[ovirt-users] Re: Alternative Perspective - Re: oVirt alternatives

On Mon, Feb 7, 2022 at 1:15 AM Patrick Hibbs <hibbsncc1701(a)gmail.com&gt; wrote: > > I wouldn't mind doing some testing. I have a little coding experience but it's mostly on the desktop (Application) side of things not Web. Although if it meant getting a proper certificate management UI I'd be willing do it. (I've been thinking about rolling up my sleeves for that exact purpose anyway.) > > The main issue as I see it is two fold: > > 1. We don't have all of the needed sources to rebuild ovirt from scratch. I.e. We're missing the oVirt Node build scripts. Thanks for asking. You are definitely not missing them. Why do you think so? I agree they are not very easy to follow, scattered around in various places, etc. But if someone comes up and says "I want to have node built on top of Alma/Rocky/Oracle/Whatever linux, and I am going to keep supporting it", this would be most welcome. I am not a node developer myself, so do not often deal with it. Some places you can start looking, if you care, are the git repos: ovirt-release, ovirt-node-ng, ovirt-node-ng-image, imgbased.

These used to be on gerrit.ovirt.org, but now either already moved, or are in the process of being moved, to github.com/oVirt. I did skim through this thread, and the other similar ones, but do not have time to reply to each and every point. I suggest that if people have concrete questions, such as "where are the node build scripts", they post them as new threads with hopefully narrow scope, that can be discussed effectively. I also have my own thoughts about the broader issues of where virtualization and the industry at large is going, but not sure they are very helpful. Best regards, > Further, we also don't have a complete set of SRPMs. I've tried getting them for backup / disaster recovery issues, and it's a huge pain to track them all down from the various repos that are involved. Keep in mind that was *before* RH started archiving repos so it's probably even harder now. (Note: You can't just do "reposync --source" that's been broken for years because CentOS didn't want to rebuild their package lists to include them automaticly. Some of them are on vault.centos.org, but some are not. Tracking down the third party repos oVirt uses is also difficult for the same reason.) Does anyone have a link to the complete set of source packages outside of oVirt's dev team? > > 2. oVirt's fate is still very much uncertain. I don't think anyone really wants to go through the trouble of creating a fork unless oVirt as a project is truely EOL'd. Currently we know that RHVM will EOL in a few years, but the oVirt project itself has made no such annoucement. All of the threads on this subject are more or less contingency planning sessions and criticism of a decision they haven't made yet. Personally, I think we should wait until oVirt has made their statement publicly before going down this path. > > As for why the criticism is being made, I can say it has some merrit. If oVirt were to continue past RHVM's EOL, or if oVirt were to be forked by the community into a new project, accepting the RH deprecations into oVirt's design and source tree is short-sighted. At best it's them trying to avoid techincal debt and loosing (unofficial) support for RHEL. At worst, it's oVirt degrading itself in deference to RH's new shiny offering at the oVirt users' expense and detriment. Again, we're now at two functionalities that have been, or will be, removed: SPICE (which is all around better than the suggested VNC replacement) and now GlusterFS (which will cause massive downtime for those unfortunate enough to have used it as their storage backend.) Given that oVirt never really supported RHEL outright, (i.e. it's not tested on that platform), and that many of the people on this mailing list have requested support for CentOS's various replacement distros. I, and others, don't see a reason for oVirt's continuing to accept these changes. A statement on the matter would be nice. > > Personally, I will wait for an official annoucement from oVirt before making any decisions as well. Although, for what it's worth, I would cast my vote to retain the GlusterFS support if it's avaiable on the hosts. I was already using GlusterFS 9 packages in oVirt 4.3 and CentOS 7 so I could connect a set of raspberry pi 4 bricks to the engine. So it's not like the support cannot exist if RH doesn't provide the packages for it. (Fun home experiment. Turns out it works just fine. I can easily run 20+ VMs concurrently with this setup, and it pays for itself via the electric bill as a bonus.) > > -Patrick Hibbs > > On Sun, 2022-02-06 at 19:07 +0000, David White via Users wrote: > > At the risk of sounding like a Red Hat or IBM fanboy, I have decided to give Red Hat the benefit of the doubt here, and to not make any decisions about switching off of oVirt until and unless an official announcement is made. > > In the meantime, I know that I need to move off of Gluster (and I made that decision before the Gluster announcement), and I would need storage with any other solution anyway, so that's where I'm going to focus my own efforts. > > In the meantime, while I realize that the optics of a company like IBM / Red Hat shutting a project like oVirt down looks bad to the FOSS community, I'm going to push back a little bit. We have had access to a FOSS application that obviously works for a lot of people. No company is required to provide their services for free, and likewise, I'm of the opinion that one needs to be willing to pay (or contribute in some way) for a quality product service. It reminds me of the mantra: "Fast, Cheap, Free - pick two". > > So here's an alternative perspective: What can the community contribute and do in order to keep the project going? Anyone could fork it, rebrand it, and run with it. > > I claim to be a software developer, and the uplink in my datacenter is only 100mbps right now (of course I can increase it when needed), so I doubt I could provide much value in terms of hosting or coding. > > But I do know security. I'm a Linux systems engineer with over 10 years of experience. I know website content management systems. And people have told me that I'm good at documentation. So I think I have a lot of skill sets that I could "offer" (albeit I don't have much time, and as we all know, time is money. I've been dealing with a serious personal matter since beginning of December, and I'm effectively an acting single parent at the moment). > > I'll end this the way I started: I'm going to wait to see what happens before I personally make any decisions to change my entire underlying virtualization infrastructure. In the meantime, I'll continue to work on what I can control - the underlying storage. And if oVirt does shutdown in the future, I'd love to have a conversation with anyone interested in helping out to fork the project and keep it running. > > > Sent with ProtonMail Secure Email. > > > _______________________________________________ > Users mailing list -- users(a)ovirt.org > To unsubscribe send an email to users-leave(a)ovirt.org > Privacy Statement: https://www.ovirt.org/privacy-policy.html > oVirt Code of Conduct: https://www.ovirt.org/community/about/community-guidelines/ > List Archives: https://lists.ovirt.org/archives/list/users@ovirt.org/message/2HY52CQ3UXC... > > > _______________________________________________ > Users mailing list -- users(a)ovirt.org > To unsubscribe send an email to users-leave(a)ovirt.org > Privacy Statement: https://www.ovirt.org/privacy-policy.html > oVirt Code of Conduct: https://www.ovirt.org/community/about/community-guidelines/ > List Archives: https://lists.ovirt.org/archives/list/users@ovirt.org/message/IS7TT2QEUA7... -- Didi