Stable staging logic : Should be an RC !!!

I don't grasp the logic of Stable staging branch.

I thought the path was : Unstable > Testing > Stable staging > Stable
So I shifted to Stable staging, both to have quicker updates and to help report issues.

But then, several month ago, I had an instance where my Stable staging was... late to Stable as something (I think Firefox) made it directly Testing > Stable.

So I went back to Stable.

Now, we see a massive Stable staging update, but I'm not willing to shift to reshift afterwards.

Why couldn't all updates from Testing be pushed to Stable staging (@ the same time as to Stable if no need of a staging phase) ?

Or did I get the story wrong and it's already the procedure ?

1 Like

It's just another layer - some things will go there before stable, some things won't.... but I think it just makes it easier to shuffle stuff around for a test in stable.

Stable staging is like 'stable with a few tests'.

Just relax, it's just a matter of time - the delay to reach stable is the same. This will apply mostly to folks stressing over not getting enough updates but not wanting to go unstable.

Hi !
My question is not a problem, I don't mind being on a stable that sometimes scarcely updates ! (Even if poeple don't believe me when I say that my Debian stable updates much more frequently than my Manjaro stable ^^ )

My question is indeed about the staging policy. I could be eager to be part of the "prod" testers with it, but if it is just a classical testing, only used in doubt cases and sometimes late on stable, I'll not install & reinstall it.

Nope !
And you're spot on my question : I'd like an RC, not a staging in fact.

If some upgrades bypass staging, it's just a staging (isolated test for testers).

To add to the confusion: If something breaks on stable and a (packaging/package)-fix get's pushed to stable, even testing can be "behind" stable at that moment.
Seems to happen to staging more often though.

1 Like

I have never used it but I was under the impression that stable-staging is only used when it is needed and when it isn't needed, it doesn't get updated.

1 Like

You're kind of spot on.
It reflects the thread title : It should be an RC and not a "vanilla" staging" coz then it's just a dead branch that sometimes (when???) has a sudden revival...

..or it shouldn't exist. (Dead branches randomly revived are time consuming for admins & users).

1 Like

What are you talking about ?

Wo got regular security updates in the past month, several of them for Firefox, for example.

Aren't you paying attention or is it just for trolling ?

2 Likes

Isn't it mostly one person complaining loudly about security updates in many topics?

Also, which critical security updates are you missing?

3 Likes

Oooops ! come back to the thread !

No complain on updates frequency, just branch maintenance methodology.

1 Like

I think this is the case...

I don't see why it's not updated at the same time as stable, though. What's the point of having an extra "testing" branch, asking people to use it, then leaving it unmaintained?

At a minimum, all that needs to happen is that testing is snapped to stable-staging before the release announcement post is written, then once that is posted stable-staging can be snapped over to stable.

Or, stable-staging is used as a "backup" for testing, so a snap is made before a large update is brought over from unstable and if there are going to be lots of issues to fix then there's a smaller update set available.

4 Likes

Well, those people are wrong. There have been several security updates on Stable (mostly for Firefox) in the last month.

Stable is updating properly, everything is fine. If you want more frequent updates, switch to Unstable. :man_shrugging:

That's a bit odd to me, too. My guess is that stable-staging was never meant to be used on a working machine, just for testing, i.e. unlike Unstable and Testing, which are fully usable by themselves.

2 Likes

When it's not 'in use' for an extra level of testing, why doesn't it simply mirror stable? Would that be more difficult?

People complaining about lack of stable / security updates etc just use it to watch YouTube I think :smiley:

1 Like

Personally, after a couple of years with Ubuntu, then a longer period with Mint, I'm not so sensitive about it - if it's important, it'll get pushed through. If it's pushed through faster we lose stability. I vote for stability... so I really don't get the fuss.

1 Like

yeah.. After being on Linux for years, I know that for me at least being on stable means I will get whatever updates / software when it is deemed stable no matter what the timeline is.. can be twice in a month or nothing for 2 months. The distro hopping kid I was 10 years ago would have complained about "something released 2 days ago and still not having it." but not anymore.
P.S. I know this is slightly off topic from OP but just my thoughts about people posting in announcement thread

2 Likes

...That's probably only half untrue...! :wink:

Hmm, yes, slightly off topic.

The issue is not about updates all the time (unstable is there for it).
The issue is about : how can I test if latest is ok & how can I contribute.

A dying staging is useless or even dangerous, so the choice seems easily manichean :

  1. terminate stable-staging : it's time consuming & dangerous.
  2. change it in RC : those who choose it know that they can sometimes be guinea pigs and are ok whith it.

I would argue that if your goal is to contribute, the project would get the most value from having you run testing instead of stable-staging. The stability difference between testing and stable is not that much.

If so, Stable-staging is useless....

Forum kindly sponsored by