#sysops

canoodle@nerdpol.ch

One day either JavaScript or AutoComplete will start ww3

which pretty much reads like:

it probably has to do with: https://wiki.mozilla.org/ServerJS/Introduction

(used in firefox and/or thunderbird?)

(checkout this search for more javascript rants X-D)

one day either JavaScript or Autocomplete will cause the 3rd world war:

now imagine Biden and Putin are texting:

Biden: Dear Putin, very sorry what happened in the past.

But no worries, we will invade you for dinner.

Putin: Invade?

Biden: Sorry autocomplete. Invite you for dinner.

#linux #gnu #gnulinux #opensource #administration #sysops #javascript #js #argh #wtf #ahoh

Originally posted at: https://dwaves.de/2022/06/25/one-day-either-javascript-or-autocomplete-will-start-ww3/

canoodle@nerdpol.ch

One day either JavaScript or AutoComplete will start ww3

it probably has to do with: https://wiki.mozilla.org/ServerJS/Introduction

(used in firefox and/or thunderbird?)

one day either JavaScript or Autocomplete will cause the 3rd world war:

now imagine Biden and Putin are texting:

Biden: Dear Putin, very sorry what happened in the past.

But no worries, we will invade you for dinner.

Putin: Invade?

Biden: Sorry autocomplete. Invite you for dinner.

#linux #gnu #gnulinux #opensource #administration #sysops #javascript #js #argh #wtf #ahoh

Originally posted at: https://dwaves.de/2022/06/25/one-day-either-javascript-or-autocomplete-will-start-ww3/

canoodle@nerdpol.ch

Russia's Conti working on exploits for Intel ME BMC AMT IPMI - Intel ME the biggest security fuck up in computing history - sue Intel

“The biggest network security threat today is a remote code execution exploit for Intel’s Management Engine.”

“Every computer with an Intel chipset produced in the last decade would be vulnerable to this exploit, and RCE would give an attacker full control over every aspect of a system.

If you want a metaphor, we are dinosaurs and an Intel ME exploit is an asteroid hurtling towards the Yucatán peninsula.” (https://hackaday.com/tag/intel-me/)

Intel might have installed – over the course of at least a decade (to this day?) a closed source backdoor in your computer’s firmware, that might never receive updates and is hard to remove.

Once this backdoor is fully cracked, everyone (Russia, China and North Korea) can use it.

Having remote control over a server down to the BIOS is a neat feature.

https://dwaves.de/2018/12/03/intel-bmc-java-jviewer-kvm-remote-control-with-linux-and-intel-mainboard-s2600cwr/

Hackers think so too.

Because a firmware is sometimes hard to update.

BIOS-UEFI updates need to be as easy to install than OS updates.

There are even parts of Intel ME that can not be updated at all (yet) because they are encrypted & signed and the system won’t start if they are missing (security by obscurity).

another dramatic way to put it:

The biggest network security threat today is a remote code execution exploit for Intel’s Management Engine.”

“Every computer with an Intel chipset produced in the last decade would be vulnerable to this exploit, and RCE would give an attacker full control over every aspect of a system.

If you want a metaphor, we are dinosaurs and an Intel ME exploit is an asteroid hurtling towards the Yucatán peninsula.” (https://hackaday.com/tag/intel-me/)

Currently the OS must be hacked or USB access (“physical access equals root access”) before it is possible to malware very deeply in the system = having this guy say: “only solution” “shredder mainboard”

But it might be just a matter of time, until new attack vectors are found, that allow exploit over network, maybe even in the security-nightmare language that every browser runs: JavaScript? X-D (just turn it off globally thanks!)

https://www.golem.de/news/conti-ransomware-gruppe-arbeitet-an-exploit-fuer-intel-me-2206-165848.html

(Conti is a hacker group associated with Russia)

https://www.golem.de/news/security-das-intel-me-chaos-kommt-2003-147099.html

https://www.golem.de/news/security-hackern-gelingt-vollzugriff-auf-intel-me-per-usb-1711-131065.html

Intel ME: Will Intel deliver updates? I hope so. Otherwise: Seriously sue intel to put your IT hardware at danger of being destroyed PERMANENTLY.

As it has happened with those KA-SAT satellite modems.

solutions anyone?

Yes multiple hard one’s.

  • try turning AMT BMC Intel ME off in the bios
    • some BIOS even allow to disable Intel ME permanently
  • on some systems BMC can be disabled with a jumper on the motherboard
  • Flash GNU Linux to the BIOS! (CoreBoot, LibreBoot with the Lenovo x60s no problem, with newer notebooks / PCs probably more effort)
  • “Just use AMD”, yes better but according to this video only a partial solution?

intel needs to work with it’s damaged customers to fix this mess

Or be sued for every hack and every ransomware attack and every downtime and every destroyed motherboard.

seriously.

open source backdoors instead of closed source backdoors!

ok ideally no backdoors at all?

“Intel had already found the vulnerability (CVE-2019-0090) itself last year, described only as a privilege escalation and tried to fix a possible attack vector.

According to PT, however, there are probably other attack vectors and the real problem in ROM still remains, as this part cannot be updated.

This is the preliminary culmination of an embarrassment on the part of Intel, which the manufacturer is trying to sell as security.

By now at the latest, Intel should mothball the concept of a proprietary ME and work on the open hardware security chip that Google

“Customers, users and Intel would be the winners.”

translated from: https://www.golem.de/news/security-das-intel-me-chaos-kommt-2003-147099.html

“Together with partners, Google has announced the Open Titan project.

The goal is a completely openly designed chip based on RISC-V, which is to be used as root-of-trust in many different devices.”

https://www.golem.de/news/open-titan-google-startet-oss-projekt-fuer-hardware-security-chip-1911-144816.html

links:

https://www.rapid7.com/blog/post/2013/07/02/a-penetration-testers-guide-to-ipmi/

#linux #gnu #gnulinux #opensource #administration #sysops #cybersec #itsec #cyber #intel #intel-me #intelme

Originally posted at: https://dwaves.de/2022/06/10/russias-conti-working-on-exploits-for-intel-me-bmc-amt-ipmi-intel-me-the-biggest-security-fuck-up-in-computing-history-sue-intel/

canoodle@nerdpol.ch

The Deep Fake Crisis destroying trust in online conferencing - From Fake Klitschko conferences to Fake Tom Cruise videos - "Massive challenge for Europe" - is technology "neutral"?

update: major of Berlin fooled by Klitschko LIVE deep fake video conference (actually not so funny)

  • it would mean: do not trust phone calls, even if the number is shown as “police” “doctor” “friend”
  • trust no video conference
  • what can be trusted?

https://www.youtube.com/watch?v=CDMVaQOvtxU

https://www.youtube.com/watch?v=CDMVaQOvtxU

how deep fakes work:

  • computer program analyzes facial expressions of person A in footage
  • computer program then applies this, to make the facial expressions of person B, look like person A
    • (it is easier for the program to generate good results, if person A and person B look alike, but in the future not even this might need to be necessary)
  • this is great news, for fake news, but bad news, for truth, as it adds to confusion of the masses (land of confusion)
  • that means, for authenticity, other ways to truly identifying a person in a video have to be found

“Massive challenge for Europe”

‘A “massive challenge for Europe” is when citizens are confronted with disinformation through misleading or openly false information. AI could exacerbate the effects of such “hybrid threats”, as such lies and propaganda could be brought to the people en masse “without technical expertise”. Trends in “deepfake technology” also increased concerns here. According to reports, criminals have already taken advantage of such procedures. AI is also likely to play into the hands of “traditional organized crime” and terrorists.”‘

  • ‘5G also poses problems for investigators’
  • ‘more difficult to identify SIM cards and their users and to eavesdrop on or locate them afterwards’
  • ‘mandatory end-to-end encryption could make eavesdropping even more difficult’
  • ‘virtualization of networks with the new standard does not taste good to the law enforcement officers’

auto translated from: https://www.heise.de/newsticker/meldung/KI-5G-und-Krypto-Europol-warnt-vor-disruptiven-Technologien-4547719.html

https://www.heise.de/news/Europol-Deepfakes-drohen-zum-Standardwerkzeug-Krimineller-zu-werden-7069235.html

#linux #gnu #gnulinux #opensource #administration #sysops #fake #fakenews #news #deepfake #berlin #major #video #conference #fakes

Originally posted at: https://dwaves.de/2022/05/01/the-deep-fake-crisis-destroying-trust-in-online-conferencing-from-fake-klitschko-conferences-to-fake-tom-cruise-videos-massive-challenge-for-europe-is-technology-neutral/

canoodle@nerdpol.ch

Deutsche Telekom and LANCOM hardware partnership - how to reach lancom support - VPN IKEv1 vs IKEv2

https://www.amazon.de/LANCOM-884-VoIP-Over-ISDN/dp/B01ERY8A8S/

https://www.amazon.de/LANCOM-884-VoIP-Over-ISDN/dp/B01ERY8A8S/

LANCOM is a major parnter of Deutsche Telekom.

“Würselen, August 26, 2002 – LANCOM Systems GmbH and Deutsche Telekom AG have agreed to cooperate on the sale of LANCOM Systems products.”

https://my.lancom-systems.de/newsroom/presse/archiv/archivansicht/lancom-systems-kooperiert-mit-deutscher-telekom/

that’s how a lot of lancom products end up at Telekom customers.

They could also have partnered with AVM that builds those pretty nice fritzboxes, why not?

just in case anyone wonders, this is NOT the official lancom forum: https://www.lancom-forum.de/

it is a forum run by social incompetent hobbyists and not paid by lancom.

so don’t expect to reach lancom tech support via this forum.

to reach the official lancom support, the user will need to register at LANCOM’s Jira instance here: https://support.lancom-systems.com/servicedesk/customer/portal/3/user/signup

the lancom support has answered… it only took them 5 weeks…

English:

  • Since 2019, IKEv1 has been described as deprecated and insecure by the IETF (Internet Engineering Task Force) and should therefore no longer be used.
  • LANCOM Systems recommends using the current IKEv2 standard instead.
  • The IKEv1 functionality is retained in LANCOM devices and can therefore continue to be used for scenarios with devices without IKEv2 support.
  • However, LANCOM Systems will no longer provide support for the analysis of connection problems with IKEv1 connections.
  • Also, there will be no bug fixes or new features in the firmware for IKEv1.

German:

  • IKEv1 wird seit 2019 durch die IETF (Internet Engineering Task Force) als veraltet (deprecated) und unsicher bezeichnet und sollte daher nicht mehr verwendet werden.
  • LANCOM Systems empfiehlt stattdessen den aktuellen Standard IKEv2 zu verwenden.
  • Die IKEv1 Funktionalität bleibt in LANCOM Geräten erhalten und kann somit weiterhin für Szenarien mit Geräten ohne IKEv2 Unterstützung verwendet werden.
  • LANCOM Systems wird allerdings keinen Support mehr bei der Analyse von Verbindungs-Problemen mit IKEv1-Verbindungen leisten.
  • Auch wird es für IKEv1 keine Fehlerbehebungen oder neue Features in der Firmware geben.
  • In Einzelfällen kann es zu Verbindungsabbrüchen während des Rekeyings kommen.
  • In einem solchen Fall kann es sinnvoll sein, die Lifetimes anzuheben, damit die Verbindungsabbrüche nicht mehr so häufig auftreten.
  • Informationen erhalten Sie in folgenden Artikeln:

#linux #gnu #gnulinux #opensource #administration #sysops #lancom #vpn #telekom #IKEv1 #IKEv2

Originally posted at: https://dwaves.de/2022/05/18/deutsche-telekom-and-lancom-hardware-partnership-how-to-reach-lancom-support-vpn-ikev1-vs-ikev2/

canoodle@nerdpol.ch

Russia's Conti working on exploits for Intel ME BMC AMT - Intel ME the biggest security fuck up in computing history - sue Intel

Intel might have installed – over the course of at least a decade (to this day?) a closed source backdoor in your computer’s firmware, that might never receive updates and is hard to remove.

Once this backdoor is fully cracked, everyone (Russia, China and North Korea) can use it.

Having remote control over a server down to the BIOS is a neat feature.

https://dwaves.de/2018/12/03/intel-bmc-java-jviewer-kvm-remote-control-with-linux-and-intel-mainboard-s2600cwr/

Hackers think so too.

Because a firmware is sometimes hard to update.

BIOS-UEFI updates need to be as easy to install than OS updates.

There are even parts of Intel ME that can not be updated at all (yet) because they are encrypted & signed (security by obscurity).

Currently the OS must be hacked or USB access (“physical access equals root access”) before it is possible to malware very deeply in the system = having this guy say: “only solution” “shredder mainboard”

But it might be just a matter of time, until new attack vectors are found, that allow exploit over network, maybe even in the security-nightmare language that every browser runs: JavaScript? X-D (just turn it off globally thanks!)

https://www.golem.de/news/conti-ransomware-gruppe-arbeitet-an-exploit-fuer-intel-me-2206-165848.html

(Conti is a hacker group associated with Russia)

https://www.golem.de/news/security-das-intel-me-chaos-kommt-2003-147099.html

https://www.golem.de/news/security-hackern-gelingt-vollzugriff-auf-intel-me-per-usb-1711-131065.html

Intel ME: Will Intel deliver updates? I hope so. Otherwise: Seriously sue intel to put your IT hardware at danger of being destroyed PERMANENTLY.

As it has happened with those KA-SAT satellite modems.

solutions anyone?

Yes multiple hard one’s.

  • try turning AMT BMC Intel ME off in the bios
    • some BIOS even allow to disable Intel ME permanently
  • on some systems BMC can be disabled with a jumper on the motherboard
  • Flash GNU Linux to the BIOS! (CoreBoot, LibreBoot with the Lenovo x60s no problem, with newer notebooks / PCs probably more effort)
  • “Just use AMD”, yes better but according to this video only a partial solution?

intel needs to work with it’s damaged customers to fix this mess

Or be sued for every hack and every ransomware attack and every downtime and every destroyed motherboard.

seriously.

open source backdoors instead of closed source backdoors!

ok ideally no backdoors at all?

“Intel had already found the vulnerability (CVE-2019-0090) itself last year, described only as a privilege escalation and tried to fix a possible attack vector.

According to PT, however, there are probably other attack vectors and the real problem in ROM still remains, as this part cannot be updated.

This is the preliminary culmination of an embarrassment on the part of Intel, which the manufacturer is trying to sell as security.

By now at the latest, Intel should mothball the concept of a proprietary ME and work on the open hardware security chip that Google

“Customers, users and Intel would be the winners.”

translated from: https://www.golem.de/news/security-das-intel-me-chaos-kommt-2003-147099.html

“Together with partners, Google has announced the Open Titan project. The goal is a completely openly designed chip based on RISC-V, which is to be used as root-of-trust in many different devices.”

https://www.golem.de/news/open-titan-google-startet-oss-projekt-fuer-hardware-security-chip-1911-144816.html

#linux #gnu #gnulinux #opensource #administration #sysops #cybersec #itsec #cyber #intel #intel-me #intelme

Originally posted at: https://dwaves.de/2022/05/17/russias-conti-working-on-exploits-for-intel-me-bmc-amt-intel-me-the-biggest-security-fuck-up-in-computing-history-sue-intel/

canoodle@nerdpol.ch

rant: system philosophy: IT that ignore UNIX K.I.S.S will die - in Germany there are only one man teams

https://www.youtube.com/watch?v=pGFGD5pj03M

https://www.youtube.com/watch?v=pGFGD5pj03M

team, team, team team…

In Germany, everyone is his/her own team, there are only one-man, one-woman teams.

teams not necessarily help each other

the best “help” you can get from another “team” is “it is somewhere in the (incomplete, outdated, no body cares) dokuwiki”

often they also work against each other

so actually there are no teams

without social skills and development as human being

the only interface that still exists is text and money

“give me that, here the money”

a “helpful” “community” isn’t it?

which reminds me of:

have no problem with that, just please do not raise any expectations that it might be otherwise.

“we are a hip, cool and fun company” (no)

just say it out loud: “we are a boring company, our systems are a grown mess and we are about to lose control, want to fix it?”

well if you read & understood K.I.S.S (simplify, simplify, simplify, standardize, automate) it could be fixed over the course of several years, if not: no.

to ignore K.I.S.S means: leaving behind an badly documented mess, either for others to fix or to die (delete & restart).

Any newcomer’s motivation will die fast, when being stuck without help and with outdated documentation.

Aka: the company is doomed to die in a self made mess. (Don’t tell me i did not warn ya)

Some will ride the rainbow, until it dies.

You just killed your own company and your job, but not in the way McIlroy envisioned it:

“As a programmer, it is your job to put yourself out of business. What you do today can be automated tomorrow.”

Doug McIlroy

(probably quoting JFKennedy)

PS: those guys at ubuntu surely also never have heard of K.I.S.S

the only reason ubuntu is still a thing, is that it is based on K.I.S.S respecting Debian

also: systemd has become “creepware” (it used to do startup, now it also does network config, WHY? WHY? (because it want’s to rule them all)

#linux #gnu #gnulinux #opensource #administration #sysops #team #teams #it #crowd #company #germany #error #mistake #unix #kiss #simplify #simplicity #rant

Originally posted at: https://dwaves.de/2021/11/10/rant-system-philosophy-it-that-ignore-unix-k-i-s-s-will-die-in-germany-there-are-only-one-man-teams/

canoodle@nerdpol.ch

system philosophy: IT that ignore UNIX K.I.S.S will die - in Germany there are only one man teams

https://www.youtube.com/watch?v=pGFGD5pj03M

https://www.youtube.com/watch?v=pGFGD5pj03M

team, team, team team…

In Germany, everyone is his/her own team, there are only one-man, one-woman teams.

teams not necessarily help each other

the best “help” you can get from another “team” is “it is somewhere in the (incomplete, outdated, no body cares) dokuwiki”

often they also work against each other

so actually there are no teams

without social skills and development as human being

the only interface that still exists is text and money

“give me that, here the money”

a “helpful” “community” isn’t it?

which reminds me of:

have no problem with that, just please do not raise any expectations that it might be otherwise.

“we are a hip, cool and fun company” (no)

just say it out loud: “we are a boring company, our systems are a grown mess and we are about to lose control, want to fix it?”

well if you read & understood K.I.S.S (simplify, simplify, simplify, standardize, automate) it could be fixed over the course of several years, if not: no.

to ignore K.I.S.S means: leaving behind an badly documented mess, either for others to fix or to die (delete & restart).

Any newcomer’s motivation will die fast, when being stuck without help and with outdated documentation.

Aka: the company is doomed to die in a self made mess. (Don’t tell me i did not warn ya)

Some will ride the rainbow, until it dies.

You just killed your own company and your job, but not in the way McIlroy envisioned it:

“As a programmer, it is your job to put yourself out of business. What you do today can be automated tomorrow.”

Doug McIlroy

(probably quoting JFKennedy)

#linux #gnu #gnulinux #opensource #administration #sysops #team #teams #it #crowd #company #germany #error #mistake #unix #kiss #simplify #simplicity

Originally posted at: https://dwaves.de/2021/11/10/system-philosophy-it-that-ignore-unix-k-i-s-s-will-die-in-germany-there-are-only-one-man-teams/

canoodle@nerdpol.ch

The Unix philosophy - simple and beautiful (so it "just works")

“The Unix Philosophy in One Lesson”

“All the philosophy really boils down to one iron law, the hallowed ‘KISS principle’ of master engineers everywhere:”

https://homepage.cs.uri.edu/~thenry/resources/unix_art/ch01s07.html

https://homepage.cs.uri.edu/~thenry/resources/unix_art/ch01s07.html

src: https://homepage.cs.uri.edu/~thenry/resources/unix_art/ch01s07.html

“Keep It Simple, Stupid” or “Keep It Super Simple” (less offensive)

The Unix philosophy emphasizes building simple, short, clear, modular, and extensible code that can be easily maintained and repurposed by developers other than its creators.

This is what allows a system to “boom” “it just work” as Mr Jobs puts it.

The Unix philosophy favors composability as opposed to monolithic design.

Later summarized by Peter H. Salus in A Quarter-Century of Unix (1994):[1] This is the Unix philosophy:

would add:

  • Keep it as simple as possible and as complex as ABSOLUTELY necessary (more moving parts = more problems)
  • simplify & agree on common standards & automate:
    • design standards together “open standards”, so they can take care of (almost) all required use cases
  • software minimalism
    • only run as little software on a system (even if it has plenty of resources) as absolutely necessary
    • for example: don’t run full blown desktop guis on servers that do not need them
      • a lot of admins rely on the “html gui” webmin, and virtualmin
        • if that helps to keep a system in well working secure shape, that’s fine but… it is not optimal for two reasons: - less software = always better - at least enhance security with a ssh-tunnel, meaning the port to access web-console (80 or 443) should only be accessible by localhost via ssh-tunnel (if there are zero-day exploits against latest OpenSSH (aka OpenSSH hacked) it’s “all internet offline doomsday” anyway… so this very very critical piece of software needs relentless testing and fast updates/patching) - gui-admins using this software, know “the gui buttons” but not the bash commands - thus: the gui-admins do not really know, what the software is actually doing in the background when they click a button, thus they might be unable to debug the problem, if things fail (that is where this “monitor all logs” bash script comes in handy)
  • genius: “the genius is in control of chaos” (maybe true for a while, but long term it’s deadly wrong)

  • standards:

    • to standardize is a good thing, imagine a world, where every electronic device would have it’s own power plug… nobody could plugin anywhere without adapters… adding adapters increases complexity (fire hazard?) and decreasing simplicity
    • https://dwaves.de/2019/02/26/usb-cable-salad-of-madness-usb-3-0-and-wifi-interference-micro-usb-usb-c-mess-has-finally-arrived/
    • what existing standards are there?
    • can they be adapted? (try to identify & contact the creators)
      • instead of everyone making a new standard -> mess-of-standards (maximum incompatibility to each other)
    • is it really necessary to create a new standard? (e.g. one fine day, someone might decide to use /config for config files and not /etc, because what does etc stand for? (it stands for “all other files that do not fall in the /bin /dev /lib /usr /sbin… category” it stands for “et cetera” (lat. “the rest”))
      • it evolved into “the directory where config files are stored” and everyone knows that, so it has “grown” into a standard
      • it is agreed, that things should be called what they are and the naming of this directory is just bad (not self speaking)
      • but changing that standard would mess up this “grown-into-standard” and get a lot of users and programs confused… worth it?
      • imho THE worst idea is to add another directory and place another network config file, so that there are 2, 3, 4, 5 different places to configure network = complete mess

Malcolm Douglas McIlroy: “Everything was small… and my heart sinks for Linux when I see the size of it. […]

The manual page, which really used to be a manual page, is now a small volume, with a thousand options…

We used to sit around in the Unix Room saying, ‘What can we throw out? Why is there this option?’

It’s often because there is some deficiency in the basic design — you didn’t really hit the right design point.

Instead of adding an option, think about what was forcing you to add that option.” (src: https://en.wikiquote.org/wiki/Doug_McIlroy)

https://archive.org/details/DougMcIlroy_AncestryOfLinux_DLSLUG

Ken Thompson and Dennis Ritchie, key proponents of the Unix philosophy.

“we are trying to make computing as simple as possible – in the late 1960s Dennis Richie and I realized that the then current Operating System where much way too complex – we attempted to reverse this trend by building a small simple operating system on a minicomputer” (Ken Thompson)

“What we wanted to preserve was not just a good programming environment in which to do programming – but a system around which a community could form – fellowship – we knew from experience that the essence of communal computing – as supplied by remote access time sharing systems – is not just to type programs into a terminal instead of a key-punch – but to encourage close communication” (Dennis M. Ritchie)

https://homepage.cs.uri.edu/~thenry/resources/unix_art/ch01s06.html

https://en.wikipedia.org/wiki/Unix_philosophy

“As a programmer, it is your job to put yourself out of business. What you do today can be automated tomorrow.”

Doug McIlroy

Damn this guy is a philosopher.

Working in IT seems to be just like capitalism itself: working to make one’s job obsolete. “great” outlook.

There need to be alternative lifestyles that make sense and are sustainable.

https://ytpak.net/watch?v=JoVQTPbD6UY

[video width=”586″ height=”436″ mp4=”https://dwaves.de/wp-content/uploads/2017/05/Ken-Thompson-and-Dennis-Ritchie-Explain-UNIX-Bell-Labs.mp4″\]\[/video\]

what happens if companies can not agree on standards:

  • probably the best example:

    • Apple and the “one button vs two button mouse”
    • “Steven P. Jobs, once said that one was the obvious answer because that made it impossible to push the wrong button” (src)

      • turns out: two button mouses are “easy enough” for users that used them for more than 3 days straight
      • by now, approx 99% of Apple’s one-button mouses are in the (recycling?)trash and got replaced by at least two-button mouses (what a waste of resources, right?)
      • GNU Linux users would want a 3 button mouse (the middle button is for pasting stuff)
      • Why Apple hates USB?

        • probably: so Apple can sell more power adapters
        • turns out: shipping iphones without adapters while charging the same price makes even more money
        • finally in 2020 (after aprox. 10 years?) they came along and adopted USB-C Macbook ARM M1 CPU

          - they are still refusing to make the iPhone charge via USB-C
          - [![https://dissectiontable.com/best-chargers-iphone-12-pro-mini-max/](https://dissectiontable.com/wp-content/uploads/2020/10/Best-chargers-for-iPhone-12-iphone-12-Pro-12-Mini-iphone-12-pro-max.jpg)](https://dissectiontable.com/best-chargers-iphone-12-pro-mini-max/)
          
                    > <https://dissectiontable.com/best-chargers-iphone-12-pro-mini-max/>
          - so the user STILL in 2022 needs a special Apple cable to plug into “everyone’s else’s” chargers 
                    - it is… 
                                - 1) truly ridiculous
                                - 2) wasting resources (in 2022 the electronic trash recycling quotas are still pretty bad, only [13% to 35%](https://duckduckgo.com/?t=ffab&q=how+much+of+electronic+waste+get%27s+recycled) of disposed electronics becomes new electronics)
                                - 3) #wtf Apple seriously?
          

#linux #gnu #gnulinux #opensource #administration #sysops #unix #philosophy #m #mcilroy #philosophie #torvalds #itsec #cybersecurity #security #kernel #thompson #ritchie #apple #jobs #standards #standard #gnu-linux #simplify #open #source #openstandards

Originally posted at: https://dwaves.de/2017/05/02/the-unix-philosophy-simple-and-beautiful-so-it-just-works/

canoodle@nerdpol.ch

Rant: wordpress & privacy - self hosted instances upload all content to their CDN servers i2.wp.com imho without consent and no possibility to delete?

correct if wrong but…

checkout this picture, just uploaded to this self-hosted wordpress instance recently:

it “suddenly” and without consent it also exists on this server i2.wp.com, the “cool” wordpress CDN, that is supposed to speed up loading time of a blog… well… not this blog eh?

https://i2.wp.com/dwaves.de/wp-content/uploads/2022/05/S21_Ultra_Sample_Test_Photo_RawTherapie_post_processed_20220525_sky.jpg

if the user would have wanted that, instead of going self-hosting, the user could have just spared all the hazzle and used wordpress.org.

so even when the author-user deletes that image from the user’s blog… (just tested) it will be still there on https://i2.wp.com/dwaves.de/wp-content/uploads/2022/05/S21_Ultra_Sample_Test_Photo_RawTherapie_post_processed_20220525_sky.jpg

“great” isn’t it? so much for “control over one’s data”.

“This is happening because you’re using Jetpack’s built-in Image CDN that relies on WordPress.com. If you like to disable that feature go to your Dashboard > Jetpack > Settings > Performance and toggle off the Enable site accelerator option under the Performance & speed section. You can read more about this feature here:

https://jetpack.com/support/site-accelerator/” (src)

no – this blog is NOT using jetpack site accelerator.

also the option described is not available on self hosted wordpress.

#wtf?

wordpress – a giant content “sucking up ur content and storing it forever” machine?

time for alternatives.

if the visitor Firefox -> F12 -> network checks where the parts of this blog are coming from, they are all coming from dwaves.de and not some wordpress CDN for “faster loading time”.

the only external content are the smileys: https://s.w.org/images/core/emoji/14.0.0/svg/1f61c.svg

(those evil smileys probably “report back” to wordpress… how much that article was visited… and they (probably) sell it again… to G*** and M$ and whoever is willing to pay for that data)

…ah there it is:

https://wordpress.org/about/privacy/

https://wordpress.org/about/privacy/

https://wordpress.org/about/privacy/

so this is how wordpress argues, that they have “legitimate interests” to store all pictures uploaded to self-hosted wordpress instances on: ipv4.de-cix.fra.de.as2635.automattic.com (80.81.193.69)

hosted by: https://www.northdata.de/de-CIX+Management+GmbH,+K%C3%B6ln/HRB+51135

associated with:

https://www.northdata.de/deutsche+medienakademie+GmbH,+K%C3%B6ln/HRB+57192

https://www.northdata.de/7P+Business+Solutions+GmbH,+K%C3%B6ln/HRB+65133

#linux #gnu #gnulinux #opensource #administration #sysops #wordpress #privacy #cdn #cms #web #www #html

Originally posted at: https://dwaves.de/2022/06/06/rant-wordpress-privacy-self-hosted-instances-upload-all-content-to-their-cdn-servers-i2-wp-com-imho-without-consent-and-no-possibility-to-delete/

canoodle@nerdpol.ch

so wordpress uploads all content to their CDN servers i2.wp.com - even when self hosted

correct if wrong but... checkout this picture, just uploaded to this self-hosted wordpress instance recently: it "suddenly" and without consent it also exists on this server i2.wp.com, the "cool" wordpress CDN, that is supposed to speed up loading time of a blog...[...]

#linux #gnu #gnulinux #opensource #administration #sysops #wordpress #privacy #cdn #cms #web #www #html

Originally posted at: https://dwaves.de/2022/06/06/so-wordpress-uploads-all-content-to-their-cdn-servers-i2-wp-com-even-when-self-hosted/

canoodle@nerdpol.ch

so wordpress uploads all content to their CDN servers i2.wp.com - even when self hosted

correct if wrong but…

checkout this picture, just uploaded to this self-hosted wordpress instance recently:

it “suddenly” and without consent it also exists on this server i2.wp.com, the “cool” wordpress CDN, that is supposed to speed up loading time of a blog… well… not this blog eh?

https://i2.wp.com/dwaves.de/wp-content/uploads/2022/05/S21_Ultra_Sample_Test_Photo_RawTherapie_post_processed_20220525_sky.jpg

if the user would have wanted that, instead of going self-hosting, the user could have just spared all the hazzle and used wordpress.org.

so even when the author-user deletes that image from the user’s blog… (just tested) it will be still there on https://i2.wp.com/dwaves.de/wp-content/uploads/2022/05/S21_Ultra_Sample_Test_Photo_RawTherapie_post_processed_20220525_sky.jpg

“great” isn’t it? so much for “control over one’s data”.

“This is happening because you’re using Jetpack’s built-in Image CDN that relies on WordPress.com. If you like to disable that feature go to your Dashboard > Jetpack > Settings > Performance and toggle off the Enable site accelerator option under the Performance & speed section. You can read more about this feature here:

https://jetpack.com/support/site-accelerator/” (src)

no – this blog is NOT using jetpack site accelerator.

also the option described is not available on self hosted wordpress.

#wtf?

wordpress – a giant content “sucking up ur content and storing it forever” machine?

time for alternatives.

if the visitor Firefox -> F12 -> network checks where the parts of this blog are coming from, they are all coming from dwaves.de and not some wordpress CDN for “faster loading time”.

the only external content are the smileys: https://s.w.org/images/core/emoji/14.0.0/svg/1f61c.svg

(those evil smileys probably “report back” to wordpress… how much that article was visited… and they (probably) sell it again… to G*** and M$ and whoever is willing to pay for that data)

#linux #gnu #gnulinux #opensource #administration #sysops

Originally posted at: https://dwaves.de/2022/06/06/so-wordpress-uploads-all-content-to-their-cdn-servers-i2-wp-com-even-when-self-hosted/

canoodle@nerdpol.ch

stress vs work

if you are in control of all the details, stress becomes “just a lot of work”

  • because you pay attention to the details, others might not

it also means:

  • if the program was written by you
    • 1) you know what it can do
    • 2) you know what it can not do
    • 3) and you (should) know why it fails in certain cases
  • if the program was not written by you
    • all the 3 above has to be extensively tested to see if a system or program can do all the requried use cases – secure, reliable and fast

stress can mean:

  • too much work

then the solution is: simplify, simplify, simplify (to simplify what can be simplified is ALWAYS a good idea)

it can also mean:

  • not in control of the details
    • meaning: someone else is in control of the details and the quality of care for details is unkown hence the quality of the software is unkown
      • too much unkowns = stress
        • which brings us back to case 1) 2) and 3)

otherwise it is “just a lot of work”

#linux #gnu #gnulinux #opensource #administration #sysops #stress #work #vs

Originally posted at: https://dwaves.de/2022/06/06/stress-vs-work/

canoodle@nerdpol.ch

update 2022-06: hardkernel.com how to install xfce4 on ODROID XU4 can do 1080p Youtube and Super Mario Kart@50fps on 8x ARM7 (32Bit) - basic setup - cpu tuning and monitoring

update: 2022-06

  • by now under xfce4 (+lightdm + firefox) it can do 720p and 1080p Youtube (see below screenshots and video)
  • with passive cooling ARM CPU getting very hot rendering:
    • 720p Youtube video 70C (158F)
    • 1080p Youtube video 80C (176F)
  • active cooling recommended, preferably with a quiet fan (default odroid fan is said to be noisy)

720p

1080p

[video width=”742″ height=”436″ mp4=”https://dwaves.de/wp-content/uploads/2017/12/ODROID\_XU4\_Youtube\_1080p\_Screenshot-at-2022-06\_works\_but\_running\_80C\_hot.mp4″\]\[/video\]

how to install xfce4 on odroid XU4

make sure network connectivity is established:

<span style="color: #00ffff;">su - root</span><span style="color: #00ffff;">apt update</span>
# install basics
<span style="color: #00ffff;">apt install console-data ssh rsync</span>

# did not quiet do it
<span style="color: #00ffff;">apt install armbian-bullseye-desktop-xfce</span>

# should do it
<span style="color: #00ffff;">apt install xfce4 lightdm xorg tango-icon-theme gnome-icon-theme firefox-esr pavucontrol pulseaudio</span>

update: 2019-10: what it does:

  • one can install a very decent Debian and have okay working speed so yes it can be seen as a very energy efficient (5V based) and still usable (definately faster than pi2) desktop system (99% of all packages available
  • clock for performance: one can manually set the CPU to “performance” with the tool: <span style="color: #00ffff;">gksu odroid-cpu-control</span>

or

  <span style="color: #00ffff;">/usr/local/sbin/odroid-cpu-control</span>

(maximum clock speed at all times) but then things get very fast but also 90°C and more hot (very very hot) so not recommended, so better to select “ondemand” and “from 200 Mhz to 1400 or 2000 Mhz
- if one needs maximum performance consider installing active cooling fan and set clock to permanent maximum.

Super Mario Kart (retroarch) can be played pretty smooth (50-60fps at about 60-75°C) with a little cpu tuning.

this blog post update is written on the xu4 Debian Jessie 8 (Game Station Turbo Image, (direct download link) then recommended to download 500MBytes of updates then firefox stable) and Firefox 60.9.0esr (32-bit)

one is pretty new to Kodi (it comes with older version Javis) can’t really say anything about it… have never used it.

odroid@gamestation-turbo:~$ <span style="color: #00ffff;">hostnamectl</span> 
   Static hostname: <a href="https://forum.odroid.com/viewtopic.php?t=7322">gamestation-turbo</a>
         Icon name: computer
           Chassis: n/a
  Operating System: Debian GNU/Linux 8 (jessie)
            Kernel: Linux 3.10.106+
      Architecture: arm

what it does not:

  • it does not 1080p full hd with 100 frames
    • it is okay at 720p but still one is wondering: where is the GPU hardware acceleration? the 8x ARM CPUs seem to do all the work on Firefox playing Youtube on Debian 8 (hd is stuttering)
    • kodi (latest version on android) plays stuff fine
  • netflix.com refused to work (well this guy has managed to even on an older C1)

tried all possible variants (if one got it working let me know)

suspected problem: no Widevine Support? (Google Chrome will drop 32-bit machines but Chromium is still going. (Source))

so this could be the problem… that 32Bit ARM is simply not supported by Widevine?

(Raspberry Pi 3 can run it… the older Raspberries seem too slow for the job)

neither Debian 8 (Jessie) Firefox (60.9 esr) nor Android + NetFlix App (won’t install straight from Play Store because it says “hardware not supported” but even when one exports the App via AppExtractor from a Android Phone and installs it on Android on xu4 Netflix refuses to work 🙁

who needs netflix if one can have Scott Manley’s Youtube Channel about Space stuff? HURRAY 🙂 (one is serious about this… really love it)

the hardware

you can get the ODROID with HDMI (monitor/screen with audio) output:

“The Odroid XU4 comes outfitted with 2GB of DDR3 RAM, and an Samsung Exynos 5422 CPU, the Odroid beat out many SBCs including the ultra-popular Pi 3 in benchmark tests from Love Our Pi.”

(2.1GHz Quad-Core (Cortex®-A15 (32Bit)) + 1.4GHz Quad-Core (Cortex®-A7 (32Bit)))

with active cooling: https://www.pollin.de/p/odroid-xu4-einplatinen-computer-samsung-exynos-5422-2-gb-2x-usb-3-0-810409

with passive cooling: https://www.pollin.de/p/odroid-xu4q-einplatinen-computer-samsung-exynos-5422-2-gb-2x-usb-3-0-810750

for 80€

bootup dmesg output of recent Armbian:

odroid.xu4.dmesg.txt

odroid.xu4.lsusb.txt

recommended addons:

  • a fast microsd card with 64GBytes of storage
  • a USB Hub (will run out of ports (only 3x) fast (mouse, keyboard, wifi stick, joypad1 and joypad2)
  • usb wifi adapter like 148f:5572 Ralink Technology, Corp. RT5572 Wireless Adapter

      <span style="color: #00ffff;">apt-get install firmware-ralink</span>
    

    In Debian 9: Install the firmware-misc-nonfree package:

      <span style="color: #00ffff;">apt install firmware-misc-nonfree</span>
    
  • <span style="color: #00ffff;">lsusb</span> 
    Bus 006 Device 002: ID 0bda:8153 Realtek Semiconductor Corp. 
    Bus 006 Device 001: ID 1d6b:0003 Linux Foundation 3.0 root hub
    Bus 005 Device 001: ID 1d6b:0002 Linux Foundation 2.0 root hub
    Bus 004 Device 002: ID 05e3:0616 Genesys Logic, Inc. 
    Bus 004 Device 001: ID 1d6b:0003 Linux Foundation 3.0 root hub
    Bus 003 Device 006: ID 0079:0011 DragonRise Inc. Gamepad
    Bus 003 Device 005: ID 046d:c00e Logitech, Inc. M-BJ58/M-BJ69 Optical Wheel Mouse
    Bus 003 Device 004: ID 1a40:0101 Terminus Technology Inc. 4-Port HUB
    Bus 003 Device 003: ID 0461:0010 Primax Electronics, Ltd HP Keyboard
    Bus 003 Device 002: ID 05e3:0610 Genesys Logic, Inc. 4-port hub
    Bus 003 Device 001: ID 1d6b:0002 Linux Foundation 2.0 root hub
    Bus 002 Device 001: ID 1d6b:0001 Linux Foundation 1.1 root hub
    Bus 001 Device 002: ID <span style="color: #ff9900;"><strong>148f:5572 Ralink Technology, Corp. <a href="https://www.ebay.de/sch/i.html?_from=R40&_trksid=m570.l1313&_nkw=RT5572+Wireless+Adapter&_sacat=0">RT5572 Wireless Adapter</a></strong></span>
    Bus 001 Device 001: ID 1d6b:0002 Linux Foundation 2.0 root hub
    

lsmod|grep rt28
rt2800usb 17720 0
rt2800lib 75674 1 rt2800usb
rt2x00usb 10723 1 rt2800usb
rt2x00lib 43390 3 rt2x00usb,rt2800lib,rt2800usb
mac80211 567210 3 rt2x00lib,rt2x00usb,rt2800lib

dmesg|grep rt2x
[ 23.250784] [c6] ieee80211 phy0: rt2x00_set_rt: Info - RT chipset 5592, rev 0222 detected
[ 23.286987] [c6] ieee80211 phy0: rt2x00_set_rf: Info - RF chipset 000f detected
[ 24.978846] [c4] ieee80211 phy0: rt2x00lib_request_firmware: Info - Loading firmware file 'rt2870.bin'
[ 25.010855] [c4] ieee80211 phy0: rt2x00lib_request_firmware: Info - Firmware detected - version: 0.29

dmesg|less
# then search for /ieee80211
[ 23.250784] [c6] ieee80211 phy0: rt2x00_set_rt: Info - RT chipset 5592, rev 0222 detected
[ 23.263058] [c5] cfg80211: (57240000 KHz - 63720000 KHz @ 2160000 KHz), (N/A, 0 mBm), (N/A)
[ 23.286987] [c6] ieee80211 phy0: rt2x00_set_rf: Info - RF chipset 000f detected
[ 23.296854] [c6] ieee80211 phy0: Selected rate control algorithm 'minstrel_ht'
[ 23.298225] [c6] usbcore: registered new interface driver rt2800usb
[ 24.958894] [c4] IPv6: ADDRCONF(NETDEV_UP): eth0: link is not ready
[ 24.978846] [c4] ieee80211 phy0: rt2x00lib_request_firmware: Info - Loading firmware file 'rt2870.bin'
[ 25.010855] [c4] ieee80211 phy0: rt2x00lib_request_firmware: Info - Firmware detected - version: 0.29


cpu frequency clock speed mhz monitoring script: (works on intel x86 CPUs as well)
==================================================================================

this will monitor “on demand” changes to cpu clock in real time… (1 sec interval)

vim /scripts/mon\_cpu.sh

while true; do
echo "===== arm based cpu frequency monitoring ====="
sort /proc/cpuinfo | uniq
for d in /sys/devices/system/cpu/cpu* ; do
if [ -f "$d/cpufreq/scaling_cur_freq" ]
then
echo "$d";
cat $d/cpufreq/scaling_cur_freq;
fi
done
sleep 1; clear; done
```

sample output:

===== arm based cpu frequency monitoring =====
/sys/devices/system/cpu/cpu0
300000
/sys/devices/system/cpu/cpu1
300000
/sys/devices/system/cpu/cpu2
800000
/sys/devices/system/cpu/cpu3
800000
/sys/devices/system/cpu/cpu4
600000
/sys/devices/system/cpu/cpu5
600000
/sys/devices/system/cpu/cpu6
600000
/sys/devices/system/cpu/cpu7
600000

what does it look like?

What SoC are you using?

The SoC is a Samsung Exynos5422 Octa.

What GPU does it include?

An ARM Mali-T628 6 Core.

heat

Complex components such as the XU4 processor may reach temperatures as high as 95°C. At high temperatures, the processor will throttle itself and operate slower so that temperatures do not continue to increase.

robotics

The 30-pin GPIO port can be used as GPIO/IRQ/SPI/ADC, and the 12-pin GPIO port can be used as GPIO/I2S/I2C for electronics and robotics. The GPIO pins on an ODROID-XU4 are a great way to interface with physical devices like buttons and LEDs using a lightweight Linux controller. If you’re a C/C++ or Python developer, there’s a useful library called WiringPi that handles interfacing with the pins, which is described in Chapter 4. Note that all of the GPIO ports are 1.8Volt, and the ADC inputs are limited to 1.8Volt. If a sensor or peripheral

needs higher voltage, the GPIO ports may be level-shifted to 3.3V or 5V using the XU4 Level Shifter Shield.

Serial console port

Connecting to a PC gives access to the Linux console. You can monitor the boot process, or to log in to the XU4 to perform root maintenance.

Note that this serial UART uses a 1.8 volt interface, and it is recommended to use the USB-UART module kit available from Hardkernel.

A Molex 5268-04a (2.5mm pitch) is mounted on the PCB, and its mate is Molex 50-37-5043 Wire-to-Board Crimp Housing.

RTC (Real Time Clock) backup battery connector

If you want to add a RTC functions for logging or keeping time when offline, just connect a Lithium coin backup battery (CR2032 or equivalent). All of the RTC circuits are included on the ODROID-XU4 by default. It connects with a Molex 53398-0271 1.25mm pitch Header, Surface Mount, Vertical type (Mate with Molex 51021-0200).

Gigabit Ethernet

The Realtek RTL8211F is a highly intergrated 10/100/1000M Ethernet transceiver that complies with 10Base-T, 100Base-TX, and 1000Base-T IEEE 802.3 standards.

USB MTT hub controller

The Genesys GL3521 is a 2-port, low-power, and configurable

USB 3.0 SuperSpeed hub controller.

documentation

community support forum: https://forum.odroid.com/

https://wiki.odroid.com/odroid-xu4/odroid-xu4

https://wiki.odroid.com/odroid-xu4/application_note/software/headless_setup

https://magazine.odroid.com/wp-content/uploads/odroid-xu4-user-manual.pdf

download images firmware os software

first of all if you want to use the MicroSDCARD you will have to flip the switch to the left:

android 4.4.4

https://wiki.odroid.com/odroid-xu4/os_images/android/android

debian “armbian”

i used this image of stretch (debian9): https://forum.odroid.com/viewtopic.php?f=96&t=30552

https://www.armbian.com/odroid-xu4/

overall the xu4 is more powerful hence a better better/ quicker desktop replacement than raspberry pi 2 (did not test version 3) but it is not perfect.

You can clearly feel a lack of computing power and speed when you fire up gimp to “quickly” edit some screenshots – it works – but even 8x

model name  : ARMv7 Processor rev 3 (v7l)
BogoMIPS    : 36.00

cpuinfo.txt + SANDISK Ultra microSDXC 64 GB MicroSD card as “harddisk” can not compete with i5+SSD.

    • gnome mate works 🙂 (this is what you see screenshot above)
    • LibreOffice works 🙂
    • hdmi video and sound works! (tested with Samsung TV)
    • firefox (could not get youtube hardware encoding to work yet, also not DRM so netflix not working 🙁
      • youtube works
    • thunderbird does not work “Segmentation fault” but mail client evolution does 🙂
    • use retroarch for gaming / game emulation
    • vlc does not work, you can use cvlc /path/video.mp4
    • gimp works 🙂
    • blobby volley works perfectly with 75 FPS
    • SuperTux2 works
    • chromium does not work:
      • chromium –version
        Chromium 70.0.3538.110 built on Debian 9.6, running on Debian 9.6
      • here is the log: chromium crash.txt
  • webcam cheese works!

    • HD Pro Webcam C920 as /devices/platform/soc/soc:usb3-0/12000000.dwc3/xhci-hcd.3.auto/usb3/3-1/3-1.2/3-1.2:1.0/input/input6
  • /etc/apt/sources.list looks like this.

  • list all installed software: dpkg -ldpkg-l-installed-software

<span style="color: #00ffff;">hostnamectl</span> 
Static hostname: odroid
Icon name: computer
Operating System: Debian GNU/Linux 9 (stretch)
Kernel: Linux 4.14.81+
Architecture: arm

<span style="color: #00ffff;">uname -a</span>
Linux odroid 4.14.81+ #1 SMP PREEMPT Fri Nov 16 14:06:58 UTC 2018 armv7l GNU/Linux

games:

some really manage to turn this thing into a major gaming emulator machine 🙂 nice job!

ODROID GameStation Turbo https://magazine.odroid.com/article/os-spotlight-odroid-gamestation-turbo/

https://forum.odroid.com/viewtopic.php?f=98&t=7322

ubuntu

http://com.odroid.com/sigong/blog/blog_list.php?tag=ODROID-XU4

leads to:

ubuntu-16.04.2-mate-odroid-xu4-20170510.img.md5sum2017-05-10 21:1078ubuntu-16.04.2-mate-odroid-xu4-20170510.img.xz2017-05-10 21:141.2Gubuntu-16.04.2-mate-odroid-xu4-20170510.img.xz.md5sum2017-05-10 21:1481ubuntu-16.04.2-minimal-odroid-xu4-20170516.img.md5sum2017-05-17 21:1881ubuntu-16.04.2-minimal-odroid-xu4-20170516.img.xz2017-05-17 21:18257Mubuntu-16.04.2-minimal-odroid-xu4-20170516.img.xz.md5sum2017-05-17 21:1984ubuntu-16.04.3-4.9-mate-odroid-xu4-20171025.img.md5sum2017-10-30 07:2882ubuntu-16.04.3-4.9-mate-odroid-xu4-20171025.img.xz2017-10-30 07:281.2Gubuntu-16.04.3-4.9-mate-odroid-xu4-20171025.img.xz.md5sum2017-10-30 07:2885ubuntu-16.04.3-4.9-minimal-odroid-xu4-20170824.img.md5sum2017-08-24 14:0185ubuntu-16.04.3-4.9-minimal-odroid-xu4-20170824.img.xz2017-08-24 14:03280Mubuntu-16.04.3-4.9-minimal-odroid-xu4-20170824.img.xz.md5sum2017-08-24 14:0388ubuntu-16.04.3-4.14-mate-odroid-xu4-20171212.img.md5sum2017-12-13 15:5983ubuntu-16.04.3-4.14-mate-odroid-xu4-20171212.img.xz2017-12-13 15:591.3Gubuntu-16.04.3-4.14-mate-odroid-xu4-20171212.img.xz.md5sum2017-12-13 15:5986ubuntu-16.04.3-4.14-minimal-odroid-xu4-20171213.img.md5sum2017-12-15 16:1786*ubuntu-16.04.3-4.14-minimal-odroid-xu4-20171213.img.xz*2017-12-15 16:17302Mubuntu-16.04.3-4.14-minimal-odroid-xu4-20171213.img.xz.md5sum2017-12-15 16:1789load it onto sdcard

warning! POTENTIAL DATALOSS! selecting the wrong device can OVERWRITE DATA ON YOUR COMPUTER ATTACHED HARDDISKS!

insert sdcard into your sdcard reader..

<span style="color: #00ffff;">wget http://de.eu.odroid.in/ubuntu_16.04lts/ubuntu-16.04.3-4.14-minimal-odroid-xu4-20171213.img.xz</span>; # download image<span style="color: #00ffff;">
lsblk</span>; # make sure you identify the right hardware. 
NAME MAJ:MIN RM SIZE RO TYPE MOUNTPOINT
sda 8:0 0 119.2G 0 disk 
├─sda1 8:1 0 49G 0 part /projects
├─sda5 8:5 0 67.4G 0 part /
└─sda6 8:6 0 2.9G 0 part [SWAP]
sr0 11:0 1 1024M 0 rom 
<span style="color: #ff6600;">mmcblk0</span> 179:0 0 14.9G 0 disk 
└─mmcblk0p1 179:1 0 14.9G 0 part
# mmcblk0p1 is the first partition on device <span style="color: #ff6600;">mmcblk0 (the sdcard)
</span><span style="color: #00ffff;">apt-get install xz-utils</span>; # install maybe missing unpacking software <span style="color: #00ffff;">md5sum ubuntu-16.04.3-4.14-minimal-odroid-xu4-20171213.img.xz</span>; # check if md5sum matches <span style="color: #00ffff;">unxz ubuntu-16.04.3-4.14-minimal-odroid-xu4-20171213.img.xz</span>; # unpack the image
<span style="color: #00ffff;">umount /dev/mmcblk0p1</span>; # unmount sdcard time
# write directly
<span style="color: #00ffff;">md5sum -c ubuntu-16.04.3-4.14-minimal-odroid-xu4-20171213.img.xz.md5sum</span>; # verify downloaded file has no errors 
<span style="color: #00ff00;">ubuntu-16.04.3-4.14-minimal-odroid-xu4-20171213.img.xz: OK
</span>
# two possible ways to continue
# write directly unpack on the fly<span style="color: #00ffff;">
xzcat /download/folder/ubuntu-16.04.3-4.14-minimal-odroid-xu4-20171213.img.xz | sudo dd bs=41 of=/dev/mmcblk0
watch kill -USR1 $(pgrep ^dd)</span>; # watch progress, it will copy 1.8GByte
<span style="color: #00ffff;">sync</span>; # run this to write disk cache to disk

# or
# unpack first<span style="color: #00ffff;">
unxz ubuntu-16.04.3-4.14-minimal-odroid-xu4-20171213.img.xz</span>; # unpack 1.8GByte
# possibly this tool was called un-xz, if unxz does not work try un-xz?

# umount the sdcard
<span style="color: #00ffff;">umount /dev/mmcblk0*
</span># write the image<span style="color: #00ffff;">
dd bs=1M if=/download/folder/ubuntu-16.04.3-4.14-minimal-odroid-xu4-20171213.img of=/dev/mmcblk0</span>; # write image to sdcard
<span style="color: #00ffff;">watch kill -USR1 $(pgrep ^dd)</span>; # watch progress, it will copy 1.8GByte <span style="color: #00ffff;">
sync</span>; # run this to write disk cache to disk

<span style="color: #ff0000;"># NOTE!</span>
# during first boot it sits for 3-5 minutes on this message: 
<span style="color: #ffff00;"># "random crng init done"</span>
# it then reboots and sits again on this message...
# this is <span style="color: #ffff00;">NORMAL! it does some partition resizing! this takes time.</span>
# (using the full size of the sdcard)
# please wait paitently.
# if one hour later nothing has changed
# and you do not see a login screen
# restart try again...
# or
# either powersupply is insufficient
# or
# sdcard has problems, check it for bad sectors and/or exchange
<span style="color: #00ffff;">umount /dev/mmcblk0*
badblocks -n -v /dev/mmcblk0</span>

setup

depending on if you can connect a monitor or not:

  1. insert micro sdcard…
  2. LAN connect ODROID XU4 to a DHCP-server ((DSL) router/fritzbox with DHCP enabled)
  3. LAN connect PC/Laptop to the same DHCP-server
  4. start wireshark on your laptop and power on ODROID XU4.

networking: dhcp and finding odroid’s ip

can be a little pain in the a….fternoon.

the ubuntu ODROID XU4 image does not come with a fixed IP, so you will have to connect it to some sort of DHCP-server (router) in order to ssh into it, or install the dhcp-server-service under debian 8. (maybe as VirtualBox VM).

you can also use wireshark and to listen to network activity related to the odroid (dhcp-offer) and find it’s IP this way.

you are looking for simething like:

15 2.286036000 <span style="color: #ff0000;">Wibrain_</span>30:cf:9b Broadcast ARP 60 Who has 192.168.0.1? Tell <span style="color: #ff0000;">192.168.1.101</span>

or use nmap network scanner to find it’s ip…

nmap -n -v -p 1-255 -n -sS 192.168.<span style="color: #ff0000;">10</span>.0/24
Initiating SYN Stealth Scan at 22:21
Scanning <span style="color: #ff0000;">192.168.10.10</span> [255 ports]
Discovered open port 22/tcp on 192.168.10.1

modify the above line depending in what kind of subnet your router is operating. (192.168.XXX.XXX)

ssh into it

usr: root / usr: odroid
pwd: odroid

root@odroid:~# hostnamectl 
   Static hostname: odroid
         Icon name: computer
  Operating System: Ubuntu 16.04.3 LTS
            Kernel: Linux 4.14.5-92
      Architecture: arm

CONGRATULATIONS! YOU LOGGED IN SUCCESSFULLY THE FIRST TIME TO YOUR EMBEDDED POWERHOUSE!

fix the ip

to not have to hassle with dhcp servers again one can change the ip to fixed via:

<span style="color: #00ffff;">nano /etc/network/interfaces</span>

fill it with this content:

# The loopback network interface  
auto lo  
iface lo inet loopback  

# The primary network interface  
auto eth0 
iface eth0 inet static  
address 192.168.<span style="color: #ff6600;">0.111</span>
netmask 255.255.255.0
gateway 192.168.<span style="color: #ff6600;">0.1</span>
dns-nameservers 192.168.<span style="color: #ff6600;">0.1 
</span>

modify the orange parts so it fits your LAN settings – then type:

<span style="color: #00ffff;">reboot
</span>

if everything went straight… you should be able to ping the odroid on 0.111 and ssh into it on 0.111

messing with the odroid

the first thing you probably do with a new system is check for updates…

if your network settings are correct your odroid should be able to access the internet.

<span style="color: #00ffff;">ping yahoo.com</span>; # check if internet is working
<span style="color: #00ffff;">apt update</span>; # update package index
Hit:1 http://ports.ubuntu.com/ubuntu-ports xenial InRelease
Hit:2 http://ppa.launchpad.net/saiarcot895/myppa/ubuntu xenial InRelease 
Get:3 http://ports.ubuntu.com/ubuntu-ports xenial-updates InRelease [102 kB] 
Hit:4 http://deb.odroid.in/5422-s xenial InRelease 
Get:5 http://ports.ubuntu.com/ubuntu-ports xenial-backports InRelease [102 kB] 
Get:6 http://ports.ubuntu.com/ubuntu-ports xenial-security InRelease [102 kB]
Get:7 http://ports.ubuntu.com/ubuntu-ports xenial-updates/main Sources [286 kB]
Get:8 http://ports.ubuntu.com/ubuntu-ports xenial-updates/main armhf Packages [587 kB]
Get:9 http://ports.ubuntu.com/ubuntu-ports xenial-updates/universe armhf Packages [499 kB]
Fetched 1678 kB in 2s (667 kB/s) 
Reading package lists... Done

<span style="color: #00ffff;">apt upgrade</span>; # upgrade system
The following packages have been kept back:
 linux-image-xu3
0 upgraded, 0 newly installed, 0 to remove and 1 not upgraded.

<span style="color: #00ffff;">apt install linux-image-xu3</span>; # force upgrade of this package

# what follows is this scary message that one accepts by selecting "No"
<a href="https://dwaves.de/wp-content/uploads/2017/12/odroid-scary-kernel-upgrade-yes-no-question-screen.png"><img alt="" class="alignnone wp-image-12775 size-full" height="246" src="https://dwaves.de/wp-content/uploads/2017/12/odroid-scary-kernel-upgrade-yes-no-question-screen.png" width="1677"></img></a>

# then an <a href="https://dwaves.de/wp-content/uploads/2017/12/apt-install-linux-image-xu3-output.txt">output like this should follow...</a>

<span style="color: #00ffff;">apt autoremove</span>; # remove packages that have become obsolete

<span style="color: #00ffff;">reboot</span>; # reboot now to load new kernel
Connection to 192.168.0.111 closed by remote host.
Connection to 192.168.0.111 closed.
<span style="color: #00ffff;">ping 192.168.0.111</span>; # after round about 30 seconds the odroid xu4 should have rebooted and respond to your pings
PING 192.168.0.111 (192.168.0.111) 56(84) bytes of data.
64 bytes from 192.168.0.111: icmp_seq=27 ttl=64 time=1.12 ms
64 bytes from 192.168.0.111: icmp_seq=28 ttl=64 time=0.513 ms
64 bytes from 192.168.0.111: icmp_seq=29 ttl=64 time=0.480 ms

<span style="color: #00ffff;">apt install htop vim rsync</span>; # install cool software task monitor htop, vim, rsync
<span style="color: #00ffff;">htop</span>; # and start it
<a href="https://dwaves.de/wp-content/uploads/2017/12/Screenshot_odroid-xu4-htop-task-monitoring.jpg"><img alt="" class="alignnone wp-image-12778 size-full" height="552" src="https://dwaves.de/wp-content/uploads/2017/12/Screenshot_odroid-xu4-htop-task-monitoring.jpg" width="1024"></img></a>
# as you can see the 8x CPUs of the odroid xu4 are pretty bored right now

cpu benchmarking:

<span style="color: #00ffff;">apt install sysbench</span>; # let's install them some work
<span style="color: #00ffff;">mkdir /scripts</span>
<span style="color: #00ffff;">vim /scripts/bench_cpu.sh
</span>
#!/bin/bash

NUM_CORES=$(grep -c ^processor /proc/cpuinfo)

echo "============ CPU MIPS and FLOPS"

cat /proc/cpuinfo | grep -ie hardware;
cat /proc/cpuinfo | grep -ie model;
cat /proc/cpuinfo | grep -ie mips;
cat /proc/cpuinfo | grep -ie flops;

echo "============ CPU BENCHMARK"

sysbench --test=cpu --cpu-max-prime=20000 run --num-threads=$NUM_CORES

:wq save and quit or download it.

<span style="color: #00ffff;">wget <a href="https://dwaves.de/wp-content/uploads/2017/05/bench_cpu.sh_.txt">https://dwaves.de/wp-content/uploads/2017/05/bench_cpu.sh_.txt</a></span>; # download script<span style="color: #00ffff;">
mv bench_cpu.sh_.txt bench_cpu.sh</span>; # rename<span style="color: #00ffff;">
chmod +x /scripts/bench_cpu.sh</span>; # make script executable
<span style="color: #00ffff;">/scripts/bench_cpu.sh</span>; # run it

============ CPU MIPS and FLOPS
Hardware : ODROID-XU4
model name : ARMv7 Processor rev 3 (v7l)
model name : ARMv7 Processor rev 3 (v7l)
model name : ARMv7 Processor rev 3 (v7l)
model name : ARMv7 Processor rev 3 (v7l)
model name : ARMv7 Processor rev 3 (v7l)
model name : ARMv7 Processor rev 3 (v7l)
model name : ARMv7 Processor rev 3 (v7l)
model name : ARMv7 Processor rev 3 (v7l)
============ CPU BENCHMARK
sysbench 0.4.12: multi-threaded system evaluation benchmark

Running the test with following options:
Number of threads: 8

Doing CPU performance benchmark

Threads started!
Done.

Maximum prime number checked in CPU test: 20000

Test execution summary:
total time: 37.4554s
total number of events: 10000
total time taken by event execution: 299.4689
per-request statistics:
min: 23.26ms
avg: 29.95ms
max: 44.59ms
approx. 95 percentile: 40.59ms

Threads fairness:
events (avg/stddev): 1250.0000/327.26
execution time (avg/stddev): 37.4336/0.01

so the benchmark ran in 37 seconds. (rerun with the debian jessie 8 turbo game station image and performance governer: execution time (avg/stddev): 56.9216/0.01…)

on an QuadCore Intel(R) Core(TM) i5-4200U CPU @ 1.60GHz the benchmark run in 14.60 seconds.

so the 8x core ODROID X4U is 2.5 times slower than the 4x core i5 at 1.60 GHZ using all available cores.

while doing this

<span style="color: #00ffff;">vim /scripts/cpu_temp.sh</span>
#!/bin/bash

while true ; do

cpu_temp=$(< /sys/class/thermal/thermal_zone0/temp)
cpu_temp=$(($cpu_temp/1000))
echo $cpu_temp C

sleep 1;
clear;

done # executes COMMAND every second

i monitored the temp of the cpu for around 30minutes while running all 8 cores full speed.

the system is PASSIVELY cooled and seems to throttles the CPUs automatically when 80 degrees Celsius are reached.

maximum measured temp was 82 Celsius (179.6 F).

the passively cooled case got warm but not uncomfortably hot. i have heard raspberry had problems with heat.

when the benchmark was over the temps dropped instantly to 53 C.

Introduction Videos:

https://youtu.be/xxl4ORk4SJo

https://www.youtube.com/watch?v=wtqfC9v0xB0

https://www.youtube.com/watch?v=lUchfyTpOjU

Real Time Clock: RTC

just as the Raspberry pi, the odroid needs a battery to keep clock when it is turned off.

https://www.hardkernel.com/main/products/prdt_info.php?g_code=G137508214939

ODROID-HC1 no HDMI but SATA version

there is/was a NAS/SATA variation (headless, no GPU, no HDMI, no screen/monitor) : https://www.pollin.de/p/odroid-hc1-einplatinen-computer-fuer-nas-und-cluster-anwendungen-810766 for 60€.

#linux #gnu #gnulinux #opensource #administration #sysops #hardware #odroid #embedded #desktop #desktopreplacement #mariokart #snes #segamegadrive #genesis #sega #sonic #mario #game #games #retrogames #retroarch #debian #hardkernel

Originally posted at: https://dwaves.de/2017/12/19/update-2022-06-hardkernel-com-how-to-install-xfce4-on-odroid-xu4-can-do-1080p-youtube-and-super-mario-kart50fps-on-8x-arm7-32bit-basic-setup-cpu-tuning-and-monitoring/

canoodle@nerdpol.ch
canoodle@nerdpol.ch

GNU Linux bash - setup and modify basics script

it is part of a scripts package straight after (Debian based) GNU Linux OS setup, usually doing those modifications: vim /scripts/install_basics.sh #!/bin/bash echo "=== install the basic tools needed to work with a system ===" echo "== disable cdrom apt repo =="[...]

#linux #gnu #gnulinux #opensource #administration #sysops #bash #gnu-linux #setup

Originally posted at: https://dwaves.de/2022/03/03/gnu-linux-bash-setup-and-modify-basics-script/

canoodle@nerdpol.ch

GNU Linux bash - get distribution OS version

this might sound trivial, but there are multiple approaches:

# for (theoretically every) Debian, but not every Debian based system:
<span style="color: #00ffff;">cat /etc/debian_version</span> 
11.3

# for Debian 10 and prior
<span style="color: #00ffff;">hostnamectl|grep Op</span>
  Operating System: Debian GNU/Linux 10 (buster)

# for Debian 11 and higher
<span style="color: #00ffff;">lsb_release -d</span>
Description:    Debian GNU/Linux 11 (bullseye)

# get kernel version (sometimes also includes hints to the distribution)
<span style="color: #00ffff;">uname -a</span>
Linux hostname 5.4.106-1-pve #1 SMP PVE 5.4.106-1 (Fri, 19 Mar 2021 11:08:47 +0100) x86_64 GNU/Linux

# yes Debian is running conservative mode on kernel updates
# it has it's pros and cons
<span style="color: #00ffff;">uname -a</span>
Linux hostname 5.10.0-14-amd64 #1 SMP Debian 5.10.113-1 (2022-04-29) x86_64 GNU/Linux

#linux #gnu #gnulinux #opensource #administration #sysops

Originally posted at: https://dwaves.de/2022/06/03/gnu-linux-bash-get-distribution-os-version/

canoodle@nerdpol.ch

Rant: Open Source and the concept of: Release early, release often or publish early & publish often -> continuous development/continuous integration (CD/CI) -> tight loops ok but still - linking to nirvana without redirection & badly written software that everyone uses - another case of - nothing works "ok" - klarer fall von "nichts funktioniert ok"

https://administrator.de/forum/wol-geht-nicht-mit-broadcast-adresse-101944.html

-> it’s catastrophic, when webpages change their url setup…

https://www.heise.de/netze/Wake-on-WAN–/artikel/89304/0

because it will result in

“nothing works” “ok”

this does not have nothing to do with luck, but with:

  1. bad url management:
    • wordpress does an pretty good job there, as whenever the user changes the url (more keywords?) it will also redirect from the older past urls to the new url
      • that is how it is SUPPOSED to be for EVERY website of the (not so) “ethernal” part of the internet called www
  2. elastic search seems to be a very very badly written software that does not do any sort of software quality checks?
    • or maybe it’s wrongful integration? (but maybe it just sucks)
    • why is every developer-user using it?
  • PS: as mankind still ponders and evolves (by making mistakes) how to best deal with computers
    • yes someone said “publish early” & “publish often” (doing this with the blog… also… often too often and too early X-D)
      • or: “Release early, release often” (wiki)
        • “tight feedback loop between developers and testers or users” (wiki) - yeah sure as a developer that might be a good thing, as a user… really doubt it… - there are highly intelligent respected developers that pioneered this concept… it might work for small teams… (of one)
        • “This philosophy was popularized by Eric S. Raymond in his 1997 essay The Cathedral and the Bazaar, where Raymond stated “Release early. Release often. And listen to your customers”.[4]”“This philosophy was originally applied to the development of the Linux kernel and other open-source software, but has also been applied to closed source, commercial software development.””The alternative to the release early, release often philosophy is aiming to provide only polished, bug-free releases.[5] Advocates of RERO question that this would in fact result in higher-quality releases.[4]
      • has this lead to every developer going in the: continuous development/continuous integration direction? (definately sounds like it)
        • it really should be called CD/CI not CI/CD because first comes the development, then the integration (but well hewego: CI/CD@RedHat)
        • still pondering if it’s really a good idea - well if software quality sticks to UNIX principles of K.I.S.S (most do not and have NO IDEA what non-K.I.S.S means for their software-project or company: - it is the difference between: - lost in chaos of complexity = dysfunctionality - vs a lean stream of running smooth software-company - src: https://homepage.cs.uri.edu/~thenry/resources/unix_art/ch01s07.html - plus test-driven development: 100.000 use case checks tested afterwards automatic & semi-automatic & manual - than that probably works (but then that is what needs to be done anyway to ensure good software quality) - plus: maybe a feedback channel that does not de-motivate - always say something positive first - then the critique
        • signal.org is a very cool mobile & desktop messenger (that usually works pretty well) but: - what is already annoying: if updates per program are 100MBytes and more… (always downloads the full thing (signal.org desktop client) no differential updates?)
  • word of advice: never blindly follow “the trends”
    • always think for yourself, “does it make sense”?
      • test it if it works for you, if not, drop it, what’s the point?

imho gotta to do both…

#linux #gnu #gnulinux #opensource #administration #sysops #rant #software #quality #mess #archive #heise #url #urls #redirects #ci-cd #cd-ci #CICD #CDCI #dev #systems #system #company #developers #developer #buckminster #buckminister

Originally posted at: https://dwaves.de/2022/02/03/rant-open-source-and-the-concept-of-release-early-release-often-or-publish-early-publish-often-continuous-development-continuous-integration-cd-ci-tight-loops-ok-but-still-linking-to-n/