RealRancor from the ownCloud forum mentioned recently ( https://central.owncloud.org/t/strip-down-supported-major-releases-of-owncloud/1586?u=tflidd ), that the current release and maintenance schedule, ownCloud ends up with 5 supported versions. This is a lot of work for everybody, for core developers, app developers, support, and users. I Nextcloud followed this maintenance cycle, we would end up supporting NC 9-14 in Dec 2017/Jan 2018!
I`m a little surprised that here is no discussion. For me as a user who has very limited knowledge of programming it sounds like very much work to support so much versions at the same time. Why not release just one stable and create some more better controllable little steps. I know, nextcloud is a very complex thing with a lots of third party programs.
All found bugs must be fixed in all possible 5 stable versions. All the work more than third or four five times.
All this things let me think is would be easier to only support one stable. Can please some of the core developer explain what benefits have the current release model?
As a company we use ownCloud at the moment. We want to migrate to Nextcloud soon.
At the moment Nextclouds Maintenance and Release-Schedule does not contain any EOL entries.
So I would like to second, that an up to date Maintenance and Release-Schedule like the one existing for ownCloud would be helpful to plan our updates and support.
Thanks.
We donāt have a public plan yet but right now the plan is to offer support for 1 release plus some months to migrate. So we would soon drop support for Nextcloud 9. Again, preliminary plan.
Customers can of course purchase our promised up-to-15-year support and updates.
OK. Thanks for the answer. From what I can see on Maintenance and Release-Schedule the NC-Releases come out roughly every 4-5 months. Is that the planned cycle? That would mean that we would have to upgrade roughly every half year to have security support. That is quiet frequently looking at that we would have to test, inform users, maybe update documentation, have downtime, take risks that still after testing something breaks due to the update.
Is there a LTS-version planned and if so an ETA when it will be released?
@TomFernandes every release is essentially LTS for customers - we offer 10 years support to business users. For home users, who donāt have a lot of users to inform and documentation to update, a faster cycle means features faster so for them it is fine. I guess weāll discuss details at the January hackathon, this is pretty much my opinion on how it should go - not decided at all!
To get things clear: you mean, that with a support contract we could (if it goes the the way you think it should go) stay on e.g. NC 11 and you (the company I purchased the support contract from) would provide us with security-updates for that release for e.g. 10 years?
That is correct. It is one of the main benefits of a support contract. It goes up to 15 year, I believe, though it depends a bit on platform/OS releases. Our goal is to maintain Nextcloud releases in alignment with RHEL and SLES releases, so you can pick a platform (RHEL 7) and a Nc release (eg 11, 12 etc) and stay on it until the end of the lifetime of the platform. As RHEL and SLES are typically supported for 5-15 year, we support Nc for that long, too.
Oh, thatās very interesting. Iād of course prefer some kind of free LTS-version but from a marketing point of view this makes very much sense. Thanks for clarification. Iād be grateful if the final decision from the hackathon will be pushed to this thread or to a newsletter or so.
Note that FOR SURE we offer 10-15 year support to our customers, that was decided when we started the company. So if you need that you should contact sales anyhow
The only open question is how long the support will be for free. There wonāt be an LTS release, 99% sure, but maybe we will support versions for longer than 6 months for free. Maybeā¦
That would be a huge drawback for me. I try to avoid all x.0 releases, there are usually bugs and 3rd-party apps need some time to release a suitable version. This effectively cuts the time down to 5 months. Currently, Iām doing major upgrades every 9-12 months (I updated through some major versions). Major versions come with a lot of changes, e.g. NC 11 removes php support for versions <5.6, in some cases you need to update the underlying operating system or even exchange devices (NAS boxes) or find other solutions.
The risk is that many people end up using unsupported nextcloud versions.
Sadly, weāve done statistics on usage a few times and you wonāt believe the number of people on a release like ownCloud 8.0.2 or something like that. Totally insecure and full of known and fixed bugsā¦
Our plan is to work towards an auto-updating Nextcloud. That would need a lot of testing and work and I suspect we will keep releases around a bit longer until we get there. Remember, weāre talking about proposals/thoughts here, not plans. Youāre right that there are serious drawbacks for some users and we want to avoid that, of course. Maybe a 9-12 month lifecycle is fine, if we offer customers 3-5-10 years.
The auto-updating in web browsers works really good. That was often a huge problem that people didnāt update their browser and plugins. But since they have this auto-update in place and they also disable extensions and plugin with security problems (e.g. flash), that is really a huge step forward.
For Nextcloud this would be a huge step. Updates were very problematic and first we need to make sure that the process works better and more reliable (the new updater app is a large improvement). I was hoping we get all the release cycles and update procedures (e.g. skip major upgrades) in such a way that Linux distributions can easily package Nextcloud. Then it would be very easy for all the Linux users to just use an apt-get/yum/whatever update which they are already used to run on a regular basis.
We donāt have a public plan yet but right now the plan is to offer support for 1 release plus some months to migrate. So we would soon drop support for Nextcloud 9. Again, preliminary plan.
[ā¦]
The only open question is how long the support will be for free. There wonāt be an LTS release, 99% sure, but maybe we will support versions for longer than 6 months for free. Maybeā¦
Version 9 is supported for year now, NC 10 for 7 months now and NC 12 is approaching (which is all great!).
But not being clear about when support will be dropped creates uncertainty, as one is not able to plan/predict upgrades well.
It would be helpful if the project would have the same clarity about the EOL like it has about the release schedule
Found this thread while doing some research on how we can migrate Mail-in-a-Box from ownCloud to Nextcloud which is what many users in that community suggested.
The situation with MIAB for instance is that itās running on Ubuntu 14.04.5 LTS which is still officially supported by Ubuntu until April 2019. 14.04 however only ships PHP 5.5 which makes it unsuitable for Nextcloud 11+ unless PHP is upgraded manually.
I agree that it would be great to have an official commitment by the Nextcloud team on EOL policy. Even if itās āwe will only support the current release for non-paying usersā that would be much better than no info.
Yeah, I know we should bring some clarity in this. We simply havenāt discussed it extensively - but I would expect that Nc 9 will not be supported long after 12 is out. How it goes in the future I really canāt say. It in part depends on if businesses are willing to pay for support or if many/most donāt want to. If companies donāt contribute it becomes hard to keep everything free, of course.
If a company decides to pay for support of a version, will the support updates only provided for this company or to everybody?
Another option could be crowd-funding: Letās suppose that NC provides a 12 month support, then each additional year costs e.g. $1000 (cheapest subscription is 1900 ā¬). Would there be enough people that the contribution of everyone is acceptable?
but from a marketing point of view this makes very much sense. Thanks for clarification. Iād be grateful if the final decision from the hackathon will be pushed to this thread or to a newsletter or so.
