Eletriarnation posted:Sure, I know that they couldn't just wholesale lift closed source software - what I'm saying is that if Poettering set out to emulate those examples knowing how they work, it doesn't really make sense that what he developed would just be worse for no reason instead of working in the same way as the examples. At the very least, we have to claim that he didn't know what he was doing or that he made bad decisions on how to make necessary-for-Linux changes instead of just replicating what was in front of him. The point I'm obviously failing to make is that if systemd is still as full of holes as it obviously is, and if it's still seeing scope creep that doesn't fix the issues and adds new ones, he's not solved anything. I'm not saying they did it deliberately, I'm not into conspiracy theories. I'm saying they unintentionally benefited from him being in in a particular place at a particular time. Whether it's right is debatable. EDIT: Also, don't be so quick to assume RedHat won't replace systemd - they're already replacing pulseaudio with PipeWire, despite the fact that that was written by Lennart Poettering too. BlankSystemDaemon fucked around with this message at 15:30 on Apr 11, 2021 |
|
# ? Apr 11, 2021 15:28 |
|
|
# ? May 8, 2024 05:19 |
|
The problem is, again, that systemd is replacing things that don't need replacement, constantly changing how they are replacing things. Like with changing the network management again after everyone was almost used to as it was implemented.
|
# ? Apr 11, 2021 18:10 |
Anywho, I think enough have been said to exhaust this. The battle lines have been drawn pretty clearly on this topic for a long time, there's no changing anyone's mind - so let's just have a drink and shitpost about something else.
|
|
# ? Apr 11, 2021 18:32 |
|
BlankSystemDaemon posted:I'm pretty sure he's said he set out to emulate launchd, so assuming that he didn't seems to be pretty pointless. Regardless of the systemd discussion because, agreed, no one's mind is gonna be changed; I'm not sure I've ever heard a single person defend pulse audio. It's a dumpster fire that while kind of works now, needs to be replaced; if only because of the negative stigma it has. I've only heard good things about pipewire.
|
# ? Apr 11, 2021 20:52 |
|
Mr. Crow posted:Regardless of the systemd discussion because, agreed, no one's mind is gonna be changed; I'm not sure I've ever heard a single person defend pulse audio. It's a dumpster fire that while kind of works now, needs to be replaced; if only because of the negative stigma it has. I've only heard good things about pipewire. While I've heard only positive things about pipewire too, I'm a bit more reserved about it now, since I am using Fedora 34 that comes with it by default. I have an application (written in Qt) that I use to change the default audio output device. It's more than 5 years old and it works just fine with pulseaudio. Since upgrading to F34, even though pipewire has a pulseaudio compatible API (wrapper or whatever it is), it behaves a bit differently. For example, when the DE starts (KDE), it tells me that it connected to the server, but that there are no audio outputs available. If I quit the app then restart it it works just fine. Now, maybe the pulseaudio layer is not that stable yet, so I've been meaning to write it against pipewire directly. The documentation, however, of that API is ... not quite existent to be honest. I haven't spent that much time on it, but meh I kinda gave up for now. At the moment I just live with it not starting when the DE starts and I start it manually after. I hope they'll stabilize it soon.
|
# ? Apr 11, 2021 23:01 |
Volguus posted:While I've heard only positive things about pipewire too, I'm a bit more reserved about it now, since I am using Fedora 34 that comes with it by default. I have an application (written in Qt) that I use to change the default audio output device. It's more than 5 years old and it works just fine with pulseaudio. Since upgrading to F34, even though pipewire has a pulseaudio compatible API (wrapper or whatever it is), it behaves a bit differently. For example, when the DE starts (KDE), it tells me that it connected to the server, but that there are no audio outputs available. If I quit the app then restart it it works just fine. Now, maybe the pulseaudio layer is not that stable yet, so I've been meaning to write it against pipewire directly. The documentation, however, of that API is ... not quite existent to be honest. I haven't spent that much time on it, but meh I kinda gave up for now. At the moment I just live with it not starting when the DE starts and I start it manually after. I hope they'll stabilize it soon. I have WINE apps that don't see any sound devices either in Fedora 34.
|
|
# ? Apr 11, 2021 23:14 |
|
This is a weird question that I'm having trouble getting an answer to on Google; I've used ESXi and KVM for PCI-E passthrough on my system that has IOMMU support and what not, but I'm trying to find out if PCI non-E cards are also capable of being passed through to a guest OS. I have a Windows 10 installation using an old 8-port coaxial cable input card (PCI, the OG kind) and would love to virtualize it.
|
# ? Apr 30, 2021 00:53 |
|
Less Fat Luke posted:This is a weird question that I'm having trouble getting an answer to on Google; I've used ESXi and KVM for PCI-E passthrough on my system that has IOMMU support and what not, but I'm trying to find out if PCI non-E cards are also capable of being passed through to a guest OS. I have a Windows 10 installation using an old 8-port coaxial cable input card (PCI, the OG kind) and would love to virtualize it. You could get a pci-e to pci adapter card and do it that way?
|
# ? Apr 30, 2021 01:09 |
|
Wibla posted:You could get a pci-e to pci adapter card and do it that way? LOL you're no help. I guess I'll make a bootable ESXi drive and just boot that host, and see if the PCI card shows up as available to passthrough.
|
# ? Apr 30, 2021 16:14 |
|
Less Fat Luke posted:LOL you're no help. I guess I'll make a bootable ESXi drive and just boot that host, and see if the PCI card shows up as available to passthrough. Old school PCI doesn't support IOMMU IIRC so you'll probably have to pass through everything that's on the same PCI bus, and it may only work if the PCIe-to-PCI bridge is wired into a PCIe port that supports ACS.
|
# ? Apr 30, 2021 16:34 |
|
Just a guess but I'd say it would come down to the PCI bus itself and how it presents devices attached to it. I'd assume there's some kinda standard for that poo poo so it shouldn't matter how the devices behind the bus are connected and as long as the bus supports passthrough it would work in theory except if the hardware does DMA or some poo poo who loving knows love to know how you get on.
|
# ? Apr 30, 2021 16:40 |
|
Less Fat Luke posted:LOL you're no help. I guess I'll make a bootable ESXi drive and just boot that host, and see if the PCI card shows up as available to passthrough. Wow. You're welcome
|
# ? Apr 30, 2021 19:10 |
|
Pile Of Garbage posted:Just a guess but I'd say it would come down to the PCI bus itself and how it presents devices attached to it. I'd assume there's some kinda standard for that poo poo so it shouldn't matter how the devices behind the bus are connected and as long as the bus supports passthrough it would work in theory except if the hardware does DMA or some poo poo who loving knows love to know how you get on. My thoughts too, we'll see. I went to install ESXi and realized nothing I have network card wise is compatible so I'm going to grab a cheap Broadcom before testing further.
|
# ? Apr 30, 2021 19:17 |
|
This may be a stupid question that's answered in the docs but I'm making the switch from vmware to proxmox to get rid of the vcenter overhead and I'm having some trouble adjusting to the way it works in some cases. In esxi-land I had an NFS share that contains a ton of ISOs separated out in directories, e.g., /nfs/windows/server, /nfs/windows/desktop, /nfs/linux/centos etc. Proxmox lets me mount the NFS but it seems to insist I throw all ISOs into /nfs/template/iso and doesn't let me subfolder things. Is the intent that I have everything in one big bucket and just use tags to search for what I want? Transition has been fairly smooth otherwise though honestly the only reason I trimmed down was to ditch the bloated pig that is vcenter for a one-node homelab and still retain the ability to create templates, etc.
|
# ? May 6, 2021 19:04 |
What overhead? I'd love to see some numbers that prove which one has more overhead, because unless you're looking at OccamBSD (just as an example, because it's what I know; it's FreeBSD stripped to its absolute minimum to only run bhyve and nothing else (not even networking)), I'd be surprised if there's a measurable difference on comparable workloads.
|
|
# ? May 6, 2021 19:26 |
|
Free ESXi can do everything I needed *except* templates so dedicating 10+gb RAM and whatever size storage for VCSA on a single node server didn't make a whole lot of sense.
|
# ? May 6, 2021 19:30 |
|
I continue to preach the good word of XCP-NG
|
# ? May 6, 2021 23:33 |
|
Martytoof posted:Free ESXi can do everything I needed *except* templates so dedicating 10+gb RAM and whatever size storage for VCSA on a single node server didn't make a whole lot of sense. Ram is cheap, esp compared to slamming your dick in a drawer dealing with proxmox idiosyncrasies. Every time i tried jumping from esxi to prox, I’ve jumped back fairly quick.
|
# ? May 7, 2021 21:24 |
|
Yeah, that’s definitely true. That being said, I’ve moved a few things over to proxmox and it’s seemingly doing the job well enough. Hardware is an old R620 with 128gb ram and 1.2tb in an array of six 300gb spinners I got from work a while back so I don’t actually want to put any money into upgrading it or anything, so I’m fine just reclaiming the disk space and ram from VCSA in favour of proxmox. My end goal is to deploy as code anyway and not touch the GUI so who knows, maybe I’ll go back to bone stock esxi once I have my poo poo sorted out.
|
# ? May 8, 2021 00:41 |
|
Terraform works pretty well with proxmox and cloud init. Combine it with packer for building your templates and you'll be pretty happy.
|
# ? May 8, 2021 05:35 |
|
Quick warning, virtuallyghetto is now williamlam.com, articles are still here. It looks like there have been some external push given how sudden it was.
|
# ? May 8, 2021 14:36 |
|
SlowBloke posted:Quick warning, virtuallyghetto is now williamlam.com, articles are still here. It looks like there have been some external push given how sudden it was. My guess would be that he was trying to stop using the word ghetto for his website. He does work for VMware, and while it isn't an official company site it does kind of reflect on the company. He is a good guy, but that name did always kind of bug me.
|
# ? May 8, 2021 19:10 |
|
I didn’t even make it a week into proxmox. For some reason I felt the performance of my VMs was really sluggish. Not sure why but I bet it was some human error on my part. Either way, just resigned to continuing the VCSA life forever I guess. I thought about going the xen route but man, that needs a management VM too if I want to do any fun template stuff, and if I’m doing to do that I may as well just stick with the one I have licensed now. RIP experimenting I guess.
|
# ? May 12, 2021 20:36 |
|
Martytoof posted:I didn’t even make it a week into proxmox. For some reason I felt the performance of my VMs was really sluggish. Not sure why but I bet it was some human error on my part. You can talk directly to the hypervisor via the xe commands, and is compatible with all the known automation tools I know of. Plus, XCP-NG comes with the Xen Orchestra virtual appliance. https://www.criticaldesign.net/post/automating-lab-builds-with-xenserver-powershell-part-3-unlimited-vm-creation
|
# ? May 12, 2021 20:47 |
|
That's a good writeup, thanks! I might give it one more kick at the can. I still have to reformat everything one more time since I have larger SD cards coming form Amazon so I can install ESXi7.
|
# ? May 12, 2021 20:51 |
|
It’s been a while since I actively did any networking in the vSphere world — is there anything out of the box in vSphere 6.7 or 7 that provides layer 3 switching between networks or is throwing a vyos/pfsense/tiny linux box with ip forwarding on two vSwitches still the way to go? I don’t need anything complex, just simple routing to keep my home network and lab networks from crowding each other.
|
# ? May 13, 2021 02:21 |
|
Martytoof posted:It’s been a while since I actively did any networking in the vSphere world — is there anything out of the box in vSphere 6.7 or 7 that provides layer 3 switching between networks or is throwing a vyos/pfsense/tiny linux box with ip forwarding on two vSwitches still the way to go? I don’t need anything complex, just simple routing to keep my home network and lab networks from crowding each other. Nsx-V is their out of the box solution, which might be a bit overkill if vcsa was too much overhead for you. SlowBloke fucked around with this message at 07:24 on May 13, 2021 |
# ? May 13, 2021 07:21 |
|
SlowBloke posted:Nsx-V is their out of the box solution, which might be a bit overkill if vcsa was too much overhead for you. NSX-V is EOL you should be looking at NSX-T now which is a complete rewrite and much much much better. Given the use case though just install a router VM.
|
# ? May 14, 2021 23:35 |
|
Oh ya, I was somehow hoping that L3 routing was maybe just a button they added to a distributed vswitch in vsphere or something but I should have figured it would be functionality hidden behind an expensive overkill product. Just going to throw whatever the modern equivalent of floppyfw is between these two vswitches, thanks gang. Decided to not really waste any more time on researching hypervisors and just stick with the one I know. Free Xen sounds tempting but my experiment with Proxmox just left me not really wanting to re-learn the whole thing so I guess I’ll stick with VMware for better or for worse.
|
# ? May 15, 2021 04:08 |
|
I just use a barebones pfsense for stuff like that.
|
# ? May 15, 2021 08:17 |
|
Martytoof posted:Oh ya, I was somehow hoping that L3 routing was maybe just a button they added to a distributed vswitch in vsphere or something but I should have figured it would be functionality hidden behind an expensive overkill product. Just going to throw whatever the modern equivalent of floppyfw is between these two vswitches, thanks gang. The problem is that the DVS is a distributed construct so adding routing there also needs to be distributed so it’s not quite as trivial as doing a simple routing appliance. There’s a fair amount of trickery going on under the covers in NSX to make distributed routing work.
|
# ? May 15, 2021 09:20 |
|
I seem to recall this thread having a negative opinion on Virtualbox, so if not that what's my best free option to run a basic Linux VM on an Intel Mac running Big Sur with USB passthrough? I have a device with a USB console port that doesn't have drivers for modern Mac OS, a Debian VM with the adapter passed through seems like the easiest answer.
|
# ? May 17, 2021 22:16 |
|
wolrah posted:I seem to recall this thread having a negative opinion on Virtualbox, so if not that what's my best free option to run a basic Linux VM on an Intel Mac running Big Sur with USB passthrough? You could try UTM https://mac.getutm.app
|
# ? May 17, 2021 22:38 |
|
wolrah posted:I seem to recall this thread having a negative opinion on Virtualbox, so if not that what's my best free option to run a basic Linux VM on an Intel Mac running Big Sur with USB passthrough? It works, it's just if you've used Parallels or VMware it sucks.
|
# ? May 18, 2021 13:10 |
IOMMU passthrough in Hyper-V is embarrassingly bad.
|
|
# ? May 18, 2021 15:18 |
|
wolrah posted:I seem to recall this thread having a negative opinion on Virtualbox, so if not that what's my best free option to run a basic Linux VM on an Intel Mac running Big Sur with USB passthrough? I use it all the time with vagrant and it works well. The UI isn't great but I rarely deal with it anyway.
|
# ? May 18, 2021 15:35 |
|
SlowBloke posted:You could try UTM https://mac.getutm.app Bob Morales posted:It works, it's just if you've used Parallels or VMware it sucks.
|
# ? May 18, 2021 17:05 |
|
Martytoof posted:Oh ya, I was somehow hoping that L3 routing was maybe just a button they added to a distributed vswitch in vsphere or something but I should have figured it would be functionality hidden behind an expensive overkill product. Just going to throw whatever the modern equivalent of floppyfw is between these two vswitches, thanks gang. I loooove proxmox. VMware is like if I want to do more I have to pay and for my homelab I'm not going to pay the cost of non free esxi to do what I can do just as well with proxmox. I use VMware and hyperv at work and they are great in their own ways but for a homeserver I really don't think I'll ever change off proxmox.
|
# ? May 31, 2021 01:09 |
|
Yeah I think it was probably fine and vSphere is way overkill for what I want. In the end though I just wanted to work on my lab rather than the hypervisor and already had a license so it was actually more effort to take the few days to re-learn everything I know in proxmox. Not a huge time investment but honestly I just couldn’t be bothered.
|
# ? May 31, 2021 02:35 |
|
|
# ? May 8, 2024 05:19 |
|
Is there a good way to back up proxmox? been using it for about 6 months now and would like to do something just in case.
|
# ? May 31, 2021 15:56 |