Error on auto update

    • Offizieller Beitrag

    why did Linus Torvald bother to integrate atomic Live Patching in the 5.1 kernel?

    Live patching has been around longer than that. But since you ask, that is meant for critical systems that can't afford the downtime (think healthcare). It is also something setup and maintained by experienced Linux admins. Why do you think it is not enabled by default on any Linux distro? And unfortunately, it isn't as good as it sounds. Most of the speculative execution patches still require reboots. And while RedHat and Canonical offer decent support (not really free, five systems for Canonical if you have a home account) for this, you don't get that with Debian.

    omv 7.0.5-1 sandworm | 64 bit | 6.8 proxmox kernel

    plugins :: omvextrasorg 7.0 | kvm 7.0.13 | compose 7.1.4 | k8s 7.1.0-3 | cputemp 7.0.1 | mergerfs 7.0.4


    omv-extras.org plugins source code and issue tracker - github - changelogs


    Please try ctrl-shift-R and read this before posting a question.

    Please put your OMV system details in your signature.
    Please don't PM for support... Too many PMs!

  • just for a study of feasibility, assuming OMV would be build on Ubuntu and a home account with Canonical was created.

    Are the goals:

    1) "no additional maintenance effort is caused for the owner to operate it in a secure way"

    2) "no service disruption due to reboot"


    technically achievable?

    omv 6.9.6-2 (Shaitan) on RPi CM4/4GB with 64bit Kernel 6.1.21-v8+

    2x 6TB 3.5'' HDDs (CMR) formatted with ext4 via 2port PCIe SATA card with ASM1061R chipset providing hardware supported RAID1


    omv 6.9.3-1 (Shaitan) on RPi4/4GB with 32bit Kernel 5.10.63 and WittyPi 3 V2 RTC HAT

    2x 3TB 3.5'' HDDs (CMR) formatted with ext4 in Icy Box IB-RD3662-C31 / hardware supported RAID1

    For Read/Write performance of SMB shares hosted on this hardware see forum here

    • Offizieller Beitrag

    technically achievable?

    Nope. Did you read my comment about still having to reboot to protect against speculative execution? There is no such thing as a NAS with no additional maintenance effort. You can only have that if you pay someone else to do the maintenance.


    Can you buy a car that requires no maintenance? I don't understand why someone would buy an OMV box and expect to do no maintenance. I would likely not have a job if computers could run without maintenance.

    omv 7.0.5-1 sandworm | 64 bit | 6.8 proxmox kernel

    plugins :: omvextrasorg 7.0 | kvm 7.0.13 | compose 7.1.4 | k8s 7.1.0-3 | cputemp 7.0.1 | mergerfs 7.0.4


    omv-extras.org plugins source code and issue tracker - github - changelogs


    Please try ctrl-shift-R and read this before posting a question.

    Please put your OMV system details in your signature.
    Please don't PM for support... Too many PMs!

  • having to reboot to protect against speculative execution

    well KernelCare doesn#t share this view https://blog.kernelcare.com/ke…d-spectre-without-reboots



    There is no such thing as a NAS with no additional maintenance effort

    I know, but there are no longer technical limitations that would prevent such a NAS to be build.

    Marketing terms this a "killer" feature. We are talking about a "game changer".


    buy an OMV box and expect to do no maintenance. I would likely not have a job if computers could run without maintenance.

    this IT paradigm is about to become obsolete. I worked for companies that are very close already.

    omv 6.9.6-2 (Shaitan) on RPi CM4/4GB with 64bit Kernel 6.1.21-v8+

    2x 6TB 3.5'' HDDs (CMR) formatted with ext4 via 2port PCIe SATA card with ASM1061R chipset providing hardware supported RAID1


    omv 6.9.3-1 (Shaitan) on RPi4/4GB with 32bit Kernel 5.10.63 and WittyPi 3 V2 RTC HAT

    2x 3TB 3.5'' HDDs (CMR) formatted with ext4 in Icy Box IB-RD3662-C31 / hardware supported RAID1

    For Read/Write performance of SMB shares hosted on this hardware see forum here

    • Offizieller Beitrag

    well KernelCare doesn#t share this view https://blog.kernelcare.com/ke…d-spectre-without-reboots

    I'm glad you research little items I mention as exampes instead of the whole picture. Never heard of kernelcare. Not free from the link and not sure I would trust it. And that is only two of the issues that might require kernel reboots. There have been more. Good luck getting this working on an sbc.

    I know, but there are no longer technical limitations that would prevent such a NAS to be build.

    Marketing terms this a "killer" feature. We are talking about a "game changer".

    Marketing is a bunch of shit. You need hardware that can do this. I've worked with servers with "self healing" ram and a million other high availability features. What you are dreaming of doesn't happen on an OMV budget.


    this IT paradigm is about to become obsolete. I worked for companies that are very close already.

    Sure. And cars that fly are right around the corner. I have worked with computers for 40+ years. I am impressed by how far they have come but they are nowhere near self-maintenance. You must have worked in the PR department because even fully automated plants need people to maintain the automation. And since I spend a lot of time automating things, I'm not too worried.

    omv 7.0.5-1 sandworm | 64 bit | 6.8 proxmox kernel

    plugins :: omvextrasorg 7.0 | kvm 7.0.13 | compose 7.1.4 | k8s 7.1.0-3 | cputemp 7.0.1 | mergerfs 7.0.4


    omv-extras.org plugins source code and issue tracker - github - changelogs


    Please try ctrl-shift-R and read this before posting a question.

    Please put your OMV system details in your signature.
    Please don't PM for support... Too many PMs!

  • many new & unrelated topics have been brought up but the basic "what if" question in #22 is still open

    omv 6.9.6-2 (Shaitan) on RPi CM4/4GB with 64bit Kernel 6.1.21-v8+

    2x 6TB 3.5'' HDDs (CMR) formatted with ext4 via 2port PCIe SATA card with ASM1061R chipset providing hardware supported RAID1


    omv 6.9.3-1 (Shaitan) on RPi4/4GB with 32bit Kernel 5.10.63 and WittyPi 3 V2 RTC HAT

    2x 3TB 3.5'' HDDs (CMR) formatted with ext4 in Icy Box IB-RD3662-C31 / hardware supported RAID1

    For Read/Write performance of SMB shares hosted on this hardware see forum here

    • Offizieller Beitrag

    many new & unrelated topics have been brought up but the basic "what if" question in #22 is still open

    Unrelated topics have been brought up because you ignore my answers. I answered the #22 question. If you choose to ignore my answer then I am done here.

    omv 7.0.5-1 sandworm | 64 bit | 6.8 proxmox kernel

    plugins :: omvextrasorg 7.0 | kvm 7.0.13 | compose 7.1.4 | k8s 7.1.0-3 | cputemp 7.0.1 | mergerfs 7.0.4


    omv-extras.org plugins source code and issue tracker - github - changelogs


    Please try ctrl-shift-R and read this before posting a question.

    Please put your OMV system details in your signature.
    Please don't PM for support... Too many PMs!

  • I've seen LivePatching working and avoiding reboots since 3 years for large scale ORACLE database servers and your view is that reboot are unavoidable.


    The comparison with cars and NAS is invalid, because a NAS doesn't have components with mechanical wear and thermal stress comparable to a combustion engine.

    omv 6.9.6-2 (Shaitan) on RPi CM4/4GB with 64bit Kernel 6.1.21-v8+

    2x 6TB 3.5'' HDDs (CMR) formatted with ext4 via 2port PCIe SATA card with ASM1061R chipset providing hardware supported RAID1


    omv 6.9.3-1 (Shaitan) on RPi4/4GB with 32bit Kernel 5.10.63 and WittyPi 3 V2 RTC HAT

    2x 3TB 3.5'' HDDs (CMR) formatted with ext4 in Icy Box IB-RD3662-C31 / hardware supported RAID1

    For Read/Write performance of SMB shares hosted on this hardware see forum here

    • Offizieller Beitrag

    I've seen LivePatching working and avoiding reboots since 3 years for large scale ORACLE database servers

    Good grief man. Now you're being obtuse. We're talking about "home admin's" who may not understand basic concepts like file and folder permissions. Their use of OMV may be their first exposure to Linux. This has already been explained, previously, in so many words.

    LivePatching is OBVIOUSLY done by EXPERIENCED Linux admins. The clue there is "large scale ORACLE database servers". LivePatching is NOT something that is trusted to novices, with very little or no experience, or something to be attempted on autopilot. Here is a LivePatching -> tutorial for Developers. (Note that a Developer is NOT a novice.)

    _____________________________________________________________

    Going against my own better judgement:

    The comparison with cars and NAS is invalid

    And so is your comparison between large ORACLE database servers, that are managed by professionals, and a NAS package intended for home and small business use, being managed by novices and PC enthusiasts. There's NO comparison in that scenario. Further, your proclivity toward "selective reading", in this thread, is accomplishing nothing. If you chose to read selectively, picking out what supports your case while ignoring the remaining context, further communication is pointless.

    When it comes to security updates:
    A Linux server that is not exposed to the internet has a good security profile. (Meaning no ports are forwarded at the router.) That includes OMV. It is LAN clients AND servers with exposed ports that are a high security risk. Without exposed ports, there are users who are running OMV 2.X servers that are doing fine, because they're running behind a router's firewall.

    Here's the bottom line:
    With manual updates (security included) if a beginner does an update and something goes wrong during or after the update, they'll know the problem is associated with the update. As things are now, if notifications are not enabled or mis-configured, users may not know that a automated security update occurred. If something goes wrong with the update (this thread has demonstrated that it can happen) it may appear to have happened spontaneously. There won't be an association, just oddball errors or functional issues that will leave a novice scratching their head wondering, "what happened?"


    That's when they'll come to the forum looking for support. Since they might not know that the issue came from an update, they won't include that information in the description they give to forum supporters. (And as you noted prior, you'll be no help with that, quote; "Unfortunately no idea.") They may skip the forum altogether and simply dump OMV due to a perceived "instability".

    With the support end of it considered, there is no upside to this. That leaves the task of writing something up, telling novice users how to turn this "new feature" off. "That" is going to require getting novice users on the command line. I'll be frank, I'm not happy about that.

    With that said, this topic has been belabored with pages of text, yet again, so I'm done here.

  • "selective reading"

    interesting is that when I respond to comments made (ie speculative execution) its flagged as "would not trust".

    Also I'm not saying such a change will be a "piece of cake", but instead am looking for potential issues to watch for (minesweep).


    The mentioned "supportability" topic I'd paraphrase as "make history of events visible that lead to an issue and HW details (aka telemetry)" is a good point and could be addressed with a "create support question" script.

    Weeks ago I proposed improvements for this topic but got "shut down" with arguments around "never approached it this way".

    So be it.

    omv 6.9.6-2 (Shaitan) on RPi CM4/4GB with 64bit Kernel 6.1.21-v8+

    2x 6TB 3.5'' HDDs (CMR) formatted with ext4 via 2port PCIe SATA card with ASM1061R chipset providing hardware supported RAID1


    omv 6.9.3-1 (Shaitan) on RPi4/4GB with 32bit Kernel 5.10.63 and WittyPi 3 V2 RTC HAT

    2x 3TB 3.5'' HDDs (CMR) formatted with ext4 in Icy Box IB-RD3662-C31 / hardware supported RAID1

    For Read/Write performance of SMB shares hosted on this hardware see forum here

    2 Mal editiert, zuletzt von mi-hol ()

    • Offizieller Beitrag

    I've seen LivePatching working and avoiding reboots since 3 years for large scale ORACLE database servers and your view is that reboot are unavoidable.

    Do you read anything I type??? I never said live patching doesn't work. The free versions of kernel live patching on Ubuntu need reboots for certain patches. Oracle is a paid product (basically RHEL) with support for live patching that it seems can do that without reboot. Debian does not have this.

    I'm not saying such a change will be a "piece of cake"

    You seem to have no idea what it would take to implement. If it was easy, it would be standard in every distro.


    could be addressed with a "create support question" script.

    Write one.

    omv 7.0.5-1 sandworm | 64 bit | 6.8 proxmox kernel

    plugins :: omvextrasorg 7.0 | kvm 7.0.13 | compose 7.1.4 | k8s 7.1.0-3 | cputemp 7.0.1 | mergerfs 7.0.4


    omv-extras.org plugins source code and issue tracker - github - changelogs


    Please try ctrl-shift-R and read this before posting a question.

    Please put your OMV system details in your signature.
    Please don't PM for support... Too many PMs!

  • Do you read anything I type???

    yes I do, but many times it appears to me that what you had in mind to answer has apparently never made it into writing.

    Did you ask yourself, if such an aggressive tone is helpful to have complex, technical conversations that lead to a conclusion at the end?

    omv 6.9.6-2 (Shaitan) on RPi CM4/4GB with 64bit Kernel 6.1.21-v8+

    2x 6TB 3.5'' HDDs (CMR) formatted with ext4 via 2port PCIe SATA card with ASM1061R chipset providing hardware supported RAID1


    omv 6.9.3-1 (Shaitan) on RPi4/4GB with 32bit Kernel 5.10.63 and WittyPi 3 V2 RTC HAT

    2x 3TB 3.5'' HDDs (CMR) formatted with ext4 in Icy Box IB-RD3662-C31 / hardware supported RAID1

    For Read/Write performance of SMB shares hosted on this hardware see forum here

    • Offizieller Beitrag

    yes I do, but many times it appears to me that what you had in mind to answer has apparently never made it into writing.

    What didn't make it into writing? I said Debian doesn't have live kernel patching and you responded that Oracle has it. Who cares if Oracle has it if OMV doesn't run on Oracle.



    Did you ask yourself, if such an aggressive tone is helpful to have complex, technical conversations that lead to a conclusion at the end?

    If you notice, my tone gets more aggressive caused by irritation of you ignoring my comments, your constant insistence on changing things that aren't broken, and/or asking the same question again.

    omv 7.0.5-1 sandworm | 64 bit | 6.8 proxmox kernel

    plugins :: omvextrasorg 7.0 | kvm 7.0.13 | compose 7.1.4 | k8s 7.1.0-3 | cputemp 7.0.1 | mergerfs 7.0.4


    omv-extras.org plugins source code and issue tracker - github - changelogs


    Please try ctrl-shift-R and read this before posting a question.

    Please put your OMV system details in your signature.
    Please don't PM for support... Too many PMs!

Jetzt mitmachen!

Sie haben noch kein Benutzerkonto auf unserer Seite? Registrieren Sie sich kostenlos und nehmen Sie an unserer Community teil!