Could someone suggests a future proof approch for the app? portainer in a VM?
bye
Olindo
Could someone suggests a future proof approch for the app? portainer in a VM?
bye
Olindo
If you’re using iX apps, just wait; they promise they’ll all be seamlessly migrated to the new docker-compose framework with the releease of EE. If you’re using TrueCharts apps, just wait; they say they’ll also have a migration solution (which I’m guessing will be something like Talos on a VM) by then.
Or if you want to tear it all out and replace it with something of your own making, of course you can do whatever you like.
Thanks for your suggestion, but i think that my way in jail, hoping that ix doesn´t cancell it
You should, of course, do whatever you think will best meet your needs. But it really seems that a lot of people are overreacting to iX’ announcement[1]. iX is going to have a migration path with their apps. TrueCharts is going to have a migration path with theirs. We don’t as yet know much about what either path is going to look like[2].
Electric Eel isn’t due out until October, and Morgan’s said (somewhat surprisingly, I’ll admit) that Dragonfish will still be supported for a good time afterward. Nobody (in terms of users) needs to be taking immediate action on this. In other words,
Now, if you were wanting to toss the apps anyway, hey, knock yourself out–there are lots of options, each with its own pros and cons. But it’s premature (IMO) to believe you need to do that now on account of this announcement.
and I’m one of those who believes strongly that this is a very bad move on their part ↩︎
Sure, we know that iX is going to move everything to docker-compose, we know TC is going to stick with Helm/Kubernetes, and it looks like TC’s not planning on Kubernetes-in-sandbox–but there are still a lot of unknowns. ↩︎
I agree & gave you thumbs up…even though I’ve also just tossed the apps and gone with docker compose on account of this announcement
I’m a bit slow to notice these things and like to be late to adopt so all the bugs are ironed out but it seems like application virtualisation has been ever changing from bhyve to iocage to kubernetes and now to docker. Is this the trend?
Also the VM platform has changed once or twice over the years too, hopefully they don’t decide to move to something else as I don’t want to have to rebuild those desktops too.
bhyve is a hypervisor; the previous app/jail platform to iocage was Warden.
Only once, really. CORE had bhyve; SCALE has kvm–bhyve works with FreeBSD; kvm works with Linux.
To clarify, as Dan mentioned, you only had to migrate the VMs if you switched from Core (freeBSD) to Scale (Linux). That’s a major change of OS so should be understandable.
The good thing about the change to docker is it opens up so many super easy deployments that were more difficult previously, especially for newer users IMHO.
As far as i know core VMS should Auto migrated during the Sidegrade from core to scale.
Yes, if you go through that process, some like to fresh install Scale and obviously then, they have to do their own.
Pretty much, migrate VM, choose cpu mode, update nic/PCIe pass through.
Job done.
I would highly recommend considering in your plans for future-proofing a system for backing up and restoring your applications. Currently this isn’t supported by ix but is being developed.
I’ve written about it in the following thread.
It seems the backup can restore mechanism that’s being developed might not be able to back up / restore individual applications. Meaning if one of your applications is corrupt you’ll have to restore everything back to a previous state even though it’s only one app that’s affected.
I hope I’m wrong but something to consider if you’re going to future proof.
Warden of course!
Final question from me on this topic.
@dan you seem pretty dialled in on this. Is there a future proof route now to manually migrate to?
I’ve already migrated a couple of apps from TC to TN apps and have a couple of TC ones left which there are docker containers out there for but I don’t know how to install on Dragonfish.
Thanks
IMO, nothing iX controls should be considered anything like “future-proof.” They have a history of discarding features that they consider “too much work,” and I don’t see any reason to believe the new incarnation of apps will be any different.
So that leaves a few options:
My own answer is to wait and see what TC comes up with. Based on that, I’ll see if I want to follow their path or do something else. I’m currently only running one TN app (Storj), and expect I’ll get rid of it before too long.
Based on their latest news on their homepage, their solution is talos Linux in a vm and a migration tool that migrates all truecharts apps from “bare metal” to the vm. Timeframe is :
ClusterTool is not fully ready yet. Here is the official roadmap (please note that these dates may change without prior notice):
Thanks for the update; I hadn’t checked their news page in the last few days. But what I don’t see mentioned there is how host paths will be handled–I presume they’re going to need to move to network (likely NFS) mounts. Edit: Yep.
With the future migration to Talos using ClusterTool, you won’t technically be able to reach these folders through “hostPath” anymore. Please ensure to make a seperate NFS share on TrueNAS SCALE for each of those folders and edit the TrueNAS SCALE App to point towards said share directly.
They already have docs on setting this up on Proxmox; if their tool will migrate the apps to that version of the cluster, that’d be a big win IMO–I don’t currently use VMs on SCALE and wasn’t really planning to start.
If you need NFS mounts in a VM will that require a bridged network device?
Assuming you’re mounting from the VM host, then yes.
I go through precisely this scenario in this video
TrueNAS Scale: Setting up a Static IP and Network Bridge // Access NAS host from VM