Re-visiting update control on the desktop

Thanksss snap auto-refresh, you have break my entire cluster with:
/snap/microk8s/2338/bin/dqlite: symbol lookup error: /snap/microk8s/2338/bin/dqlite: undefined symbol: sqlite3_system_errno

Whyyy, I’m a DevOps Engineer & Linux SysAdmin, i need to do upgrade of stable services before in anothers servers and then if all work, then go to the production one… but not, snap is so high that cant disable the auto-update… Please devs (i think that you dont see the point of the sysadmin point of view and a ton of issue about auto update)

FYI this is the exact problem that got me to this thread as well, but there is absolutely 0 developer will and 0 political will because they don’t see this as a problem… yet put server software (like microk8s) on there, and promote it right-front-and-center in their server installer.

That’s the thing, a dev that never worked as SysAdmin will not see autorefresh as a problem… theoretically is good, because is updating “is great and get security updates”, but practically and in enterprise production, auto-update crash things, whe are humans, and the stable branches has issues too (like this one with dqlite & microk8s), so the advice that everyone get is to put the 127.0.0.1 api.snapcraft.io into the servers hosts file, really??? this is the way for a “production” ready tech??? i think that is a joke… please devs, you’re doing a great job, but don’t think only in your environment, so, let the most configuration options to allow the soft be configured by us the sysadmins too. Because even i can’t change the snap repository URL to even create a caché of packages… ufff really is a mess…

there is:

This is ok for caché you’re right… but is the same snap store…

The thing is… is so hard put a simple config option to:

    • Disable snap autorefresh
      or
    • Pin a snap version

Because: Now how i can tell to the user that a WRONG “stable” microk8s update break a cluster of 10 servers with + 100 services… only because devs think that is “the best” way to take care of productions things update…

This happen to me too with the bird daemon with +10 linux gateway acting as software routers, because snap get auto-refresh… great breaks many of that gateways…

Ok, i known that i can put the 127.0.0.1 thing, but that is for a Microsoft App that i cant touch, but Ubuntu really?? an “OpenSource” thing??? i think you’re thinking about things that you need to leave the user thinking, because you dont known who is using snap, can be a simple desktop user, but can be a professional SysAdmin too and the minimun requirements for an admin for a software is get the control… some friends are tell me that this Ubuntu things are converting to MicroUbuntu things, with this kind of crash and autoupdates i thinking this too…

Ok, frankly here’s the problem:

Anyone on the Canonical payroll who is capable of enabling this, won’t do it, because they have an ever-growing list of other things to do. Getting update control on their to-do list requires going through the Guardians of the Galaxy, and prioritising it above whatever else they’re doing. That means architect / design work.

Anyone who isn’t on the Canonical payroll isn’t motivated to do it, because chances are (see above) whatever implementation they come up with may be rejected, or left in pull request limbo.

I started this thread over a year ago, in the hope someone on the snapd team might pick up on the various hints from the community that forced updates with no off-switch was a frequent, frustrating, pain point. Sadly the issues is blocked. I failed internally to get any movement on this. Unless someone has very strong technical and negotiation skills, this won’t happen any time soon. Sorry.

10 Likes

It’s a feature I hope to support with Project Kebe. The specific manifestation has yet to be determined but it will be on the list.

If you can do this on Ubuntu Core:

We should be able to do it on the desktop. Why not?

1 Like

Because this is what I say… Canonical must listen the community, because if you dont listen your community, you’re Microsoft, so i think Canonical is converting into Micrononical

These kind of statements add nothing to the discussion and convince no-one. It only adds noise which drowns out the valid concerns and the discussion about how to fix it.

Really, really… I have yet to read any real technical impediments that developers are relying on to not enable that functionality. All I read is whether it is better for the user or not, please not all users are dumb and there are many who manage technical services and not having that option is like telling them “You don’t know what you are doing let us deal with it”.
Solutions?? yes are tons of them… why the installer of the operating system allow the user to format the disk if “the user dont known about it”??? pleaaaaase, in the moment that you are a base for doing “things” that “is ready for production” you must listen the users… not are 3 or 4 users, are a lot more that currenty are production managers of servers…

I, like most people in this thread, am not a Canonical employee and have no power about whether or not update control gets changed. We’re on the same side. I’m trying to tell you this kind of approach is very unlikely to result in anything useful.

I am trying to make sure the thread doesn’t spiral down into a reddit-like group of people shouting in a void where any useful discussion is drowned out.

3 Likes

but that’s the thing… many people give to the devs real facts where auto-refresh breaks things, but the devs doesnt give to us a real technical problem to avoid the include of that option to snapd… is a simple option, where the default is true, a normal user will not touch that… but some that need it, then touch it…
So ok, if the user touch that and forgot to update is a User issue, but then if the user found how to disable then end with:

  • Touch /etc/hosts file
  • Install another tool (snap-proxy)
  • Download the snap locally and use --dangerous to install
  • And so on…

So where is the real technical explain of this… if they dont see the point, another lot of people see it, and into that set of people are sysadmin working in critical systems, so who knowns more of their system, the devs of snapd or theirself?

@lfdominguez no one is arguing with you about what you’re saying, it’s how you’re saying it. As @popey hinted at in his response, this isn’t actually a technical problem. Canonical leadership just doesn’t want it, which is why no Canonical engineer will find time for it, and why any non-Canonical engineer will need to be a master negotiator if they want to write it themselves. I’m afraid your style of negotiation will not be convincing.

5 Likes

Agree… thanks and sorry with the thread… but is frustrating this kind of things jejeje, asking anyone has tried to push a PR with a proposal??? because if not when i have a time i will try send one…

I don’t believe so, please feel free! Just remember what you’ll be facing. You must either convince someone to merge something they know their bosses don’t want, or convince the bosses that they want it. Neither will be easy; you will need to be eloquent.

2 Likes

Or fork it and carry a very small set of patches against it. Which is what I’ll have to do / have done for Project Kebe anyway. :slight_smile: But it’s a non-zero amount of work and you have to keep it up-to-date yourself.

5 Likes

I just checked in here to see what the status is and whether it’s improved at all and I find it’s not.

I’m very disappointed in Ubuntu and Snap on this. I feel hugely betrayed. I have not upgraded from 16.04 and I will not until I can set it up do updates only when I am in a situation when it’s possible to do the updates and I initiate the process manually.

In my opinion Snapcraft is being very ugly on this.

I have the same shitty problem with my phone so when I’m home on my shitty Verizon Inet connection my phone has WiFi connection turned off so it does not wedge my bandwidth and use up my quota with updates.

I sure hope someone else has already made this observation:

This whole automatic update controversy indicates that the developers insist on violating the KISS principle. It’s just one of the symptoms of bad engineering, as well as other kinds, of judgment demonstrated by the developers, and probably their bosses, on this thread. Not only do I have to leave Ubuntu because of Snap but because I can no longer trust the Ubuntu corporate judgment.

I will look closer at the Linux Mint distribution and any others which might be without Snap and I will switch away from Ubuntu the very next time I visit a place with decent Internet bandwidth and when I have enough connect time.

This is very easy to fix with Snap. Open the settings, go to the Wifi network settings, click the gear icon and check “Metered connection: has data limits or can incur charges”. Snap will not update anymore when on that network.

image

1 Like