Has there been any thought given to having a community led version of CORE?
Many software vendors have this.
I think the term youâre looking for is âfork.â And yes, there has.
Iâm aware of the âforkâ but Iâm more suggesting an IX backed CE.
I recall a negative response from someone working in iX on the old forum; might be wrong.
Fork?! Forks are difficult because on one-hand you are stranded on an island but on the other your free to do what you want. But a fork does send a ripple through the user-space and really can have detrimental effects (the end-user being offered a choice is not always a good thing). A collaboration can seem better (more support, and resources, essentially) in some respects, but now developers are essentially working for free where iX collects the profits.
1 Like
I donât think ix is interested in a community edition of CORE.
The only reason CORE was âcommunityâ up to now is really because itâs their BETA testing platform for enterprise.
Now that theyâre all-in with SCALE, SCALE is basically the new âcommunity editionâ whether you like it or not.
Many software vendors have this.
Many vendors have this exactly the way ix does it. Itâs a testing ground for their enterprise version (ie. CentOS Stream, Fedora Core). Obviously bigger vendors can afford to have more variants of community editions, but Iâm guessing, that is just not the case with ixSystems.
Fork?! Forks are difficult because on one-hand you are stranded on an island but on the other your free to do what you want. But a fork does send a ripple through the user-space and really can have detrimental effects (the end-user being offered a choice is not always a good thing).
Yep, it could be especially dangerous because updates from upstream (particularly security updates) may not make it to the fork due to difficulty merging (tons of conflicts) once it diverges enough. FreeBSD pf fork is one good example. CVE-2023-4809 had already been fixed in OpenBSD in 2013, yet it continued to exist on FreeBSDâs fork until it was fixed in 2023.
3 Likes
Took a quick look at the github repo and itâs confusing (I donât know if itâs just disorganized, messy, or just that complicated)!!
Might be easier to attach your wagon to something like this:
GitHub - chermenin/xigmanas: XigmaNAS is an open source NAS / OS software appliance
Which illustrates the point. XigmaNAS is a fork from a long time ago 
1 Like
Not really. XigmaNAS, formerly NAS4Free, is a continuation of the product formerly called FreeNAS. iXSystems bought the name and wrote a new codebase.
And that was a fork of XigmaNAS (which is hosted on sourceforge).
1 Like
Waiting for zVault to take off⌠
1 Like
This somewhat reminds me of when citrix took stewardship of Xen. After years of benefitting from that package (and that community), they moved features to the pay-only version. I think it took 24 hours for a new repo to be set up for XCP-ng and the pay features made available free. Regardless of the details, the code moved back to free open source due to a ânon comitted stewardâ.
Would be nice to see that happen to freenas/truenas. I assume any and all changes made by ix over the years are unencumbered and the project could go back to open source?
Iâm one of the folks that really wants/needs vm support too. I dont need a heavy virtualization setup, just a couple vmâs. Truenas had all my bases covered with zfs/smb/bhyve. Them skipping out before updating to 14 causes me to never have pci passthrough which Iâve been anxiously waiting for.
I was assuming Iâd move to scale to get better virtualization, but having been in freebsd since 2.1 Iâm a little concerned about stability for my appliance on scale. HexOS from what Iâve read does not interest me, and seems just another way IX is going to get utility from the freenas/truenas code/community.
So am I correct that there will be no ânew featuresâ after 13.3?
It has always been, and it currently is, open source.
Only CVEs.
True. I mean, âgo back to community managed versionâ.
Oh wow, I did not know they started another OS project. The last one (TrueOS) didnât survive that long. Shame really, because Lumina had such great potential. Itâs one of the few DEâs that donât need stuff like DBus, procfs, or systemd.
HexOS is not a TrueNAS/IX project.
Near as I can tell, theyâre a customer.
Weâre confident that even though this is a hard decision, itâs also the correct decision because of the exciting new projects that weâre all becoming more involved in like TrueNAS CORE.
Did this age well?
All things considered, itâs fascinating that the technical risk not that long ago was assessed to be high enough that Scale would get relegated to âyeah, weâre also working on this other thingâ status.
To some extent, it feels like a very Linux-y experience. You can get it to work for you, but you need to put in effort to work around the oddities (be they kernel or due to the wider community that uses the kernel). At the same time, clean water (i.e. ZFS) trickles in despite the objections of the E. Coli-lovers.