This is AFTER debloating all the MS bs as much as I can.
The amount of MS telemetry is just mindboggling.
While telemetry is bad the problem here is probably that this windows service pings the server but doesn’t get a response because it got stuck in your pihole. So it tries to pings again and again and again and again…
deleted by creator
Is that user unfriendly? In the desired path where the resource isn’t blocked the options are display an error or try again.
When a request goes to PiHole it just gets redirected to a bad address, so the application hasn’t got any idea why it isn’t working. In most cases a programmer just adds a retry with exponential back off and goes on with their work, because there’s no real cost to doing that for analytics or advertising code.
Retry with exponential back off would not produce a large number of requests.
The best solution would be to disable telemetry to the capacity you are capable or is possible without breaking something and block pings as backup to when it’s enabled again with updates and repeat.
Yeah, that’s what I was thinking…
Exponential backoff is baked in most libraries. So probably not.
Sure but if they’re trying to send new requests while the others are still retrying, then it will result in the same thing
Is there a service whit which you can automatially send fake answers to certain events?
Teams no linger does that. It just goes to an error page if it fails to load twice
As someone who has designed and used telemetry systems, I’ll never quite understand the strong aversion some people have to them. Telemetry is what lets me tell my boss “yes people really do use our software this way and we can’t break it” or “90% of crashes happen right after the player uses a grenade”. And despite what some conspiracy theorists would have you believe, telemetry data for software from reputable companies does not get sold or used for marketing purposes. Our lawyers make sure of it, and also make us go through privacy reviews to make sure that data isn’t leaking PII.
To me, telemetry would be like a sofa company wanting to put some cameras in your home to see if you’re using the sofa the way they thought you would. It just feels… off.
“90% of crashes happen right after the player uses a grenade”.
Imo, a simple opt-in crash report gets the job done. Technically it is telemetry, but a crash report is more justified than a “where have you clicked” report.
telemetry data for software from reputable companies does not get sold
There’s just no trust in companies to not sell my data. I cannot trust Microsoft nor Google nor any other company to not sell my data, having seen the shenanigans every single company is willing to pull off to get a cent more a year.
Oddly the “where you clicked” report does drive decisions for updates. We (as a developer) use that information to drive UI decisions and determine which flows are more important and should be more easily available
Yes, and it makes sense that it would, and I’d happily give such data to developers I trust to make the programs I love better.
However, that same click data (maybe with some how-long-have-you-looked-at-this data) can definitely be used to target ads at me, which in my book is not okay. And as I said in my comment before, it’s really difficult (at least for me) to trust companies.
In general I agree, but users should be able to make that decision themselves. I do not understand why you can’t turn off telemetry, when it would be trivial to offer that option and so few users would bother to use it.
users should be able to make that decision themselves.
Agreed
I do not understand why you can’t turn off telemetry,
It should be opt-in, not opt-out.
I agree, it should be opt in. Where I disagree is with how strongly people react against it.
If you buy a car and it only comes with Bluetooth do you have a meltdown? Do you kick, scream, and cry that Big Car is going to steal your data? No that’s fucking ridiculous. But grown ass man children on Lemmy act that way when Microsoft wants to know how many gigs of ram they have.
This is a terrible analogy. First of all, Bluetooth isn’t sending data back to headquarters. It’s a short range wireless connection between your gadgets. It couldn’t even if it wanted to.
Secondly, no one is kicking, screaming, and crying. You said it yourself “that’s fucking ridiculous”, and you’re right, because you just made it up to attack those you disagree with.
Finally, would you be upset if your car sent info on where and how you drive back to Ford? Or do you consider the places you go to be your own private information? Because despite your exaggerations, that’s actually what we’re talking about here. Some people would, and some people wouldn’t. You’re obviously in the latter group. But I don’t consider what I do in my free time to be anyone’s business but my own, and I’m offended that any business thinks it’s entitled to it.
As a programmer: “your data is boring. I am not interested in leveraging this for anything besides getting the service you are using to work as well as possible”
Also me as a programmer: “yo, you don’t need that data, stop asking for it. Ohh, your app is broken because it can’t access permissions? Yeet.”
It’s not about what the programmers want, it’s about the sales and marketing departments. They are the ones who use and abuse that data
As a UI designer, no it’s not just them. We need to know how people are using features to know if we should prioritise or deprioritise work on them and what work we might want to put in them.
Thank you. I’ve personally not worked at a company yet that actually sold this data or moved it outside of our internal systems. It exists purely to drive business decisions
We need to know how people are using features to know if we should prioritise or deprioritise work on them and what work we might want to put in them.
So ask them to take the survey, instead of spying on them.
Biases data towards only users willing to do such things.
They’re able to autocorrect for that, in the way they present the surveys and the type of questions they are asking, they actually have trick questions to check for those kind of things.
As a programmer: “your data is boring. I am not interested in leveraging this for anything besides getting the service you are using to work as well as possible”
Also me as a programmer: “yo, you don’t need that data, stop asking for it. Ohh, your app is broken because it can’t access permissions? Yeet.”
It’s not about the programmers. It’s about the company and the ability to make money off of data they get from you. You should be the one who gets money for your data. Not Microsoft, not Google etc.
Is Microsoft making money off of this particular telemetry data? Maybe not. It should always be opt-in
My company collects a ton of data, but it doesn’t leave our servers. We use it purely to drive internal decisions based on how people actually use the software
My problems with telemetry:
Scope: if you provide a service which is a “wrapper” for doing other things, I do not want you to collect usage data. Example: an entire fucking operating system
Opt-out by default (or completely unable to turn it off) even if the service or software I’m using is paid: I want to have the ability to say no. Communicate properly what you collect when I get access to the service, allow me to say no and don’t hide it in 300 pages long TOSes. I don’t want to become your free UX tester when I already pay for the service.
Telemetry-driven development: I absolutely hate this both as a user and a developer. We see there are thousands of users using a feature, but it’s a low % in general, so lead decides we need to remove it from our product. I know that those x thousand people will be annoyed, and so am I when I’m on the receiving end of this.
Another reason that is not universal but service specific is making decisions that purposefully keep you on the platform, over optimizing the interface for maximizing profit.
Me, after spending an entire day making sure we don’t set any cookies until we get consent and actually need them, while fighting off managers who want to install a spyware X, Y and Z just to track the amount of sales, visiting random ass page that could’ve been entirely replaced by just an image, seeing half-page banner saying “we have already set cookies, serviceworker and all of the trackers because the internet does not work without them” be like: Fuck you, Artemiy, my site works fine even without javascript and no cookie header at all. It’s only yours that shits itself at any mention of privacy.
I don’t want you to know anything about me or my device. Simple as
How about shit breaking because everyone at some point is a bad programmer? Even Apple Music doesn’t work when I walk into the elevator until halfway through presumably because hitting play sets of a bunch of useless blocking network calls for music I have saved locally.
What those calls are, I can’t say for sure. Downloading artwork, license checks or telemetry. I’d venture to guess it’s the latter since music will play with placeholder artwork on a slow connection and license checks aren’t required if the subscription was recently validated (works offline for days).
But who really knows. I never bothered to inspect the traffic. The point is, if a company like Apple is creating such a crummy experience for a function so absurdly basic, you can imagine how easy and prevalent telemetry based user degradation is. Go browse the web with a tracker blocker and tell me it isn’t snappier.
PS: I’m also a programmer and collect error reports. So many developers will forego using connection pools, much less collect data with async api’s.
And let’s not even get into how telemetry is a shit tool that is misused 99.99% of the time and only used to surface popular features that aren’t necessarily good features only because we attach causation to every metric (x feature is highly used, therefore it must be good).
So why not ask, “hey you want to share some telemetry to help us improve the product” then?
It’s what all reputable companies or projects, I am aware of, do.
Do you consider Microsoft a “reputable company?”
Yes, but maybe “reputable” isn’t the right word. Realistically, it’s anyone who would potentially face billions in a class-action lawsuit and could actually afford to pay up without going bankrupt. It’s just not worth the risk to getting a few extra $million to pull in telemetry data to the already expansive list of marketing data they collect and monetize.
For example, I would doubt that Hearthstone (Blizzard, revenue $8.7B) sells their app telemetry data. But I could definitely believe that Hill Climb Racer (Fingersoft, revenue $30M) does, or at least integrates it with ad targeting products.
Microsoft has more to lose than almost any other tech company. They also have more process, legal enforcement, and bureaucracy than most other tech companies.
There’s no fear of a lone engineer moving fast and breaking things at Microsoft. If someone at Microsoft had an idea for how they’d use your data they’d have to pass it through 5 chains of command, 2 tech orgs, and Legal just to begin the process.
How many times have I heard this sort of talk about companies only to eventually learn of their unreported flagrant disregard for the rules…
Sure they probably have some bureaucracy but forgive me if I’m not so confident on their unimpeachabilty.
You are absolutely right. There’s some glaring hacks in Azure to cross tenancies or get into the fabric that they’re completely silent on…
I mean, it does have a reputation…
A few things (and I enable it usually, for the record):
- Not really the user’s job to help you with anything especially related to your boss
- “from reputable companies does not get sold or used for marketing purposes. Our lawyers make sure of it,” fuck man, this made me laugh. Good one
Policies get updated, companies are bought and sold, laws change, and most crucially of all, data gets leaked. It doesn’t matter how airtight your asshole is puckered up or how many isolated networks are involved. It gets leaked. Leave the decision up to users about it and in particular maybe let them have full control of their networks.
It doesn’t matter how airtight your asshole is puckered up or how many isolated networks are involved. It gets leaked.
That’s why our lawyers make us make sure that data is sufficiently anonymized before it’s even put on the wire.
Yes I’m sure the lawyer is personally running the analytics to ensure compliance.
No, I’m the one adding the analytics.
Oh thank God, someone reasonable.
Before, I was lost.
That would be good, considered the user could actually have a choose to opt out the telemetry. Windows don’t ask you about gathering data for telemetry or “other” reasons.
Don’t get fooled by the insane amount of hits, when it’s not reachable it will try over and over
Yup. You need more like a logger to get the actual measure of the data.
Which could in turn make their pc run worse because it’ll constantly be retrying telemetry calls
Except this isn’t 1982 and that computer is running Windows 11, not Multix. The CPU probably has at least 8 cores and the NIC can do at least a gigabit. A network retry is not going to matter.
Depends, if you zoom in, most of the time it’s like 3 calls a second or something similar.
The future:
“Malwarebytes has flagged your entire operating system as potential malware”
Lol… :/
…pay 9.99 to have us do nothing
Lol. They are a shit company. (Microsoft & Malwarebytes)
First time I’ve heard someone say Malwarebytes sucks. Is this more of a “Recent update makes it bad” thing or “The company behind this software is horrible”? Personally I’ve never had a problem with them.
The company behind the software has always been a piece of shit, but their detection rate (true positives) is really good.
Okay so is anyone going to elaborate, or…?
They don’t give a damn about privacy, if I’m not mistaken were once caught distributing “fake” malware to test their software, and had an episode where their database got invaded and they didn’t even bother with a simple “sorry about that”. Also, their service model has some stinky dark design patterns.
Thanks for the reply, I appreciate it!
Ditto. Happy Malwarebytes customer here. Would be very interested to hear the issue.
Edit: disregard. Was raging about AVG.
No worries, panic over! Unless someone else knows something else…
They develope proprietary solutions which makes them a no go in my book
Well blocking these calls obviously inflates the numbers due to retries.
I can concur. MS do not know the concept of retry limitation.
You can even know when I’m working from home just to the number of blocked MS tracking calls in my pihole log 🤣
Devil’s advocate: basically the only proper way to figure out how people are using your product and how you can tweak it to achieve its goal is by firing events and including relevant metadata such as how much time they spent on a screen or how far they scrolled. Telemetry is not necessarily “evil” by default.
The other side of that is that the telemetry data never gives you a “why” of something.
For example, users might spend a long time at a screen because they are thinking about what to do, or they are confused by the options and can’t figure out which option they need.
This is why a QA team coupled with a large amount of beta testers is invaluable and necessary.
Telemetry, in the context of software development and UX design, is either a decision by the misinformed or just an excuse to save costs by axing the Windows QA department.
In reality it’s likely the data is being sold off. But in either case, that’s data Microsoft isn’t entitled to (from a moral/privacy perspective).
For example, users might spend a long time at a screen because they are thinking about
… anything!
what am I gonna eat?
I should remember to feed the bicycle…
who stole my cat btw?
who am I to judge?
who am I?
what’s the meaning of life?
what’s the meaning of finding it?
what’s the meaning of figuring out what the meaning is of finding it???
You forgot about the classic, “Where do you want to go today”
Dammit! I forgot to feed my bicycle last night! No wonder it was at my bedroom door ringing its bell nonstop.
And you recently had your cat stolen because you forgot to lock it before you went into the bike shop to get more food, right?
I just couldn’t get the lock looped properly. Cats do not obey the laws of physics.
Well, that’s your hard lock!!
What is the airspeed velocity of an unladen swallow?
what does unladen swallow even mean?
I replied elsewhere but YES! Telemetry is notorious for causing devs to hyperfocus on shit features due to their high usage. Just because a user is clicking X over Y doesn’t mean Y sucks and X is better. Maybe Y is in their periphery, or camouflaged by the background artwork or worded badly. But hey, since X gets a lot of clicks, it must be good, right?
Telemetry, in the context of software development and UX design, is either a decision by the misinformed or just an excuse to save costs by axing the Windows QA department.
That’s very silly. That’s actually such a ridiculous opinion I’m pretty sure you’ve left out some assumption that would make it make sense.
Telemetry is useful, but there is no accountability on how it’s being used, so ultimately it could be used in bad faith and the average user wouldn’t ever know.
The other side of that is that the telemetry data never gives you a “why” of something.
Focus groups and customer surveys work really well for knowing the “why” of something
I totally agree, but where I have a problem (and I imagine a lot of other users here) is that you can’t fully opt out. You can only set “minimal” tracking but not none.
You can if you have enterprise version.
This is the solution to a lot of this stuff.
Also, Firefox, lemmys beloved browser, sends telemetry by default. You have to dig through menus you didn’t know existed to even find out, and then disable.
Not only to Mozilla, but third parties as well.
Sure it’s scummy, but it’s definitely not hidden. When you open the settings page Data Collection is a top level option
It’s hidden in the fact it’s not presented upon first startup, it never mentions it, and it’s at the very bottom of the settings page.
You have to discover it. And who knows how long you had it enabled before you find it.
It’s interesting. I always get a pop-up asking about opting in for sending telemetry when installing Firefox. It was never hidden or the option selected for me. I opt out and it stays opted out.
Firefox is the lesser of two evils. It turned to shit the moment they took Google’s poisoned money. The money also made the Mozilla org put on airs and think they’re some world-changing UN body or some shit and lose focus on their core business of making web browsers.
Not sure about this. When I installed Firefox, it asked me if I allowed it to collect data and run studies (I answered yes). Also, as far as I remember, I never changed the Marketing Data setting and it was off.
Mozilla is also HIGHLY political, paying large sums of money to hardcore anti-white organizations.
Are these the “anti-white” supporters you’re talking about? https://www.mozilla.org/en-US/about/leadership/
Seems odd that these anti-white warriors would be overwhelmingly white themselves. I’m guessing you think the only competent people they could find to hire were white?
Devil’s advocate: basically the only proper way to figure out how people are using your product
Focus groups and customer surveys work really well.
deleted by creator
Actually, except for the deepest debugging data that only a programmer would want, you’re incorrect. And the conversation wasn’t just about that one minority type of data set.
deleted by creator
Once again, no.
All that information is important for the dev itself sure, but also for the UI people, the UX, the product manager, etc.
I mean, I just retired from a career as a self-employed incorporated UI/UX software developer for Fortune 100 companies, but what the fuck do I know, right?
The data telemetry that you are describing is data overload and ends up being not efficient to know. The truth gets lost in the quantity noise.
You have to study the usage patterns of how people use the software, by actually watching people use the app, and you don’t get that from just some counter in memory counting how many times a certain button was pressed, there’s no gestalt in that data set. Great data for selling to third parties, but not for helping you with the UX of an app.
It’s my professional advice that I feel confident in the opinion I expressed on this matter, learned from literal experience on the ground.
deleted by creator
It’s my professional opinion, as a current software engineer that what I said is my reality.
Are you an expert on the subject being discussed?
I am.
Also, there’s just one reality.
If your telemetry was useless, it’s because it was poorly done,
How, exactly, can telemetry be collected poorly?
You totally ignore my points of collecting that much data becomes ineffective and becomes ‘white noise’, as well as how that data would benefit resale more than it would UX analysis.
what can I say?
You could just move on, Internet Warrior.
You’re trying to tell an expert on the subject we’re discussing that they’re wrong about something that they’re telling you they’re very sure of, from many years of experience.
Not in the slightest unfortunately. Often customers don’t even know what customers want, and the subgroup that actually responds to these aren’t necessarily “average”
Not in the slightest unfortunately. Often customers don’t even know what customers want, and the subgroup that actually responds to these aren’t necessarily “average”
That seems like one hell of a hand waving away the opinion.
You do realize that was used for decades before computer’s and the Internet was a thing, right?
And they do things like blind tests so they get audiences that are average.
Those events are used by all of these services:
https://learn.microsoft.com/en-us/windows/privacy/windows-11-endpoints-non-enterprise-editions
Literally everything, from Windows to Office to OneDrive to Cortana to diagnostics.
All of them(except for Cortana) I use, but barely. Hadn’t had real issue so far so Imma just keep it.
I think it’s horrible and has no place at all in a operating system.
Something is seriously wrong with your Windows 11 install. I have two Windows 11 devices on my network and a Surface Duo 2.
Conviva is “video streaming analytics.” Any site with video content is trying to track who uses it, how long they watch, etc.
It freaks me out that it’s sending what users are watching in real time, that’s exactly Spyware.
I read about some smart tvs computing a hash of what’s on screen to phone home about, in attempt with figuring out what content a user is watching when that content is 3rd party controlled.
We need some privacy laws up in this mother fucker.
Wow, do you have a lot of Amazon devices? That’s a lot of calls to Amazon.
Where are the Conviva calls coming from?
Maybe op is a dev who works with AWS?
I reckon “Device metrics” is more likely to be from Alexa devices; source; am AWS Dev and don’t see many of these
This would be correct. We have at least 7 amazon alexa/fireTV devices and a bunch of other IoT devices with Alexa capability and each of them get used regularly.
The IoT devices are on their own subnet which doesn’t have access to the other subnets. I live with my mom and Alexa devices just make her life way easier. I put in the work to make sure the alexa and IoT devices are as restricted as possible without losing functionality so she can live a bit easier.
Yeah I figured. I can’t(don’t have the will power to) point out the reason for this extremely unusual behavior. I just let it be because no devices are visibly suffering from it though.
What did you use to debloat it?
Join us… become Linux nerd, never look back. Hate that the one or two software you use that has no viable equivalent is either super janky or doesn’t work on wine even though tons of games outperform windows… with the windows build.
Or battle telemetry for several years until you get forced to subscribe to win 12.
Only reason I use Win 11 is a single proprietary DRM software I have to deal with on a daily basis. I find almost everything more comfortable in Linux than Windows. I also don’t play games so it’s honestly painless.
single proprietary DRM software
Which one?
Doesn’t matter, in the end, because it’s the same story for tons of us out there. The company I work at now makes a product that only works on Windows. It’s in most of the power plants in the country. You’ve never heard of us unless you are one of half a dozen people at each power plant. There are thousands and thousands of companies just like mine, cranking out software that only works on Windows.
I think the only thing that will change this trend is the raspberry pi and machines like it. Make it so cheap to equip your employees with a Linux machine that it’s impossible to ignore.
Even then, though, 10 hours of lost productivity a month makes the windows machine the more valuable buy for even a low paid employee.
Doesn’t matter, in the end, because it’s the same story for tons of us out there.
Yeah, actually, kind of does. It helps us determine if you just bsing and/or shilling, or if there’s really a product out there that does what you say, which in that case I would like to know so I can stay away from it.
Also I do get that there’s defined vertical markets that have a small customer base of sales, and they could target a single OS in the development for the product for that vertical market, but then to use that as an example of a problem that the majority has to deal with is not intellectually honest.
I live in South Korea and the software I use is a video drm software strictly only for domestic usage. It’s very intrusive software, even killing all the processes it thinks to have a screen capturing(sharing) ability(discord, teams, obs…) Circumventing these measures will also get me banned which is very bad. There is no way they’re making Linux version too. The market is stupendously small in here.
It’s something that a lot of people have to go through here. AntiDRM/FOSS is just not a thing here at all.
Wow, okay, good to know. I had no idea South Korea was that strict on DRM and such. Learned something new.
Wait till you plug in your cell phone to charge they start calling home like crazy
All hail GrapheneOS
Get a device that is supported by degoogled android systems and get rid of all the garbage. My recommendation is always GrapheneOS on a pixel, just because you can reinstall google play services if you really need to and preserve usability, while massively boosting privacy and security.
Unfortunately the majority of devices supported by degoogled android distros are straight garbage. LIke the pixel line. Best device i’ve seen would be Fairphone 3+, or Volla Phone 22
If you say that you must have never used one. Pixels are objectively good phones by every measure except having an SD card slot, which they don’t.
No microsd, no jack, no dualsim. So basically noone of the features i need. You know what i don’t need from my phone? A lot of processing power. Wtf would i even use that for, “Advanced Calendar”?
That’s fine to prefer those features, but it’s not accurate to call those phones bad. Mine came with a USB-C to 3.5mm headphone jack and USB-C to USB-A adapter and plenty of storage, so all my needs are met with it.
I used blokada on my phone and it blocked around 10k calls per day. I’ve since moved to use Mullvad as an always on VPN and turned all of the blockers on. Haven’t setup a pihole for my home yet because I didn’t have access to the router.
You can put it on the network anyway, you just have to set it as the DNS for each device.
Better to set the DHCP server to hand out the dns rather then set it device by device
10k calls per day…
More than 1 every 10 minutes. Seems legit.
Do you have a number that’s 1 digit away from Dominos Pizza or something?
I think they meant their smartphone “calling home” not an actual phone call
@sheogorath can you confirm?
Yep I meant API calls when saying calls.
Calls as in network requests not phone calls
Telemetry DNS/API calls, not dialer calls.
Yeah, I recently did it for a lab and it was… interesting.
My Ubuntu VM wasn’t particularly great either but it was the one that my uni provided
Ubuntu lost me years ago. I still use the server version for… servers. If you want something rawk sawlid for servers, go Ubuntu. Otherwise, go Pop. Or Elementary. Or Mint.
Don’t like your hand held? Fedora.
Hate yourself? Arch.
Draw your entire personality from knowing what a transistor is? Gentoo.
Like the concept of arch but only slightly hate useself manjaro.
Are paranoid about securiy use qubes
My fire stick does double that lol
My “smart tv” too.
Looking at you Roku… they’re at the top of my list for some reason. I have two TVs.
“bUt ThAt DaTa iS gOiNg tO mAkE ThE pRoDuCt BeTtEr”
Despite your sarcasm, it really is.
I have other ways of disabling those telemetry reports.
O&O shutup 10 and adguard for desktop allow the user to turn off a lot of bullshit.
I stopped using my Pihole because it kept eating SD cards. If that wasn’t an issue would love to be using it still.
I can recommend
for that. It logs to RAM and only writes to the SD card once a day (or more/less, if you choose).
It’s a must while using PiHole imo.
You formatted your links as images. Markdown uses ![…](…) for images, […](…) for links.
I appreciate the links to log2Ram. I’ll try when I get a new pi.
I’ve recently read about how to fix corrupted SD-cards in Raspberry Pis. Corruption can happen when the Pi is unexpectedly losing power during a write to the SD card.
To avoid corruption, you can change the (boot partition of the) SD to read-only, requiring either USB storage (flash/HDD/SSD) or a writable secondary partition on the SD card) if you need to save anything locally. The system itself will run fine without write access. Only your files could be at risk if you lose power mid-write.
You can also configure your system to boot from USB storage instead of SD card. Keeping the system partition read-only is probably still a good idea, if possible in your setup.
Modern versions of raspi-config offer a similar read-only overlayFS functionality out of the box! sudo raspi-config, go to Advanced Options, then enable the Overlay FS: Enable/Disable read-only filesystem feature.
Source: Shane S. @ https://raspberrypi.stackexchange.com/questions/7978/how-can-i-prevent-my-pis-sd-card-from-getting-corrupted-so-often
It wasn’t ever an issue of power or corruption. It was hitting the rewrite limit of the SD card itself. I had it on a UPS with my other network equipment and maintained a 100% uptime other than intentional outages with clean reboots. Literally wrote so much log info the card would die. You could still read the card but writing would no longer function.
The suggestion to do a read-only setup would be optimal, I’d prefer it be ram based though as I wouldn’t have such dire need of logs that losing them during a power cycle would cause me any grief.
I had that issue too. Had to switch to running it on my NAS.
I’m wondering what could have caused that. Uneven power supply maybe? My pihole has been chugging away for years without much more than the occasional update when I remember.
SD cards have a write limit. I don’t know what to tell you other than I probably had more shit being written.
High endurance rated cards are good for that. Don’t risk using the cheap SD cards