Thanks. I hate it.

I remember getting excited over the NES.

That’s pretty exciting, for sure. Given that you can get single slot coolers for the half height variant makes it incredibly versatile. Hopefully that trend continues.

Indeed it does. I’m looking forward to the flex series (I’m specifically waiting on the 140 because I have systems with a low profile requirement), to try to put together some GPU acceleration on my homelab cluster. I need it for transcoding in the short term but in the long term I’m hoping to put up one of those open source, self hosted “cloud” gaming services.

We still do LAN parties and if I can pick up some cheap thin clients, and connect them to a GPU accelerated VDI or something, people wouldn’t have to cart their PC’s over when we have a LAN.

I’d go for something more modest like the A380, since sparkle has a low profile version of it, but the 6G of dedicated video memory gives me pause, since I’d basically have to dedicate one whole GPU per virtual desktop, which isn’t as scalable as I would need. Even putting two users on a single GPU with 6G of memory is kind of a non-starter for me. I’ve used GPUs with 3G of memory, as recently as 2 years ago, and bluntly, it’s not a good experience. So anything less than 4-6G per user is basically rejected right out of the gate. I might pick one up just to test with a single VM in a VDI situation, but long term that’s not going to work.

At the risk of resurrecting a zombie post. I’ll respond.

I’m not sure on the specifics of xcp-ng, since I haven’t run it myself, but, I know proxmox and VMware can both do PCIe pass thru to VMs. Recently L1 techs have done videos on the Intel flex GPUs and their potential with vdi for video rendering (basically for a virtual GPU), which worked excellently. I’m not sure if there’s a large feature gap between the a380 and the flex series, but I suspect not. Given the cost of an A380 it’s probably worth the risk to try it. With all the recent updates for the Intel GPUs which have been increasing performance and stability, the a380 is a solid buy, even if it’s “only” able to be passed through to the VM …

Good luck

As a networking professional, I’ll just say: it gets worse the more you look at it.

I think others have covered most of what you wanted to know, but ask me any follow up questions that might still be lingering.

I didn’t have to read far into the documentation of pi alert to find your issue. Scans and detection is done using ARP scans. ARP or address resolution protocol operates on layer 2. VLANs span layer 3 boundaries, so: layer 2 traffic does not traverse VLANs.

Additional scanning (by pi alert) is complimentary to the ARP scan. Which to me reads like ARP scans always need to work.

The easy solution is to use a trunk port into the system, and set up multiple VLAN sub interfaces on the NIC in the OS to handle each VLAN. Alternatively, give the VM multiple NICs, one for each VLAN you wish to scan.

The bottom line is that the pi alert system needs to have a direct network link into each network that it is trying to monitor.

I do it because I don’t want to run short of IP space.

I’ve worked on networks that are reaching the limit of how many systems they can hold, and I don’t want that to happen, so I intentionally oversize basically every subnet and usually over segregate the traffic. I use a lot of subnets.

They’re not all VLANs, some are on independent switches. What I did for storage in one case is gave a single NIC to the management Network for administration, and the rest connected to a storage subnet with fully dedicated links. I was using the same switch so they were vlanned but it easily could have been done on another switch. The connections from the storage to the compute systems was all done with dedicated links on dedicated NICs, so 100% of the bandwidth was available for the storage connections.

I’m very sensitive to bottlenecks in my layer 2 networks and I don’t want to share bandwidth between a production interface and a storage interface. NICs are cheap. My patience is not.

I like Frakes.

That said, in also like pretty much all Trek. I’m not a big fan of some of the series, but I don’t hate them, nor am I going to go around saying they’re “not trek” because reasons.

I’m looking forward to seeing more from the franchise, and I feel a bit alone in my universal enjoyment of Trek. There’s so many people hating on disco or Picard or whatever… I enjoy all of it.

I also enjoy Star wars and Orville, and Stargate, and pretty much most sci-fi… The only stuff, that’s popular, that I have no opinion on is Babylon 5, mainly because I have not watched any of it. Between that, the og BSG and some of the star wars properties (like the animated shows), I’ve watched almost all of the mainstream sci-fi, and honestly, it’s all pretty damn good.

I really liked how they forced the issue about time travel in disco, where the time machine suit thing wouldn’t go unless she went back to all the points she needed to in order to bring this circumstance to happen. I thought that was spot on. I try to ignore the multitude of time paradoxes in voy, and there are many, but it’s probably my least favorite part of that specific show, too much time shit, and it’s all done very poorly.

Technically the were no trek shows between TOS and TNG, so technically the last trek show without Frakes would be when TOS ended.

It’s technically correct as long as reruns don’t count, and you exclude movies, etc… Hence “show”.

It’s all riding a line of being technically correct.

I’m personally running an SMX2000RM from APC. I added a NMC2 to it for monitoring, because I’m crazy like that. I picked up the SMX2000 because I’m running old enterprise gear for my homelab. I have a half-loaded Dell C6100, a Dell R710, a Dell Powervault NX3200, a Dell R630, and a slew of networking devices, plus some utility systems, including a miniPC running Home Assistant, several Raspberry Pi’s (usually using PoE), a few PoE switches, two gateway/firewalls, modem, at least one Cisco ISR router, and a Synology NAS as part of my setup.

Very quickly, new house, bought the R630 for “home operations” which is running some internal to the home systems, such as a backup DNS, some game servers, and PCNS, all on top of VMware ESXi 6.5 free. I have the c6100 (two nodes), and the R710 running ESXi 6.5 from vMUG advantage, which is running my lab servers, including a windows domain (there’s a domain GC replicant server on the R630), Exchange, several Linux nodes doing multiple things, some windows server based gaming hosts, Plex, netbox, and mediawiki for documentation, as well as a LibreNMS for monitoring. Home Assistant is on it’s own Core i5 mini pc (either Dell or HP, I forget), which runs my new home automation z-wave network, another mini-pc running DNS (I just wanted a dedicated non-VM system for DNS, so I didn’t have to struggle without it while I got VMware working properly after a power loss), and the NAS is for personal backup storage, just a two-bay (8TB drives in RAID 1) for my PC and my SO’s PC.

I didn’t want all of that to go down as soon as there’s a power blip, and the APC SMX2000 was a good fit, considering we’re consuming ~11 Amps on the 20 Amp circuit I installed specifically for the UPS (~1300W). I want to get a second one, and redundant power set up for everything, including networking, on the second UPS (I’ll be buying another SMX2000 for the purpose), with one feeding the primary power port of everything (or PSU1 on the servers) and the other feeding all the backups (or PSU2 on servers), and I want to add battery expansion on all units, for a total of 6 battery packs across all units, and 4000VA of power available, each UPS will have it’s own dedicated 20A receptacle.

The main reason I chose the SMX2000 is that it has a NEMA L5-20R port on the back, which I connected a 0RU vertical PDU (managed, the APC AP7930), which has a NEMA L5-20P connection for the source, and provides 24x NEMA 5-20R ports. The plan is to have two, one for each UPS. I bought the UPS off of Ebay without a battery and picked up a battery from a local supplier (non-APC battery, but a drop-in replacement, with all the APC fittings).

Specifics:

UPS: APC SMX2000RMLV2U NMC: APC AP9630 Battery: SMX2000RMLV2U compatible pack from upsbatterycenter.com or upsbatterycenter.ca (I’m in Canada, so it was .ca for me - APC’s SKU for the battery is APC RBC117) PDU: APC AP7930

The UPS is capable of putting out ~1800W. My alternative was the SMT1500 or SMX1500 (I have several of the former laying around with dead or missing batteries), but the models I have don’t have any 20A outputs since they can only handle about 15A, I really wanted to go 20A for this, so I sprung for the 20A capable version, mainly to future-proof my power delivery for the rack, so I don’t have to worry about what I’m putting in there. Before moving, I had two SMT1500 units, one was rackmount, so the equivalent 1500VA APC rackmount version of the same, and they only had to feed a modem, switch, and the C6100/R710. I added a lot to the system when moving to this house, partly because this is our “forever home” and I wanted to kick things off right. I put the PDU on the side of a new 42RU rack I purchased specifically for the house, and rackmounted everything at the same time. A lot of thought went into which products to buy, I didn’t want anything brand new, and I didn’t want anything I would have to spend a fortune on to get working. The biggest expense through all this was the new networking hardware for the house, second biggest was the new battery pack, but thankfully UPS battery center is relatively inexpensive for that (still ~$400 for the battery alone, ouch). We have pretty cheap power here ( $0.182 per kWh on-peak, and $0.087 off-peak, in Canadian dollars), so I’m not overly concerned about energy use. I just don’t want the system to go down as soon as the power has a hiccup. unfortunately, my power provider has had several outages in the past year that lasted much longer than my battery lasted, so I’m thinking to look into solar or generator backup to keep at least my rack powered up, and maybe fridges and stuff. That’s beyond the scope of your question so I’ll just leave it at that.

The key here is: how much are you powering with this and how long do you want it to last. If you’re like me, and have a decent collection of things that need to stay on, a larger UPS like the SMX2000, plus a battery expansion may be the way to go, if you have less stuff but need extra run-time, maybe a smaller 1000 or 1500VA unit with a battery expansion is fine, or if you need protection against just small interruptions in power, maybe forego the added pack. It really depends on the power delivery in your area and how much you’re drawing. Part of the reason I picked the SMX2000 was because it had a connector for an external pack (along with the mentioned NEMA L5-20R), so I can expand the pack later. The add-on cards were a big plus for me, but I would have been happy with any network management whether integrated or not.

A big note from me, if you’re looking at (especially used) APC gear, is that you will want to make sure you apply any and all available firmware updates as soon as you can, since a lot of stuff on the used market gets deployed as soon as it’s received, and doesn’t get touched apart from that; once it is decommissioned for sale on the used market, nobody bothers to update it prior to sale, so update it as soon as you get it. This is especially true for the NMC modules, the interfaces had a complete overhaul during their useful life, especially for the NMC2, making it far more stable and far more capable of a unit for monitoring the UPS. I’m sure this is true for other vendors, but I haven’t really dealt with too much from Eaton or CyberPower to know (My employers so far have used APC almost exclusively, which is why I have a handful of SMT1500’s, all of which were destined for the junk pile after the battery went bad).

If you want to spend a bit more to buy something new, I would recommend something based on Lithium, as the majority of older UPS units use lead-acid, usually AGM. Lead-acid is great for cost, as the batteries are usually pretty cheap, but Lithium should last a lot longer, I expect to replace my Lead-Acid pack in my SMX in another 2-3 years; lithium should last 5-10 at least. Up to you though.

Good luck.

I’m not sure the mods are going to care honestly. I think most are in agreement that the Orville is basically star trek adjacent, and close enough that… Honestly as long as it doesn’t take over, nobody will bat an eye at the occasional Orville meme.

Is it his birthday?

I don’t know if my opener has the dry contacts required to make a solution like this work. I’m not keen on hacking the remote or anything. I might have to replace the opener before I have the right contacts to use something like this… which might be the plan, I’m going to take a close look at the specific unit I have and find the model and manual tomorrow, and see what’s in the manual before I do anything more…

We inherited a MyQ opener when we moved, and I’ve never been very happy with it aside if these issues.

Has anyone already done the research into a smart opener that doesn’t suck and allows local control via home assistant, preferably over zwave?

I’m going to do a bit of research on it myself, just wondering if anyone has a recommendation to point me in the right direction.

I don’t have wifi in my garage, and I’d rather not rely on wifi to control the door.

To be fair, mechanically, the opener we have is fine, it’s everything else about it that I generally hate.

They’re all seemingly very strict about using their gear. They all are, they always are. They’re lying when they say that things aren’t compatible or something.

The biggest lie is that you can’t. You can, and most of the time, unless it’s creating a problem, they couldn’t give a shit less if their equipment is working correctly; as long as you pay your bill, and don’t complain, they don’t give a shit. They have your money, whether things work correctly isn’t their primary concern.

If you require v6 to be static, the ipv6 equivalent to a static IP is a static /64 subnet, aka, an entire LAN. Since it’s globally routable, it needs to be issued by an RIR, the same way an IPv4 address would be.

So yes, they would issue it.

If they don’t have facilities for ipv6, there are options, such as getting an ipv6 over IPv4 tunnel going with someone like he.net. such tunnels add complexity and more work to the set up and rely on you having a very flexible router, but can be a good alternative.

It’s very situationally dependent. In many cases though, the only thing restricting you is the ISP not giving you the information to do it yourself, and sometimes, just sometimes, some kind of code or MAC address that limits what devices can operate on their network; though that’s usually set up for billing.

If you use any kind of PPPoE, the MAC address problem usually isn’t a thing; this is normally DSL/Fiber… not all fiber, just some. PPPoE needs authentication which usually means username and password login to do AAA for the client, because of this, there’s little to no security on the last mile. All these technologies are based on standards. Fiber is usually GPON, DSL has several standards, but modern DSL is usually vDSL or vDSL2, or some variant thereof, and Cable is generally DOCSIS 3 or 3.1. There are exceptions, but they’re not common.

The key is to find which specific technology the ISP is using, and find alternatives. In the case of DSL, it’s generally finding a DSL modem that uses the same profiles and annex as the provided modem; beyond that, plug it in and authenticate with PPPoE. Many PPPoE type providers use a circuit number and/or VLAN, so that generally needs to be set along-side the PPPoE credentials.

For DOCSIS, it’s a bit sticky, since I know of many cable providers who authenticate endpoints based on the MAC address of the modem; in which case, you not only need to find a modem that can support the protocols in use, and the channel widths (eg. DOCSIS 3.1 16x8), but also one that you can modify the ISP-facing MAC address on the DOCSIS interface to match the one they gave you.

For Fiber, things can be sticky, but often aren’t. The ISP can, but often doesn’t filter on all of the following: MAC, SN, SLID. All of these values are sent to the OLT (ISP side of the fiber), and it could fail on any one of them. For me, I’ve had success with the G-010S-A SFP module, and if you look around the internet, you can find a git repo that actually has all the commands to modify any/all of these values to match them to whatever the ISP provided to you. The most difficult is getting the SLID, since it’s not published on the outside of the modem. I managed to get my local GPON’s SLID from a G-010S-A module that I hijacked from a working modem; in that case it was a string of all zeros.

The information is out there if you look hard enough, and with a little bit of cleverness and ingenuity, you can usually find anything that’s missing.

I work in Networking (aka network engineering, aka a bunch of other titles), so this all comes very naturally to me; to give you some examples, one DSL modem delete I did for myself was to pick up an EHWIC-VA-DSL-M for a Cisco ISR router, after some configuration magic, which I won’t get into here, I was able to get it to connect to my ISPs DSL line, after a bit more configuration magic, the Cisco was handling all of the traffic from my network to the DSL. It was a very clean setup, only requiring a single phone line from the wall plugged into a module on the router, then on the other side of it (over ethernet) was my network. That’s a fairly advanced one, but I’m pretty proud of it. Another case was a friend on the same last-mile provider in my country, on a fiber line, where I removed the garbage modem they gave him and replaced it with a G-010S-A GPON to SFP module, and plugged that more or less directly into the router he owned. In each case, I shaved off a few ms of latency, and bandwidth was largely unaffected. It makes the internet run just that much faster than before, and puts the control in your hands.

Needless to say, the ISPs don’t want you doing this, and they don’t approve, but in general, you can do so without their involvement and for the most part, they are entirely unaware that it’s happened.

Let me know what situation you’re in and we can probably devise a solution to the garbage ISP modem issue. Frankly, the fiber modem delete is my favorite.

depends on what you mean by “do anything”. I’ve managed to shed several ms of latency by doing a modem delete.

My two favorite stories of this were for the local DSL/fiber provider here in Canada, Bell. They use vDSL2, and GPON/XGS-PON respectively. In the former case, I set up a node at my house, which was a Cisco ISR router, with a vDSL2 EHWIC card installed; after some work, I managed to get the unit dialing into the internet via PPPoE, and I managed to drop about 5-10ms of latency simply by removing the ISP provided garbage. It was also clean… a single phone cable plugged directly into the router, and out the other side was a switch, which provided all the network connections I required… my setup was a tiny bit more complicated than I’m explaining, but the other details don’t really matter (long story short, I was operating on a Bell line through a wholesale client (third party ISP using Bell’s “last mile”), and they provided me with a /29 subnet for internet routing - the Cisco handled the WAN to WAN communication, and on my /29, I had a few devices including my primary firewall, which was between me and the internet, that then broke out onto a switch for everything to connect to… a bit more than the average joe can handle, but I work in networking). The other story is about their GPON; I managed to figure out that their GPON interface is almost entirely unprotected, and worked with a G-010S-A (a fairly common design from Nokia, but has variants from other major vendors that are largely the same), so by buying or otherwise obtaining one, and programming it very specifically, you can actually plug the SFP GPON module directly into a router, and with some clever configuring, get your PPPoE to work across it without too much trouble. There’s plenty of info about it online if you want to see more.

The only sad story I have about this is that Bell started to release a new modem that has a built in fiber module (no longer using the G-010S-A), which is compatible with both the GPON and XGS-PON systems; I have yet to find an XGS-PON version of the G-010S-A that I can use for the purpose. A friend of mine, whom I did a modem delete for with the G-010S-A, was in an area that was originally served by GPON, so the solution worked. After some time though, Bell implemented XGS-PON in his area, and actually removed compatibility for the GPON, so the solution stopped working. Until I find an XGS-PON equivalent to the G-010S-A, I’m at an impasse. In the interim, my friend has put his modem back in-line, and IIRC put it into bridged mode, which is second best to a modem delete.

I’m a network technician/engineer as my dayjob, so working with this stuff is entirely in my wheel house, I can usually give useful advice for anyone wanting to walk in my shoes to delete their modem, and make it simple enough that it doesn’t require my level of skill to maintain (like in the case of my friend), and advice/strategies about how to handle the ISP.

CG-NAT is entirely in the ISP hands, I cannot touch their fancy CG-NAT engine or route around it. My best advice for anyone facing down CG-NAT, is to use IPv6, if your ISP supports it. Simply put, the best argument I’ve seen for IPv6 adoption is CG-NAT. NAT itself was bad enough, but CG-NAT is a whole new level of evil; it breaks so many things. IPv6 takes you back to the old days of globally routable addresses, end-to-end, completely eliminating the need for any kind of NAT. A large portion of the internet uses/supports IPv6 already, pretty much all the major data carriers support it and actively use it for their own gear (people like google, facebook, apple, microsoft, cloudflare, etc). IPv6 shouldn’t be feared, as an end user, the whole thing is going to behave exactly as you expect it to. The trick is: getting it up and working on your LAN, once you can work that out, you’re laughing.

I usually do a very not ISP sanctioned modem swap/delete.

Depending on the type of modem, you may be able to simply replace it with something else and the ISP may not have any way to really differentiate between the modems.

Is this for cable, DSL, or fiber?

A model number can really clarify a lot.

OP said, without damaging it…