[HN Gopher] Wyze security incident update
       ___________________________________________________________________
        
       Wyze security incident update
        
       Author : johnkpaul
       Score  : 107 points
       Date   : 2024-02-19 14:24 UTC (8 hours ago)
        
 (HTM) web link (forums.wyze.com)
 (TXT) w3m dump (forums.wyze.com)
        
       | twisteriffic wrote:
       | > The incident was caused by a third-party caching client library
       | that was recently integrated into our system. This client library
       | received unprecedented load conditions caused by devices coming
       | back online all at once. As a result of increased demand, it
       | mixed up device ID and user ID mapping and connected some data to
       | incorrect accounts.
       | 
       | That seems like enough of a line of bullshit to steer me away
       | from ever using wyze.
        
         | pylua wrote:
         | Device id and user ids are non unique?
        
           | mistrial9 wrote:
           | hash collisions?
        
             | avg_dev wrote:
             | This sounds like a pretty decent guess to me. I bet you are
             | right.
        
             | NathanKP wrote:
             | I'd bet it's this, plus something even stupider like
             | hashing a connection timestamp millisecond as the
             | "uniqueness" of the hash. I've seen a lot of terrible code
             | implementations that assume that there will never be two
             | clients connecting in the exact same millisecond
        
         | hn_20591249 wrote:
         | Yeah I would very much likely to know what caching library has
         | a failure mode of returning content for the wrong keys, that
         | seems pretty bad if not a highly suspect explanation
        
         | tgsovlerkhgsel wrote:
         | Do you think the issue was something else? "People randomly see
         | other people's content" is an issue that would immediately make
         | me think _some_ issue with caching is the culprit.
         | 
         | Given their openness in the rest of the communications, I don't
         | see why they would make this part up.
         | 
         | Edit: Of course, I'm also curious what the actual bug was. A
         | discussion below is suggesting several plausible ways (e.g.
         | concurrency issues, insufficient entropy in some key) how a
         | problem could happen under load (although many of these would
         | also lead to the problem happening with less load, just much
         | less often).
        
           | 15457345234 wrote:
           | > I'm also curious what the actual bug was
           | 
           | Hardware. Rowhammer-type effects occurring accidentally under
           | sudden load spikes. The hardware has just got too dense.
           | 
           | (I should clarify this is speculation, but reading the recent
           | article included here on sudo using special maximum-distance
           | bitfields to hold state internally
           | (https://news.ycombinator.com/item?id=39165342)... it must be
           | a problem that's being observed in the wild)
        
             | tgsovlerkhgsel wrote:
             | I can't imagine that happening with a sufficient frequency.
             | A system making such mistakes so often would just be too
             | unstable to keep an uptime >1h.
        
               | 15457345234 wrote:
               | With the 'cattle not pets' mindset that pervades modern
               | development is the lifespan of ephemeral cache VMs that
               | closely monitored? They get spun up and down on demand in
               | most architectures. I can see this being an edge case
               | failure when the system is trying to scale up, the
               | existing VMs are getting absolutely hammered, the
               | hypervisor is trying to start up new ones, memory
               | pressure and iops on the existing ones are maxed out...
               | 
               | It just seems like the most obvious root cause to me, a
               | single bit-flip in a hashed value is going to give you
               | the wrong result data without any other error because the
               | hash value is already essentially heavily compressed,
               | meanwhile the hash table is almost certain to be 100%
               | stored in memory and very heavily accessed from multiple
               | directions in a read/write manner.
        
             | Arcuru wrote:
             | No. That's not how that works.
        
           | twisteriffic wrote:
           | > Do you think the issue was something else?
           | 
           | No, I'm not questioning whether or not it was a caching
           | issue. I'm taking exception to the lack of accountability.
           | They chose the library. They (probably) chose to ignore a
           | documented or common failure mode of caching systems through
           | either poor choice of key or lack of synchronization. They've
           | obviously designed their infrastructure in a way that isn't
           | resilient to its current level of usage (cold start is a
           | normal part of software's lifecycle).
           | 
           | They could have chosen to own that, but instead they blamed
           | everyone else. That's not a sign of a trustworthy service
           | provider.
        
             | quatrefoil wrote:
             | It's not even that: the quoted language doesn't even blame
             | the library - it appears to blame increased load.
             | 
             | "As a result of increased demand, it mixed up device ID" -
             | no, it mixed up IDs as a result of some sort of a
             | concurrency bug. I don't understand the point of deflecting
             | this far.
        
               | matthewcford wrote:
               | Likely to be a multi-threading issue; my bet is the cache
               | client wasn't thread-safe. I've seen this in some apps
               | before and the solution was to turn off multi-threading
               | while we debug the library that was causing the issue.
        
               | sabujp wrote:
               | This is the answer!
        
         | Twirrim wrote:
         | How so? I've seen caching clients exhibit some really weird
         | behaviour under heavy load. It's not beyond the pale that, eg,
         | the caching library doesn't do proper locking before writing,
         | resulting in writes stomping all over each other.
         | 
         | Caching is normally read heavy, not write heavy, so it's
         | plausible it wouldn't be something you'd see much under typical
         | operation. After an outage, they'd be dealing with a thundering
         | herd level of traffic as everything tries to reconnect, that'd
         | be very different from normal write loads, even different than
         | the write load they'd have seen when they first enabled
         | caching.
        
           | darkwater wrote:
           | Yes but either the library is seriously bugged (like,
           | expecting writes to be ordered and screwing up things if it
           | gets too many writes for different objects at the same time)
           | or there was some serious bug in their implementation. Anyway
           | the attitude and the message passed in the communication
           | seems like handwashing to me. I might be too cynic, though.
        
             | jabart wrote:
             | How else would you say a 3rd party library had a bug under
             | heavy load? 1. You don't want a defamation lawsuit your
             | way. 2. If it was vendor code, you have a contract that may
             | be under a NDA. 3. If it was a vendor, lawyers, lots and
             | lots of lawyers, they likely had to say the minimal amount.
             | The fact they sent out communications for each type of
             | incident in such a short time was great.
        
               | yjftsjthsd-h wrote:
               | The problem is how much they're pointing fingers at the
               | library in the first place.
        
               | dns_snek wrote:
               | I might be splitting hairs, but they say that the
               | incident was "caused by a third party library" when in
               | fact, the incident was caused by insufficient testing on
               | their part.
               | 
               | It sounds like they're trying to shift blame for the
               | incident but then they try to pat themselves on the back
               | for all the effort they put into security. It comes
               | across as dishonest.
               | 
               | Technical details are appreciated but they should've
               | emphasized that this is their own fault. Bonus points if
               | they commit to at least consider E2EE which would
               | sidestep the issue.
        
         | SigmundA wrote:
         | I would at least want to know the client library so I can never
         | use it for anything. Also never trust the client and I hope
         | they don't mean the actual client app such that it can access
         | other user id without server validation...
        
         | woodrowbarlow wrote:
         | sounds like a hashing function with insufficient entropy.
         | "increased demand" would lead to a higher likelihood of hash
         | collisions.
        
           | beala wrote:
           | Not sure I follow. Hash functions don't require entropy, and
           | a hash collision in a hash map shouldn't cause incorrect data
           | to be returned (it just makes them less efficient).
        
             | RandomRandy wrote:
             | I think that they are saying that the output space, i.e.
             | the list of all possible hashes, is too small. Thus, IDs
             | 1234 and 5678 lead to the same hash.
             | 
             | The collision is not in the insertion into the hash map but
             | rather in the look up.
        
         | achille wrote:
         | Same thing happened to OpenAI. Will you steer clear of OpenAI
         | forever as well?
         | 
         | https://news.ycombinator.com/item?id=35294082
        
         | wutwutwat wrote:
         | There are only two hard things in Computer Science: cache
         | invalidation and naming things.
         | 
         | -- Phil Karlton
        
           | pkaeding wrote:
           | And off-by-one errors
        
         | t0mas88 wrote:
         | The whole thing points at everyone but themselves...
         | "Originated at AWS" then "caused by a caching library"
         | 
         | Very little ownership on Wyze's side.
        
         | hamburglar wrote:
         | Coincidentally, I just cancelled my wyze service because the
         | product and support are so terrible. I wanted a simple way to
         | see if there was a package on my doorstep but instead I got
         | something that alerts me when any dog, person, or vehicle goes
         | down my street, and all I've gotten from support is robotic
         | responses suggesting I update my firmware and ignoring my
         | direct questions, running out the clock on my ability to return
         | the thing. At this point I'm not surprised their engineering is
         | bad and amused that it's caused two different security
         | incidents.
        
       | jasongill wrote:
       | > We've identified your Wyze account as one that was affected.
       | This means that thumbnails from your Events were visible in
       | another Wyze user's account and that a thumbnail was tapped. Most
       | taps enlarged the thumbnail, but in some cases it could have
       | caused an Event Video to be viewed.
       | 
       | Kudos to Wyze for doing the things noted in the thread like being
       | honest and prompt with notification etc, but "thumbnails from
       | your Events were visible... and that thumbnail was tapped" is a
       | pretty mealy-mouthed way to say "another person saw your private
       | pictures and videos taken with your Wyze cameras"
        
         | tripdout wrote:
         | Well obviously, and I think people can figure that out.
         | 
         | I actually appreciate the more specific details on how the
         | private pictures and videos were actually viewed using
         | terminology from the Wyze app.
        
       | sneak wrote:
       | Nobody should ever be surprised that sending video to someone
       | else's computer (ie "the cloud") results in third parties viewing
       | that video.
        
         | 1letterunixname wrote:
         | Or buying cheap, no-brand or upstart brand cameras with cloud
         | capabilities.
         | 
         | I had a heck of a time finding a proper POE recording DVR
         | camera system for my mom's house without online or cloud
         | bullshit, but still I isolated it on the network to not take
         | any chances of UPnP port opening or dial-home crap.
         | 
         | The only system I would trust would be one that laid out their
         | security model, source to their apps, and had a self-hosted
         | server DVR option. The captological signals of 99.9% of
         | security system websites do not instill confidence in my mind.
        
           | imglorp wrote:
           | Got any brand/model recommendations? I also want to run full
           | local.
        
             | sponaugle wrote:
             | I have 30 POE IP cameras connected to BlueIris (Running on
             | a server). It records based on activity, and is entirely
             | local (including being on a dedicated VLAN).
             | 
             | The cameras are a wide variety of IP POE cameras which is
             | helpful as I am not stuck to a single brand.
        
         | LeafItAlone wrote:
         | It's 2024. Everything is connected to the internet. Dropbox,
         | Google, and Apple all offer multiple terabyte level plans. The
         | default today is to store in the cloud. We are all storing data
         | in someone else's computer.
         | 
         | Instead of blaming the users, we must hold the companies
         | responsible. Data privacy laws must be stricter and these
         | incidents must be taken more seriously.
        
           | jasonjayr wrote:
           | And if the companies do not want to be responsible, empower
           | the users to run things on their own without lockin.
        
           | sneak wrote:
           | You can indeed make laws to hold the companies responsible,
           | but that's not going to change the situation where it's dumb
           | to store private information on systems that aren't under
           | your control.
           | 
           | You might be able to cause consequences after the fact, but
           | your data will still get leaked first. You can't undo a
           | privacy violation with tort law, and there won't ever be
           | criminal penalties.
           | 
           | If you want your information to stay private, don't store it
           | on other people's computers. IDGAF what "the default today"
           | is. (Also, that's wrong - everyone that is serious and
           | actually wants their data to remain private doesn't store it
           | in the cloud. This is why the CIA got Amazon to build a
           | custom airgapped on-prem AWS region at Langley, for
           | instance.)
        
             | LeafItAlone wrote:
             | A main draw of these "security" cameras is to be able to
             | remotely monitor the locations where they are, including
             | being able to events in the event the devices are stolen.
             | Should I build my own off site, redundant, data centers to
             | make this possible without using a cloud service?
             | 
             | > Also, that's wrong - everyone that is serious and
             | actually wants their data to remain private doesn't store
             | it in the cloud. This is why the CIA got Amazon to build a
             | custom airgapped on-prem AWS region at Langley, for
             | instance.)
             | 
             | Do you think that "everyone that is serious and actually
             | wants their data to remain private" is the default? In a
             | random sampling of 100 people, how many do you think fall
             | into this category?
             | 
             | > This is why the CIA got Amazon to build a custom
             | airgapped on-prem AWS region at Langley, for instance.
             | 
             | Is your threat model the same as Langley's? Or might there
             | just be different levels of what people's needs are?
             | 
             | There's ideals and there's practicality. It's impractical
             | in today's world to completely avoid cloud services. If you
             | can do it, congratulations, more power to you.
        
       | psanford wrote:
       | There's a bunch of things in here that don't really make sense:
       | 
       | > The incident was caused by a third-party caching client library
       | that was recently integrated into our system. This client library
       | received unprecedented load conditions caused by devices coming
       | back online all at once. As a result of increased demand, it
       | mixed up device ID and user ID mapping and connected some data to
       | incorrect accounts.
       | 
       | What? How does load on the system affect correctness?
       | 
       | > The outage originated from our partner AWS
       | 
       | What does this mean? Was there an AWS outage for a service they
       | use, or was this just a normal loss of an instance?
       | 
       | It's interesting that they blame external entities for the root
       | causes of the incident and don't take responsibility for what is
       | ultimately on them.
        
         | 0x0 wrote:
         | > What? How does load on the system affect correctness?
         | 
         | Seen this happen quite often with code that is not multi-thread
         | safe, especially in languages like c# and java, such as using a
         | static class property for data that should be request-scoped,
         | or not using the appropriate concurrent collection classes etc.
        
         | skipkey wrote:
         | It's just a WAG, but I bet someone used a timestamp as a unique
         | key, or at least part of one, so you were unlikely to get
         | collisions except under load.
        
         | CodesInChaos wrote:
         | I assume the code was always incorrect, but only exhibits the
         | problem in practice under high load. This could be a race-
         | condition/data-race, or treating short hashes as unique.
        
       | jabiko wrote:
       | I would have liked them going more into the details of the
       | caching issue. It sounds like they think the cache library was
       | responsible for the issue, but a more technical analysis of what
       | exactly went wrong within that library would be great.
        
       | tgsovlerkhgsel wrote:
       | This is one of the reasons why you want end-to-end encryption
       | wherever possible.
       | 
       | Even a bad implementation with cloud-synced encryption keys
       | (which defeats most of the benefits of e2e) would have stopped
       | this.
       | 
       | The response in this case (notifying customers and specifically
       | stating whether they were affected or not) is excellent, but this
       | seems to be a repeat of a previous incident from September 2023:
       | https://www.theverge.com/2023/9/8/23865255/wyze-security-cam...
        
         | avg_dev wrote:
         | Wow. That is decidedly not a good trend. Once is definitely bad
         | enough, but twice...
        
         | ls65536 wrote:
         | I wonder how many people would continue to so casually use
         | these services if they understood that, for the most part,
         | there is rarely proper end-to-end encryption of their data with
         | these services. It is awfully disingenuous when these
         | companies' marketing materials describe their services as
         | "encrypted" when it usually just means there are two
         | independent TLS pipes, which both terminate in their "cloud";
         | this surely gives a false sense of security to end-users who
         | may not understand the implications of such a setup.
        
           | dns_snek wrote:
           | Too many. I've consulted with friends who installed "smart"
           | security cameras and other IoT devices. I really spelled it
           | out, saying that there's a very real possibility that one day
           | they'll find out someone's been listening in on all of their
           | private conversations (audio) or watching them through their
           | own cameras.
           | 
           | Responses typically range from "I'm not that interesting" to
           | "I really don't care". I think it's too abstract of a threat
           | for most people to take seriously before it happens to them.
        
       | syntaxing wrote:
       | This is one of the things Apple does right. HomeKit working local
       | is a pretty great setup and just works. I put my HomeKit cameras
       | on a VLAN without internet and device isolation and they still
       | work seamlessly. The hard part is getting cameras that are
       | wireless. I use scrypted but even then, getting ONVIF or RTSP
       | isn't as straight forward nowadays. I also have a local frigate
       | backup which works great too. You can pipe in the detection to
       | scrypted with MQTT.
        
         | radredgreen wrote:
         | This may be a good place to introduce my HomeKit based Wyze v3
         | camera reflashing side project. Still a work in progress.
         | 
         | https://github.com/radredgreen/wyrecam
        
       | t8sr wrote:
       | A little off topic, but how is it possible that a tech startup
       | named itself "Wyze" and didn't get sued by Google over the "Waze"
       | trademark? In some accents it sounds exactly the same, and
       | they're sort of in an adjacent product space.
        
         | sqlacid wrote:
         | Am I the only one that saw Wyze and thought Wyse? Guess I'm
         | old.
        
           | t8sr wrote:
           | Right, that's another one. I get that the trademark space is
           | pretty dense, but still - trademarks exist to stop consumers
           | getting confused, and this naming really is confusing?
           | 
           | Waze could come out with a line of dash cams, and then you'd
           | have Waze cams and Wyze cams.
        
         | vel0city wrote:
         | Trademarks are about confusing names in a similar product
         | market. Just having a similar-ish sounding name doesn't mean it
         | violates the trademark. Self-driving cars are pretty different
         | market to home security cameras.
        
           | t8sr wrote:
           | You're thinking of Waymo :) Waze is a crowdsourced traffic &
           | maps thing. It's vaguely, if you squint, adjacent. I don't
           | think it would be completely crazy, for example, for Waze to
           | introduce a line of dash cams.
        
       | firefalcon222 wrote:
       | Not my project but I have had great success with
       | https://github.com/gtxaspec/wz_mini_hacks & V3 model.
       | 
       | The V3 models need to be downgraded to a specific firmware first
       | and patching it exposes RSTP streams using
       | https://github.com/AlexxIT/go2rtc. Everything doable without ever
       | installing Wyze app on an environment air gapped environment with
       | no internet.
        
         | kevinsync wrote:
         | I'm having great success with half a dozen v3's in tandem --
         | for $30 a camera, the quality is really unbeatable -- setup /
         | notes below.
         | 
         | 1. all cameras (firmware v4.36.9.139) have 64gb+ micro SD cards
         | and record to local storage -- many people seem to have issues
         | with anything greater than 32gb in v3's but I've found that
         | this Verbatim tool [0] formats FAT32 at high capacity with no
         | problems
         | 
         | 2. all cameras have wz_mini_hacks [1] on the SD card with RTSP
         | enabled
         | 
         | 3. all cameras are connected via ethernet instead of wifi using
         | this adapter [2] and wz_mini_hacks config
         | 
         | 4. network blocks all outgoing internet connections for all
         | cameras to keep them LAN-only -- this means I have to connect
         | to VPN to review video when outside the house, but I'm cool
         | with that
         | 
         | 5. all RTSP streams are also recorded over the network via
         | Agent DVR [3] to a NAS
         | 
         | 6. the Wyze app (free tier, not paid) works normally with all
         | of the above in place -- I find it much more intuitive to
         | review recent videos in-app (streamed off the SD card), and
         | then review the very occasional older video from a computer off
         | the NAS (scrubbing through in VLC on a computer)
         | 
         | For what it's worth, I don't use them like a Ring camera where
         | you're responding to realtime video events / talking through
         | the camera to a delivery person -- this is mostly just for 24/7
         | recording. I have all object/motion detection events turned
         | off, just a straight uninterrupted feed recording local and on
         | the network.
         | 
         | Links:
         | 
         | 0. https://www.verbatim.com/index/search.php?words=fat32+tool
         | 
         | 1. https://github.com/gtxaspec/wz_mini_hacks
         | 
         | 2. https://www.amazon.com/dp/B07M5X9795
         | 
         | 3. https://www.ispyconnect.com/docs/agent/about
        
           | idiotsecant wrote:
           | I did this for a long time and then realized I was dealing
           | with a lot of bugginess where I'd have to restart cameras,
           | things were wired to wall warts, etc, all to save just a few
           | bucks. I got some cheap poe cameras off Amazon with built in
           | rtsp and its been great.
        
       | uconnectlol wrote:
       | connected cameras have been easily hackable circa 2001 and this
       | isnt changing any time soon especially not with node.js smartcrap
        
       | notatoad wrote:
       | "Don't use Wyze" seems like the wrong takeaway from this.
       | 
       | I'd go with "don't put internet-connected cameras in your house
       | if you don't want those images on the internet". I've got a Wyze
       | in my garage looking over my mountain bikes, and for $35 I don't
       | really care if somebody else sees that image. But I'd never put
       | one in my living space, regardless of their security track
       | record.
        
         | prmoustache wrote:
         | What is the point of filming your mountain bikes? Do you watch
         | them from your office with fondness of your most recent ride?
         | Will that prevent them from being stolen? I doubt so.
        
           | sponaugle wrote:
           | The primary use case for a camera watching something that
           | might get stolen is to provide proof of theft for insurance.
           | In some cases it can provide clear evidence of what exactly
           | was stolen, and in some cases information about who stole it.
           | 
           | Unless the camera has lasers, it isn't very good at stopping
           | said theft. ;)
        
             | prmoustache wrote:
             | Why would your insurance require a proof of the theft? That
             | is not how insurances work usually.
        
               | chrisco255 wrote:
               | Yeah all they require is a police report.
        
               | lopkeny12ko wrote:
               | Of course you need to provide proof of theft. I can't
               | simply go to my insurance, file a claim for a high-value
               | item, and expect them to not ask any more questions. This
               | would be rife with fraud (read: not a reasonable business
               | model) otherwise.
        
               | jt2190 wrote:
               | > Of course you need to provide proof of theft.
               | 
               | Generally filing a police report will suffice, if I'm not
               | mistaken. Could you lie? Of course, but (1) it's
               | generally not in your best interest because the insurer
               | might raise your rates and/or discontinue coverage, and
               | (2) you'd be making a decision to make false statements
               | to the police for purpose of fraud, which most reasonable
               | people won't do.
               | 
               | Other than perhaps using cameras as a means to deter
               | thieves, I'm not sure that low-value (under USD 5000)
               | items like bicycles are worth the time and effort for
               | insurers to launch full investigations over.
        
               | sponaugle wrote:
               | "Other than perhaps using cameras as a means to deter
               | thieves, I'm not sure that low-value (under USD 5000)
               | items like bicycles are worth the time and effort for
               | insurers to launch full investigations over."
               | 
               | I would agree, yet I have seen it happen. It is a
               | somewhat difficult to predict path. I have seen some
               | smaller claims (~$2000) take significantly longer and
               | with more investigation than other very large (>$50,000)
               | claims. I would assume there are certain metrics and
               | algorithmic methods that effect how this plays out, so it
               | is possible that someone who has had multiple small
               | claims my get more attention than a single large claim.
               | 
               | Either way - The concept of the camera is simple in
               | providing evidence in cases where that evidence can help
               | you.
        
               | Cheer2171 wrote:
               | My home insurance requires proof of theft.
               | 
               | Often you use a police report. My local police department
               | is on an unofficial slowdown strike because they don't
               | feel appreciated post-BLM. Possibly also because if they
               | "forget" to file police reports or they "get lost in the
               | system" then the official crime rate goes down. It took
               | me over a month to get a police report I could send to
               | insurance for a simple break in, and spent more of my
               | labor by hourly wage trying to get that report than the
               | cost insurance reimbursed.
        
               | sponaugle wrote:
               | There is tremendous variance with what will be 'required'
               | to complete an insurance claim. In some cases a simple
               | statement will suffice. In others, a police report is
               | sufficient. In some cases the insurance company will want
               | to 'investigate'. In many cases the underwriter will have
               | some additional requests.
               | 
               | However it is always good to recognize the primary goals
               | of the insurance parties - which is to not pay. They have
               | strong incentives in that direction, and you providing
               | proof of theft reduces the opportunities for those
               | incentives to slow or reduce the payment.
               | 
               | It is especially true in cases of undeclared specific
               | items, like an expensive mountain bike that you do not
               | have a dedicated policy on.
        
       | creativeSlumber wrote:
       | >The outage originated from our partner AWS and took down Wyze
       | devices for several hours early Friday morning. ... As we worked
       | to bring cameras back online, we experienced a security issue.
       | Some users reported seeing the wrong thumbnails and Event Videos
       | in their Events tab. ... The incident was caused by a third-party
       | caching client library that was recently integrated into our
       | system. This client library received unprecedented load
       | conditions caused by devices coming back online all at once. As a
       | result of increased demand, it mixed up device ID and user ID
       | mapping and connected some data to incorrect accounts.
       | 
       | As an software engineer who's dealt with caches for large high
       | throughput services, this does not make sense to me why they are
       | blaming a caching client. It's your own code that will decide
       | what is the cache key, and what value to pass as the cache key.
       | Did the caching library have a bug where when you ask for a given
       | key, it returned results for a different key? Or more likely did
       | your own code have a bug where you mixed up the keys? I think we
       | need more details on what went wrong in here.
        
         | albert_e wrote:
         | Lot of blaming others - for architecture, stack, configuration,
         | and operational choices that are likey/should be own decisions
         | that should come with taking ownership.
        
       | darknavi wrote:
       | This is the sort of thing that makes me salty that Unifi Protect
       | is basically cloud locked in. No direct IP connection with
       | "local" account support on the mobile app.
        
         | wredue wrote:
         | I recall seeing setting that this cloud access can be disabled,
         | and you could just VPN to get your "local" only access back.
        
         | zzyzxd wrote:
         | direct IP connection is coming:
         | https://community.ui.com/questions/Unifi-Protect-Mobile-acce...
         | 
         | but even with this, it's still an security camera app that
         | can't send push notification without cloud access.
        
       | matrix_overload wrote:
       | Wyze cameras can actually be used very securely, as long as you
       | bother to jump through some hoops.
       | 
       | First of all, google "Wyze RTSP firmware". It's the official
       | firmware from the vendor that enables the RTSP protocol. Now you
       | can enable RTSP via the app and give the camera a fixed IP
       | address in your DHCP server.
       | 
       | RTSP is a pretty standard protocol, so you can now view the feed
       | via VNC player, record it 24/7 via ffmpeg, use tools like motion,
       | etc.
       | 
       | The camera will still try to connect to cloud, but you can move
       | it to a local-only Wi-Fi network, or outright block it from
       | reaching the outer world on the router side.
       | 
       | And if you want advanced stuff (multiple streams, organized
       | recording, etc), there is a plethora of free/open-source security
       | camera tools (iSpy for instance). It all takes time to learn and
       | configure, but you can have your own fully closed-circuit
       | surveillance network, while still using the Wyze's rather cheap
       | hardware.
        
         | Shadowmist wrote:
         | https://forums.wyze.com/t/how-to-download-wyze-rtsp-files-th...
        
           | BuildTheRobots wrote:
           | > Wyze Update 04/05/2022: RTSP was considered a beta feature
           | and we are currently assessing the path forward as the
           | firmware versions have aged quite a bit. Wyze has removed the
           | firmware files for these versions for now and will update the
           | pages when plans are finalized. Please note that firmware
           | files take a while to work on and test so you may not see an
           | update in the near future. Wyze apologizes for the
           | inconvenience.
        
         | sgerenser wrote:
         | Instead of patching, you can also just use PoE cameras that are
         | designed for this use case (local RTSP) and are only a little
         | more expensive than Wyze. I've installed an Amcrest doorbell
         | that works well with Scrypted and HomeKit, and plan on adding
         | some Amcrest cameras like these soon:
         | https://www.amazon.com/dp/B083G9KT4C
        
           | mdaniel wrote:
           | lazyweb: https://github.com/koush/scrypted (regrettably the
           | licensing is "it depends" https://github.com/koush/scrypted/b
           | lob/v0.93.0/LICENSE.md#li... )
           | 
           | and don't overlook that user's other repos, as seems like
           | there are quite a few fun things in there:
           | https://github.com/koush?tab=repositories
        
       | frognumber wrote:
       | In the meantime, Wyze has rolled back RTSP support, where it was
       | possible to use their devices locally:
       | 
       | https://support.wyze.com/hc/en-us/articles/360026245231-Wyze...
       | 
       | A good response to this might be to put it back, and to extend
       | other devices to be dual-use (Wyze Cloud or HA).
        
       | lofaszvanitt wrote:
       | Accidentally mixed the IDs!
        
       | voakbasda wrote:
       | Another in a long line of reasons to avoid low price, off-the-
       | shelf, unauditable, cloud-enabled cameras.
       | 
       | I continue to be amazed that there is not a reasonably priced,
       | open source, audited, local-first solution, which doesn't require
       | a significant personal investment of time to install and
       | maintain.
        
       | n89nanda wrote:
       | Looks like they had a similar issue in 2023.
       | https://www.nytimes.com/wirecutter/blog/wyze-security-breach...
        
       | illusive4080 wrote:
       | My work encountered this same sort of thing after an outage. Our
       | Redis instance or client got confused. If a=b and c=d in our
       | cache, a request for a returned d randomly.
       | 
       | We quickly realized that cache is fast but not infallible. Use
       | proper security on all your resources. Don't rely on UUIDs to
       | obfuscate your data as security.
        
       | pledess wrote:
       | I'm wondering about the probability that, out of all the affected
       | customers, at least one had the research skills and social skills
       | to identify another customer and successfully ask to meet. Like,
       | for an essay about "His schnauzer needed a mom. WyZettle: the
       | amazing story of a pivot from a home camera service to a dating
       | app."
        
       | spullara wrote:
       | This actually looks like a concurrency bug in their request
       | handling code that may have stored the user id and camera id in
       | shared variables, under load the wrong camera id is seen by a
       | user. At least based on the description of what they say
       | happened.
        
         | bhhaskin wrote:
         | Yeah sounds like an issue with garbage collection/freeing up
         | memory.
        
       | fatkam wrote:
       | I have my cameras blocked from the internet and doing backups in
       | the attic.
        
         | _obviously wrote:
         | Are they wifi or closed circuit?
        
           | fatkam wrote:
           | PoE / Ethernet
        
         | starttoaster wrote:
         | Do any clients work with them when set up like this? Can you
         | view your live/recorded camera streams somehow? Do you get
         | little clips/events from the cameras sensing movement? Do you
         | have any other Wyze security gear like the motion sensors or
         | external floodlights?
         | 
         | edit: Looking at another comment posted in this thread it looks
         | like you don't use Wyze cameras at all, you were just
         | commenting about your own non-Wyze setup.
        
       | gerwim wrote:
       | > Wyze blamed "a third-party caching client library that was
       | recently integrated into our system" for the trouble.
       | 
       | Yes, of course. Blame a third party library which was probably
       | created by an open source maintainer instead of testing your own
       | systems.
        
         | summarity wrote:
         | Also "was recently integrated" - by ... magic?
        
         | _obviously wrote:
         | Of course, the reason this keeps happening is the
         | infrastructure is designed to let it happen in certain cases.
         | Notice how they explicitly say, they need to fix it in the
         | front end. They can't fix it in the backend because that would
         | break eavesdropping.
        
       | scubadude wrote:
       | Why do people have cameras inside their house anyway?
        
         | arealaccount wrote:
         | Income
        
         | BugsJustFindMe wrote:
         | To make sure my cat feeder doesn't suddenly stop working and
         | kill my cat while I'm away for a few days. Or other variations
         | of similar.
        
       ___________________________________________________________________
       (page generated 2024-02-19 23:01 UTC)