about how you "just can't see the point!

Shingoshi

Hi,

Ouch!

Bloviating!

As I said 'Do It'. Don't try an lead someone else to jump of the barn.

Oh crap, I didn't realise you were so 1337.

Seriously though, I'm definitely not saying this is pointless. I'll look forward to seeing your completed system.

Hopefully a 650 gig sata qualifies as a large disk.
I've been using nothing but Linux for the past 5 years.
Not a long time but long enough to know how to run the same systems you have run,,,, with no help.
Still, with the reading I have done this all is probably over my head.
As for being "adventurous", that's part of the fun of 'nix.
Still, this is the net.
Folks are going to put their 2 cents in no matter anyone's feelings on the subject.
Why care what someone else thinks,,,, it's your box dude.

I shall make no further comment, in case I get banned again.

I'm not even sure it has merit. If one was to run the DragonFly kernel and filesystem, then you wouldn't be able to use stuff that's tightly integrated with the linux kernel such as iptables, lvm, md-adm and all that stuff that makes linux linux. You could potentially port some of slackware's userspace packages over but they'd probably be available BSD native anyway. You could customise the Dragonfly startup scripts and suchlike to be more slackware like, but it would necessarily still be more Dragonfly than Slackware.

I really don't understand the point behind the suggestion. If you want to run Dragonfly and Hammer, then run them. I can see no value in attempting to make Dragonfly look like Slackware or use parts of slackware or it's application packages.