--- Log opened Tue Jun 05 00:00:45 2018
00:01 < DoYouKnow> Well, you could, but that wouldn't allow you to universally address nodes and networks
00:02 < Aeso> ^ I think this gets closer to the real reason
00:03 < qman__> If anything, it would make more sense to use the network address for broadcasts
00:03 < Aeso> If I'm creating a single forwarding table on a networking device, how do I describe the difference between a route to 192.168.1.0 the subnet and 192.168.1.0 the host?
00:04 < DoYouKnow> Aeso: I say that, because its importance is in the traditional grouping of nodes and networks, and other sources/sinks of information, into a more powerful addressing system
00:04 < DoYouKnow> with some hierarchy as well
00:05 < DoYouKnow> but some of that is based on faith
00:05 < Aeso> note that RFC 3021 allows for /31 subnets and ignores these very principles, since you shouldn't ever need to broadcast on that network
00:06 < DoYouKnow> Aeso: hosts never use the network address
00:06 < DoYouKnow> unless they are routers
00:06 < DoYouKnow> or routing
00:07 < Aeso> forgive me for being slightly pedantic, but the default route technically counts as routing :P
00:18 < djph> Aeso: provided that you're not talking about a /31 or /32 (which are comparatively weird), the ".0" address is never a host address.
00:18 < djph> well, unless you're talking about something larger than a /24 -- but then (at least one of) the .0 address(es) is not a network address
00:18 < Aeso> djph, I know, we were positing why the subnet address couldn't be used for a host
00:19 < djph> because that's how it works (excepting /31 or /32 ... but they're "weird(tm)")
00:19 < tds> I've seen people using various hacks to make use of the network+broadcast address for hosts, those tend to be nasty though
00:20 < djph> tds: same, and they typically cause so many problems.
00:21 < tds> I've seen various setups with /32 on-link routes and having a separate gateway address, I am tempted to try a design like that
00:30 < djph> not seen that myself
00:54 < spaces> all networks sexy ?
01:14 < smallville> hey
01:14 < smallville> can a small office of 5 employees use a wifi only network?
01:15 < S_SubZero> yes
01:16 < smallville> I'm designing a small office network with 4 users and a file / application server
01:17 < Epic|> Really depends on the bandwidth requirements. The appropriate answer is "we don't know"
01:18 < smallville> i'm thinking, AC router, AC wifi cards, 100mbit internet
01:18 < smallville> running an program straight from the server, and accessing files from that same server
01:19 < smallville> the server is connected to the router with ethernet, but workstations are connected through wifi
01:21 < smallville> is cat6 faster than AC wifi at 100 mbit from the ISP?
01:28 < spaces> smallville office PC's should be wired, period
01:29 < clon3man> as long as at least 1 person is wired in for troubleshooting
01:29 < S_SubZero> what sorts of speed are you hoping to get?  What needs?
01:30 < S_SubZero> wired ethernet will likely behave better, but for a small office with light network demands either would work fine
01:30 < lupine> including keyboards and mice
01:30 < lupine> *especially* if you work in a bank
02:59 < TandyUK> [00:28] <spaces> smallville office PC's should be wired, period  +1
03:04 < banisterfiend> hi there, anyone here famliar with the 'pf' packet filter on osx?
03:08 <+pppingme> whats osx?
03:13 < light> just ask your
03:13 < light> question
03:15 < banisterfiend> ok, i came across some pf rules that look like this: pass proto { tcp, udp } group foo flags any
03:15 < banisterfiend> but i can't find the definition of 'group' anywhere or how to use it
03:15 < banisterfiend> it's not in any docs i've read
03:26 < kerframil> banisterfiend: "Similar to user, this rule only applies to packets of sockets owned by the specified group". ergo, group as in the effective gid.
03:27 < shalok> Where can I get a description of all the counters in `ethtool -S`?
03:27 < banisterfiend> kerframil where did you find that? link? thanks!
03:27 < kerframil> banisterfiend: the pf.conf man page
03:28 < banisterfiend> kerframil argh thanks, i was only looking at the 'pfctl' man page
03:28 < banisterfiend> cheers homie
03:54 < banisterfiend> kerframil can you think of any reason why an app would install its firewalls rules into /etc/pf.conf rather than just directly loading its own shipped conf files?
03:55 < kerframil> banisterfiend: as includes are supported, no - not really
03:55 < banisterfiend> kerframil you mean the shipped files could just directly include the /etc/pf.conf ?
04:08 < kerframil> banisterfiend: well, pf.conf can include directvies other files. spewing rules into a file that is definitively owned by the app in question, then just adding a single include to pf.conf would seem less invasive to me.
04:08 < kerframil> oops, directives from
04:11 < banisterfiend> kerframil thanks, but is it necessary to even bother with /etc/pf.conf - what would be the downside of just ignoring it and just loading the apps own files? just that it misses out on other rules th euser has set up i /etc/pf.conf?   surely there has got to be a canonical way of adding rules to the packet filter without directly modifying /etc/pf.conf?
04:15 < kerframil> banisterfiend: you raise a good point. I'm not even that good at pf so, quite honestly, I don't know. I'm not aware of any way to define collections of rules outside of pf.conf (or that which it includes), but it might be possible.
04:16 < banisterfiend> kerframil alright thanks :)
04:16 < kerframil> banisterfiend: perhaps by using anchors, then using pfctl to load rules that reference the anchors - without destroying the main ruleset that was loaded from pf.conf
04:16 < kerframil> banisterfiend: that's my best guess
04:17 < banisterfiend> kerframil thanks i have one final question
04:18 < banisterfiend> i notice the rule files for this app look like this: com.foobar.privacy.100.killswitch and foobar.200.exemptLAN. -- why the numbers? is this some kind of standard format?
04:19 < kerframil> banisterfiend: the numbers don't mean anything to me. if they had been at the front, I'd have ventured a guess that it is to enforce the resulting lexical order when iterating over the files but clearly not here.
04:19 < banisterfiend> alright cheers :)
04:19 < banisterfiend> thanks again
04:20 < kerframil> by the way
04:20 < kerframil> an applicable use of anchors does appear to be mentioned in the pf.conf man page. at least, in FreeBSD's.
04:21 < kerframil> can't say as I've ever used them - I use Netfilter these days. still, it looks promising.
07:50 < smallville> clon3man: hey.
07:51 < smallville> i know offices should be wired, but I'm not too good with cables. I'm more of a hardware/software support person
07:54 < smallville> I'm trying to build a small office network with only wifi, I'm looking for alternatives to ethernet that will supply 80Mbps, paying isp for 100Mbps
07:55 < DoYouKnow> some commercial satellite can have links like that, smallville
07:55 < DoYouKnow> maybe even residential
07:56 < detha> smallville: wrong solution. office PCs should be wired, period. (and cables are part of hardware, time to learn that part of it then)
07:57 < DoYouKnow> I see 25 Mbps
07:57 < DoYouKnow> from hughesnet
07:57 < smallville> detha I love everthing about networks, servers, I can dissassemble and reassemble any pc laptop phone or tablet. But I hate cables/
07:58 < smallville> ok I dont hate cables, I just dont find them intersting
07:58 < detha> Make it fiber-to-the-desktop then?
07:59 < Roq> So how are you going to connect your accesspoints to the network? How are you going to power your accesspoints?
07:59 < detha> Bit overkill for a 4-person office, but at least better than wifi
07:59 < smallville> Roq: goodpoint
08:00  * dminuoso .oO( PoE? )
08:01 < smallville> I'll just force myself to learn cabling. The hard part is getting them in the walls.
08:02 < trae32566[w]> smallville: cable lube, and if you have to make holes, I suggest sticking PVC behind them, so you can expand later.
08:02 < trae32566[w]> /s/behind/in/
08:02 < detha> smallville: it will be cheaper to have a contractor do that, than what you will have to pay for a wireless solution with as much reliability
08:02 < trae32566[w]> ^^^^
08:02 < Roq> smallville: No don't. If you don't know how, or if it's not your thing, just hire a company that does electrical / ethernet layer1 installation at your site.
08:02 < trae32566[w]> yeah, it's actually not super expensive
08:02 < trae32566[w]> I would have them do the PVC thing or make wiring chases though if possible
08:02 < trae32566[w]> it will save you headaches in the long run
08:03 < trae32566[w]> especially if you run fiber or more ethernet later
08:03 < smallville> trae32566[w]: now that is a great idea
08:03 < smallville> pay someone else to do it
08:04 < dminuoso> If you have them set up cables, be sure to lay down conduits for fiber optics as well.
08:04 < trae32566[w]> smallville: also, wireless is half duplex, and it waits on every client (except for more expensive AP's, which mitigate it somewhat with airtime fairness) so if one person brings in a slow ass computer (or you have one), it'll slow down the entire AP (for that frequency)
08:04 < trae32566[w]> slow ass meaning slow wireless connectivity
08:05 < smallville> hmm
08:05 < trae32566[w]> yeah, it sucks
08:05 < smallville> ok i'll find someone to to the cabling.
08:05 < trae32566[w]> often those fucking IOT devices slow 2.4ghz down a significant amount
08:06 < trae32566[w]> because low power / low cost = low speed 2.4 apparently
08:06 < smallville> why use 2.4 in the first place
08:06 < trae32566[w]> range
08:06 < smallville> not worth it
08:06 < trae32566[w]> the long range stuff is usually heavily 2.4
08:06 < trae32566[w]> sure it is
08:06 < trae32566[w]> for example, the city of Plano, Texas
08:06 < trae32566[w]> the *entire* city has private wifi for emergency services
08:06 < trae32566[w]> how? long distance 2.4ghz APs
08:06 < smallville> after using 5Ghz, 2.4 is like seing b/w after color
08:07 < trae32566[w]> you can get decent speed on 2.4
08:07 < trae32566[w]> and again, distance
08:07 < trae32566[w]> 5 doesn't go through anything very well
08:07 < smallville> you have a different definition of "decent" than i do
08:07 < trae32566[w]> for the majority of people, 2.4 is fine most of the time
08:08 < trae32566[w]> maybe not for you or I, but for most it is.
08:08 < smallville> you're right
08:08 < trae32566[w]> in the case of most business (and even some home) environments, range and stability is going to trump speed
08:09 < trae32566[w]> also keep in mind
08:09 < trae32566[w]> many business offices have *shittastic* wireless performance
08:10 < trae32566[w]> often because the buildings have a lot of materials that block or distort signal (ie: lots of metal in the walls, etc)
08:10 < trae32566[w]> this is especially true in multi-tenant high rise buildings
08:10 < smallville> do offices use a seperate wifi from the ethernet subnet? should be seperate, for security reasons, right?
08:10 < trae32566[w]> no not really, not unless you need to
08:11 < trae32566[w]> keep in mind that'd mean you have to manually route and allow anything that needs to communicate (for example say a printer, or a chromecast, or a TV)
08:11 < trae32566[w]> all of those being wireless devices that may need access to wired (or vice versa)
08:11 < trae32566[w]> I separate the subnets
08:11 < Roq> Plus you have to be realistic on the network's performance. If you expect 10 clients per accesspoint that is connected to a 1Gbit physical uplink, all clients won't get full 802.11 speed anyway
08:11 < trae32566[w]> but I keep them on the same VLAN
08:12 < trae32566[w]> so for example wired has 192.168.2.0/24, wireless has 192.168.3.0/24, guest has 192.168.4.0/24
08:12 < Roq> 802.11N or .11AC or what have you
08:12 < trae32566[w]> the first two are on a separate VLAN than the guest network, obviously.
08:13 < trae32566[w]> really my only point behind that is easily being able to see what is where when I look at ARP or DHCP leases
08:13 < trae32566[w]> so I know 192.168.3.x is wifi, 2.x is wired
08:19 < smallville> I told the person i'm building this network for, that he should have a wifi only network.. I totally underestimated to complexity of the network. How do I tell them that ethernet is better for them?
08:20 < smallville> this might not be the right channel for that question...
08:20 < chrustler> If the person dont know the difference, really do you need to explain?
08:20 < smallville> since they are paying for the cables, switches and patch panels, yeah they need a reason
08:21 < chrustler> Quite obviously a wired network is much better but it seems like totally different requirements
08:22 < DocScrutinizer05> tell them about WLAN == RF and if they still remember fiddling with antenna of portable TV or radio - when they want reliable network they should go ethernet
08:23 < smallville> hmm
08:23 < chrustler> I dont know one person that never complained about wifi so they should understand ;) ;)
08:24 < smallville> can I build the network on wifi, then add the cabling later?
08:24 < DocScrutinizer05> never use wireless if you as well could use cable based connections
08:25 < chrustler> that's a good rule ...
08:25 < smallville> yeah i figured as much
08:26 < DocScrutinizer05> smallville: migrating WLAN -> ethernet is basically a complete redo, there's like zilch you could re-use
08:27 < smallville> i get it
08:27 < DocScrutinizer05> ethernet -> WLAN is a more viable approach
08:27 < smallville> true
08:28 < dminuoso> chrustler: I never complain about wifi.
08:28 < dminuoso> But then again I only use it for my mobile phone at home or when sitting in a cafe.
08:29 < chrustler> I dont complain either, but spouse...
08:36 < detha> chrustler: and then there is the generation that complains 'the wifi is down' meaning that the wifi connection works perfectly well, but has no internet connectivity.
08:38 < chrustler> Oh yes... :)
08:38 < grawity> you mean the generation of people who aren't born network engineers
08:41 < detha> grawity: the generation that doesn't even know what a network is. The previous generation used to complain 'the network is down' or 'the internet is down'
09:56 < djwraith> does http to this date define anything that means "repeat the previous operation"?
09:57 < djwraith> say when a client fires a GET on a privileged URL requiring public key authentication, the server tells authentication is required. the client then shouldn't have to REPEAT the entirety of the request line
09:58 < djwraith> (and the request line could end up excruciatingly long)
09:59 < djwraith> it would be great something like "GET SP ASTERISK CR LF" could be used in its place
10:02 < jvdmr> djwraith: no, http (and http/2) are stateless by design, and the feature you're suggesting would make it stateful
10:04 < djwraith> so I should implement mine own with "GET /auth?<blah>" then?
10:05 < jvdmr> yes, if you want to do something like this you'll have to do it in the application that uses HTTP
11:00 < abernathy> Alright. My downlink to my gateway is 12.5 Mbps. All clients are on the wireless. If their link rate is anything above 12.5 Mbps, do they have maximum access to the available bandwidth, or is there something I am missing?
11:03 < djph> wifi is (speed) / (no. of clients)
11:04 < djph> so, if you have wifi with 100 mbps real-world throughput and 10 clients, each client will average out to 10 mbps throughput.
11:05 < abernathy> Interesting. Are there "keepalive" packets used for connectivity purposes, which prevent one device from properly getting everything?
11:06 < abernathy> Every bit of throughput**
11:06 < djph> huh? I never said anything about "keepalive" packets
11:06 < abernathy> i know; i'm curious what causes the (speed)/(no. of clients) rule.
11:06 < djph> wifi is a shared medium, 10 devices trying to use it = each device gets 1/10 the throughput.
11:07 < abernathy> djph: does this mean when all of them are trying to use it at once, or even when those devices are idly associated?
11:08 < djph> if the devices aren't doing anything, they don't count
11:09 < abernathy> oh, okay
11:10 < abernathy> djph: now if one device is quite far away and has low throughput, does that steal the same amount of bandwidth as others with sufficient link rates for their share of time? essentially, is time shared between all clients "equally"?
11:12 < djph> that one would possibly act as more than "one client"
11:13 < abernathy> that's interesting. seems like a super hard to manage network structure.
11:13 < abernathy> are there configurations available to boot 'troublesome' users from a specific ap?
11:13 < djph> not really
11:14 < djph> you can do some things, such as minRSSI (if available on your AP)
11:14 < djph> although the real thing to do is to turn DOWN the AP Tx power
11:17 < abernathy> djph: That's a clever idea.
11:17 < abernathy> I really need to get a multi-ap setup, but they're so damn expensive.
11:17 < djph> $100 or so (per AP) from Ubiquiti
11:18 < abernathy> what kind of latency nightmare are the new mesh network setups?
11:18 < abernathy> djph: that's pretty spendy. are there any refurbs/outlets?
11:22 < djph> I'm sorry, $100 for a commercial / business grade AP is "spendy"?
11:23 < abernathy> djph: i see ubiquiti's publicly traded. the parts can't be that spendy, eh?
11:23 < abernathy> seems like commercial/business marketing is what makes it expensive.
11:23 < djph> They're cheap as chips
11:24 < djph> the competitors (cisco, ruckus, etc.) tend to start several times that price
11:24 < abernathy> djph: perhaps if it's for an enterprise that makes money. ;D
11:24 < Maarten> ubiquity isn't "spendy" they are cheap. AP's with the same functionality from Aruba go for $600, and I am not even starting to talk about Cisco/Meraki.... :D
11:25 < Roq> ubiquiti's are great for their price
11:25 < abernathy> Maarten: your well seems pretty poisoned. price anchoring is a pretty common tactic.
11:25 < abernathy> it works across marketing campaigns, too.
11:26 < djph> I mean, you could always buy a Linksys or whatever ... which is the same price, and garbage ...
11:26 < Maarten> Yeah I bought 2 for my home. $130 a pop, have my whole house covered for $260.... you really can't beat that, unless you want to buy non-managed consumer crap such as Belkin or such
11:26 < abernathy> is it really in your best interests to say that they're NOT expensive?
11:26 < abernathy> as a consumer, shouldn't you want to minimize expense and maximize value for expenditure?
11:27 < djph> which is exactly what we got by spending $100 a pop on UBNT ...
11:27 < abernathy> i think we can all agree that while they're not as expensive as their competitors, they could still come down in price.
11:27 < djph> not really, no
11:27 < abernathy> without hurting anyone except a few executive paychecks.
11:27 < djph> at least not for the quality you get with them
11:27 < Maarten> abernathy, if you are a consumer that want's a little more than a ford fiesta, but doesn't want to pay for all the features of a mercedes..... in comes ubiquiti: offers a shitload of enterprise like features for the money of a linksys router.
11:28 < djph> ^^
11:28 < djph> it's like getting the upper-midrange T-bird, without going full on Mustang
11:28 < abernathy> Maarten: again, this is missing my point
11:28 < abernathy> you're being realistic, and accepting the price as a given
11:28 < abernathy> i'm trying to argue that, up from the cost, it's quite a way to the price.
11:29 < abernathy> therefore, for them to make a return on the devices they make, they could still lower the price quite a bit. do you disagree with this statement?
11:29 < djph> you do realize they're walking the razor's edge on profit with those, right?
11:29 < abernathy> doesn't seem so.
11:29 < avu> abernathy: why are you so sure about that?
11:29 < abernathy> avu: because i know how much plastic and silicon cost.
11:30 < avu> abernathy: and you know what parts exactly go into a machine like that?
11:30 < Roq> But those are not the entire costs factored in
11:30 < abernathy> ^
11:30 < avu> abernathy: even after you just learned about their existance a few minutes ago?
11:30 < abernathy> you're right. which is why i mentioned the executive paycheck bit.
11:30 < abernathy> avu: i know about ubnt.
11:30 < djph> plastic, silicon, the people to design it, the people to build it, shipping from china ... all cost money
11:31 < detha> abernathy: in that case, why aren't you flooding the market with competing devices at half the cost?
11:31 < abernathy> djph: shipping from china's free. ;D
11:31 < Roq> OS developtment, Community support, bug fixes,
11:31 < abernathy> it's literally cheaper to get pcbs made and shipped from china than to have them made at a local fab shop in the us.
11:31 < djph> I mean, $79 MSRP for a UAP-AC-LITE ...
11:31 < abernathy> Roq: yes, and these things are pennies.
11:31 < abernathy> the ac-lite isn't really comparable to the ac pro.
11:32 < abernathy> again, you're arguing "it's not THAT expensive."
11:32 < abernathy> i'm just arguing that it could be cheaper without hurting the company's ability to grow.
11:32 < Maarten> abernathy, I disagree with that statement, considering all the features you get. If you take all the hardware that is out there that has similar features, the ubiquiti line of products comes out as most affordable every time. Sure, if you look at the pure manufacturing costs, they are going to be made even cheaper. So are ruckus and cisco, where you pay for a name..... You simply can't beat them on pricing. If you still think its TOO expensive,
11:32 < Maarten> have fun with your D-Link's and Belkin's.
11:32 < abernathy> Maarten: you're anchoring.
11:32 < abernathy> I'm literally arguing that it's entirely possible to lower the price of ALL of those products.
11:32 < abernathy> i don't see how you've missed my point multiple times
11:32 < djph> It's 2x2 802.11ac vs. the UAP-AC-PRO 3x3.  and $80 (ish) vs. $180(ish)
11:33 < abernathy> i thought it had a worse radio in terms of sensitivity, black magic waveguiding, etc
11:33 < djph> not really, no
11:33 < Maarten> abernathy, possible... maybe. But not realistic, and certainly not by much.
11:34 < djph> I mean, if you're talking something like the UAP-HD, or the UAP-SHD ; yeah, they're chock full of black magic for their "High Density" deployments.  And yet are still like 1/4 the cost of Ruckus (etc.)
11:35 < djph> I asked about getting reseller discounts.  Wholesale prices are ... not that much less.
11:35 < djph> but maybe that's because I couldn't guarantee thousands of sales / month
11:35 < abernathy> Maarten: in the modern day, i'm surprised we even bother ordering the way we currently do. producing a bunch of units to sit on shelves when it's only 1-2 weeks out to have them fabbed is ludicrous. plenty of fat to be cut in terms of corporate structure (bullshit jobs, etc)
11:36 < Roq> I don't know why you're arguing against a product that's already the best value for price is the market, and say it could be cheaper still? You can agrue against the global economy, a national housing market, the value of gold in the same way. The market is value and demand, not a personal view on it
11:37 < abernathy> because we can do better.
11:37 < abernathy> that's as simple as it gets. we will always be able to do better as human beings, and we should demand it.
11:37 < Roq> Alright, build a better product and sell it for less
11:37 < djph> ^
11:37 < abernathy> Roq: I'm talking about the possibility of doing it, and you're telling me it's literally impossible. You're saying we've hit peak performance.
11:37 < Roq> No
11:37 < abernathy> Might as well give up on all production ever, and just use the same tech forever, now, yeah?
11:37 < Roq> That's not what im saying at all
11:37 < abernathy> Don't bullshit me.
11:38 < djph> wow, there are some pretty impressive logical leaps right there abernathy
11:38 < abernathy> That's exactly what you stated. If it's possible, then I should be able to demand that it happen, because that's literally what happens in modern society: A demand is satisfied for a price. Yes, on a PRACTICAL, RIGHT-NOW, PROBLEM-SOLVED sense, it's not available.
11:38 < abernathy> But for you to claim that I'm not allowed to say we can do better at all, and have a discussion about how it's possible to do better, is just you being a stick in the mud.
11:39 < abernathy> djph: long logical legs.
11:39 < Roq> What the fuck
11:39 < avu> abernathy: and you think that betterment can be achieved by "demanding" it from some unrelated random people in a random IRC channel in a very confrontational manner?
11:40 < abernathy> avu: No, I think that discussing it on IRC is perfectly okay, and shouldn't be met with dogpiling.
11:40 < djph> We're telling you your pipe-dream of cheaper is not gonna happen "right now", and that you can take it or leave it.  You're the one who's going on about how because right now their value/price is heavily skewed towards value, that we're "anchoring"
11:40 < abernathy> I didn't say, "FUCK YOU, DJPH! GIMME GIMME, CHEAPY WEEPY"
11:40 < Roq> Sorry for the language
11:40 < abernathy> I said, "that's expensive."
11:40 < Maarten> abernathy, I'm sure huge organizations like Cisco can trim all sorts of fat. But Ubiquiti is actually quite small and lean, their head developers till talk to the people directly, and you can get forum support from the people that design the stuff, unlike Cisco, Ruckus, etc.... that is one of the reasons they are so much cheaper, they don't run their company on "bullshit" jobs etc..... You would of course know this if you particpated regurlary
11:40 < Maarten> on the UBNT forums. It's a young company..... with no fat to trim, really. THAT is why they can offer the prices they do. Talk to them directly, you can find out a lot about them.
11:40 < djph> actually ...
11:40 < abernathy> Because it literally is expensive.
11:40 < TotallyNotKim> ya'll should smoke one and chill out for a sec
11:40 < abernathy> Maarten: and their execs still make more than the average household*5
11:40 < djph> I miss an-cheng and stig
11:40 < abernathy> Maarten: also, they're publicly traded. they're more incentivized to profit than to innovate.
11:41 < abernathy> Maarten: all corporate organization is obsolete.
11:41 < Maarten> abernathy, their exec's started the company with $30k in savings and credit card debts in 2005. Try to learn a bit about them first, maybe?
11:41 < abernathy> yeah, started.
11:41 < abernathy> that's not what glassdoor says, now.
11:41 < abernathy> i'm not saying they're terrible people.
11:41 < djph> hell, I got to speak w/ Robert Perra at one of the WISPA events.  Didn't even know it was him until someone was all like "do you even know who that was!?:
11:41 < abernathy> i'm saying there are cost optimizations available, and the fact that we need to buy from them doesn't make sense if they don't make it.
11:41 < djph> real down-to-earth kinda guy
11:42 < abernathy> furthermore, we don't need to even talk about ubnt.
11:42 < abernathy> IN GENERAL, we should have cheaper, better gear available for people. We don't because of corporate structure and modern things holding us back.
11:42 < djph> we DO have "cheaper, better gear" available.
11:42 < abernathy> batch-production of chips and assembly by robots is entirely possible for the minimization of cost, but instead, we're still doing pre-stock retailing.
11:42 < TandyUK> Also greedy capitalists want to make profit ;)
11:42 < abernathy> TandyUK: we don't have to make profit. profit only helps capital.
11:43 < djph> TandyUK: hey now, capitalist pigs have to eat too
11:43 < Maarten> abernathy, anyways.... I believe you came here asking about access points, you got some pointers.... and you start a crusade against corporate greed and how the cheapest among them should be even cheaper. How does that even help?
11:43 < TandyUK> go buy UBNT and make it a not-for-profit then lol
11:43 < TandyUK> otherwise this whole aregument is just LOL
11:43 < abernathy> Maarten: no, i started a counterargument against being told i'm a goddamned fool for saying we should have access to cheaper stuff.
11:43 < TotallyNotKim> anyone else some nachos? Got popcorn too
11:43  * TotallyNotKim snacks
11:43 < abernathy> TandyUK: not-for-profits still make profit, most of the time.
11:43 < abernathy> having a company at all is the problem.
11:44 < djph> abernathy: YOU DO. It's offered by the likes of Linksys, Belkin, TP-Link, D-LINK, Netgear, etc.  Enjoy.
11:44 < TandyUK> oh right, some bloke on his own is going to do all this then?
11:44 < abernathy> however, this is now steering off topic, since it takes a lot more than just planning the networking gear's production to make this work.
11:44 < abernathy> djph: not the same utility.
11:44 < Maarten> abernathy, what is keeping you from using your own $30k in savings, credit card debt, and doing it BETTER than Ubiquiti?
11:44 < abernathy> TandyUK: why would it need to be?
11:44 < abernathy> you're making false dichotomies.
11:45 < djph> abernathy: well, then (approx) $100 is the cost-of-entry into that level of utility.  Good day, sir.
11:45 < TandyUK> huh? you said "<abernathy> having a company at all is the problem."
11:45 < abernathy> Maarten: the lack of the ability to convince a single person that ubnt is slightly overpriced by any measure, apparently.
11:45 < abernathy> TandyUK: yes. a company is a specific entity in modern society. it's not just "a group of lads"
11:45 < TandyUK> Im aware, I run some lol
11:45 < abernathy> a firm is the more technical term
11:45 < abernathy> did you know that it's not the only way to organize production?
11:46 < abernathy> we lived for 200,000 years without it. it's not the only way.
11:46 < abernathy> djph: in the current day, sure
11:46 < djph> perhaps, because this room is (was) primarily "professionals in the networking trade", and $100 for an AP is literally cheaper than anyone else on the market.
11:46 < abernathy> yes
11:46 < abernathy> i acknowledged this
11:46 < abernathy> did you miss the parts where I said, "i'm not talking about solving a problem here"
11:46 < abernathy> this is the #1 problem with freenode
11:46 < abernathy> everyone thinks i'm here to do something more than discuss the subtleties of things
11:47 < abernathy> if you don't want to do that, it's not really fair to just jump in and tell me i'm wrong when you don't get my point/purpose.
11:47 < TotallyNotKim> if _everyone_ thinks that...
11:47 < djph> or, we don't care that you think you should get it 10% cheaper, because "plastic is cheap"
11:47 < TotallyNotKim> surely freenode is the problem
11:47 < abernathy> TotallyNotKim: ad populum
11:47 < djph> TotallyNotKim: obviously.  So anyway, back to actual networking
11:48 < abernathy> TotallyNotKim: have you never read the hacker file? it's a common issue: people think everyone is there to solve a problem.
11:48 < djph> ... need to subtly nuke a network because I'm tired of dealing with shitty unmanaged switches.
11:48 < abernathy> conversations are structured in a question->"OBJECTIVE ANSWER" fashion most of the time, and people habituate this.
11:48 < abernathy> djph: what's wrong with unmanaged switching?
11:48 < Maarten> abernathy, they are literally the cheapest out there. They are a small business with less than 50 employees. They don't make shitloads of profits, and although I don't know what their CEO makes in salary, I am sure it is comparable to industry standards. If you think you can do it for less money than Ubiquiti, than go out and start your company and do it. But until you do..... Ubiquiti will remain the cheapest out there with enterprise like
11:48 < Maarten> features.
11:49 < regdude> everything is wrong with them
11:49  * TotallyNotKim smells round 2
11:49 < abernathy> Maarten: you've restated your point 3 times. they're the cheapest out there RIGHT NOW. they're still making quite a bit of profit (quite enough to be publicly traded and have plenty of stock scandal going on)
11:49 < Roq> TotallyNotKim: Got anymore of that popcorn
11:49 < djph> Maarten: apparently, last year he made $0.00
11:49 < TotallyNotKim> Roq: ima heat some corn
11:50 < abernathy> Maarten: your entire argument is just ad populum. "everyone else does it, therefore it's right"
11:50 < abernathy> completely illogical.
11:51 < djph> Maarten: vs. Ben Moore's 300k salary + ~110k in stocks / Kevin Radigan's 350k base + 100k incentives
11:51 < abernathy> additionally, software engineers making more than 4-5 times the average HOUSEHOLD...
11:51 < Maarten> abernathy, like I said: put your money where your mouth is. Show the world how it is done, and start your own networking gear company. Then come back here in 2-3 years and laugh at us all....
11:51 < djph> Maarten: (or hire us, amirite?)
11:51 < abernathy> why would i enter into a corporate firm if i vehemently disagree with their structure
11:52 < abernathy> i'm literally whining about the structure of modern production
11:52 < Maarten> abernathy, go non-profit!
11:52 < abernathy> that's not how that works
11:52 < abernathy> lol
11:52 < abernathy> you don't just say "non-profit" and then have this happen
11:52 < Maarten> abernathy, so you are saying you can't beat them, and you refuse to join them. Ok. Now what?
11:52 < djph> ... aww, no love for destroying this network ... :(  fine, the ol' loop it is.
11:53 < abernathy> you still, in modern society, must bend to the will of others, and others are plagued by other plagues of society (student loans, etc)
11:53 < TotallyNotKim> so do I actually need a stacking cable for Catalyst 3650s?
11:53 < abernathy> Maarten: neither of those things are necessary. i came into a discussion channel to discuss, and you're instead telling me to magically conjure a solution, because you don't want to be part of the conversation.
11:53 < abernathy> you're free to merely stop responding.
11:53 < djph> TotallyNotKim: if you want to *stack* them, IIRC
11:54 < TotallyNotKim> djph: I thought they would come with some sort of click on mechanism. Seems like the cables are needed
11:54 < djph> TotallyNotKim: It's been a (long) while -- I think that makes them effectively a 6000-lite
11:54 < djph> so if that's what you want / need ...
11:55 < TotallyNotKim> ay
11:55 < TotallyNotKim> ty
11:55 < djph> good luck :)
11:55 < Maarten> abernathy, I have given you my opinion - in that Ubquiti is a pretty lean company, doesn't have super rich executives compared to other companies, and are delivering a fine product for a low price. You seem to be wanting to go against everything corporate structure and come up with things like the structure of the modern corporation and student loans..... what, is your last name Marx or something? Companies make profit, that is how it works in
11:55 < Maarten> capitalism.....
11:57 < djph> Maarten: given the report on the internet (which is suspect at best -- it wasn't the UBNT yearly review or whatever).  They look to operate like companies did in the 50s/60s -- execs "only(tm)" make 2-6x as much as the staff
11:58 < abernathy> that's quite a lot.
11:58 < Maarten> I'm not saying ANY company is perfect.... but c'mon, they are allowed to make profit, right? ;)
11:58 < abernathy> especially when the staff make that much.
11:58 < abernathy> Maarten: not that much profit. it's just ludicrous.
11:59 < abernathy> when my argument is that ALL companies need to trim fat, it's hard for me to take your claim that "they're good compared to other companies" seriously, because I know they are--I run UBNT wherever I can and advocate for it as a solution in the current day.
11:59 < abernathy> However, it should change.
11:59 < Maarten> abernathy, and who are you to determine how much profit a company can make? This is not a controlled communist economy you know.... ;)
11:59 < djph> Maarten: I would agree -- i mean, I'm "allowed" to make profit -- it's how I have savings to have replaced appliances and stuff, without needing a loan.
12:00 < abernathy> Maarten: a reasonable person saying it's unnecessary
12:00 < abernathy> you're allowed, djph, but it's not necessary, and it's through the collective force of using other people's money (like all of society--you're not special--we're all forced to do it)
12:00 < abernathy> you have to admit it'd be cool to order directly from the supplier.
12:00 < djph> Maarten: (and sorry if I'm egging on the other guy, I /ignored him a while back when things started getting too far off the rails)
12:21 < shtrb> Anyone idea what could cause ip address to be garbaled : I can see incoming traffic like 1.172.16.2 , got ip 32.172.16.2 etc
12:21 < djph> shtrb: NAT?
12:22 < shtrb> ppp from carrier
12:22 < shtrb> seems like that
12:22 < djph> ppp(oe) is a bit elusive.  I've not needed to use it in ages, so :|
12:22 < shtrb> djph, but look how the address is presented (instead of sane 172.16.2.1)
12:23 < shtrb> djph, it's 3G :)
12:23 < djph> well, 1.172.16.2, is a valid address in 1/8
12:23 < djph> granted, I think that's also the US DOD's network, but ...
12:24 < djph> nope, it's part of APNIC
12:24 < grawity> something isn't getting the network byte order conversion right?
12:24 < shtrb> I doubt my half backed 3G uses that
12:24 < shtrb> I tried , wvdial and network manager (both gave me the same IP )
12:24 < detha> network byte order, or off-by-one in parsing a header and not skipping some tag?
12:25 < djph> well, you ISP could be using 1/8 in their NAT (or have a subset of that)
12:25 < detha> either way, "software problem. shoot the developer"
12:25 < djph> or what detha said
12:25 < djph> I mean, we've seen crap ISPs use all sorts of registered public addresses inside of their CGN space before
12:25 < shtrb> It's not a school ISP , so I don't think you could apply that
12:27 < regdude> like AT7T
12:27 < regdude> &
12:27 < djph> apply what? CGNAT?  lotta ISPs run that shit these days
12:27 < shtrb> AT&T took over 1.1.1.1 which I could swallow , but not that
12:27 < shtrb> djph, shooting people
12:27 < djph> ... you're not making any sense now.
12:28 < shtrb> It was a bad joke (school shooting, shoot the developer )
12:28 < microwaved_> glad we don't allow guns in this country
12:29 < djph> meh, guns are fine.  people are not.  We should really just ban people
12:29 < shtrb> firejail get some new meaning
12:29 < TandyUK> ^^^  Agreed,  people kill people, not guns
12:29 < TandyUK> people kill people with axes, hammers, knives, and all sorts too
12:30 < TandyUK> but equally glad the idiots in this country dont have access to them
12:31 < shtrb> You can say a lot about a country where there is a term for glassing someone
12:31 < TandyUK> just proves my point really, that guns arent the problem, its the idiots holding them
12:31 < djph> TandyUK: so, some euro type was going on about how awful the US was.  2016 data, because that's all the CDC has ... excluding suicides, guns killed like 30k people ... a whopping 0.01% of total deaths in the US.  (Suicides doubled that, but ...)
12:32 < TandyUK> well that says a lot lol
12:32 < TandyUK> your gun crime rate is insane, but double the number of people killed by guns want to kill themselves, says a lot about the country tbh
12:32 < djph> NOTE - that 30k is inclusive of homicide, accidental, and whatever other categories that were broken out from "total" (only exclusion was suicide)
12:33 < TandyUK> ah kinda worse then lol
12:33 < TandyUK> but this is in a country where its illegal to cross the road
12:33 < djph> no, sorry -- it was ~30k murder / accidental / blahblah ... and like 60k (or thereabouts) if you included suicide
12:34 < djph> for comparison, auto-related deaths were also like 25-30k
12:35 < shtrb> TandyUK, how do people got machetes when police suppose to arrest people with scissors ?!
12:39 < regdude> wtf, circulatory diseases are most common causes in my country. how is that possible, we are the poorest of them all
12:39 < shtrb> cheap food is not healthy ?
12:41 < regdude> a dinner at McDonalds cost twice as a meal for the whole day, so since its a poor country, then I suppose no
12:41 < djph> TandyUK: bah, can't find '16 ... but '15 ... Total (rounded to nearest 100k) -> 2.7 million.  Diseases of heart (nearest 10k) -> 630k; Accidents (nearest 10k) -> 150k; suicide 40k .... "homicide" isn't even LISTED (or rather, it's in the 540k of "everything else", after the named stuff, and a bunch of other diseases)
12:41 < regdude> oh, I forgot, we are all alcoholics here
12:41 < djph> damn straight we are
12:41 < regdude> I mean, in my country
12:41 < TandyUK> Compare those stats with the uk (for 2016): Homicides: 577,  Deaths at work: 137,  deaths from road accidents: 1792
12:42 < TandyUK> but the US does have 5x more people from the uk, so factor that in too
12:42 < TandyUK> s/from/than
12:43 < TandyUK> 7327 died from alcohol lol
12:43 < TandyUK> so more than all accidents, and murders combined lol
12:44 < shtrb> regdude, I didn't say fast food I said cheap food (you wish to make beans, you need to soak them and then boil them, you wish to prepare meat and fish ? get ready for two hour work etc)
12:45 < shtrb> cheap food like rice/bread/potatoes are not that healthy ( I think that is the cheapest )
12:46 < regdude> shtrb: well most of people here choose a liter of heavy alcohol instead of a kilo of pork, costs about the same, but the choice is clear
12:46 < djph> bah, I can't find the CDC table that listed firearms (all), and then the breakdown :(
12:48 < djph> ah, there it is
12:49 < shtrb> Influenza is in the top reasons of death ?! what is it 1918 again ?!
12:49 < Apachez> "dinner at mcdonalds", does not compute...
12:49 < regdude> oh it does here
12:50 < regdude> I like how they built one of them 50m away from a school
12:50 < shtrb> People need to have something for the second breakfast
12:50 < Apachez> did you see the series jamie oliver made on schoolfood in murica?
12:50 < Apachez> a real tomato is forbidden
12:51 < Apachez> ketchup is considered to be a "vegetable"
12:51 < shtrb> What ? why ?
12:51 < Apachez> its like... wtf?
12:51 < djph> TandyUK: holy shit, I was way off.  the TOTAL firearm-related death toll in 2015 was under 40,000.  22,018 suicide . 12,978 were homicide.  The remaining 1200-odd were either undetermined, accidental, or legal intervention/war.
12:51 < Apachez> unsweatened milk is forbidden, sweated and colorized milk is ok
12:51 < Apachez> well a real tomato can be sprayed with things
12:51 < Apachez> but lets ignore which tomatoes are used to make that ketchup :)
12:52 < Apachez> also due to public procurement
12:52 < shtrb> "sprayed with things" , I hope they don't know how crops survive incescts
12:52 < shtrb> ketchup does not have to have tomatoes
12:52 < djph> heathen
12:52 < Apachez> so pizza was considered "healthy food" because it had veggies on it
12:53 < Apachez> shtrb: it does around here
12:53 < myrat> wossap people
12:53 < Apachez> so it was a funny series
12:53 < Apachez> after like 3 months he gave up and went back to UK :)
12:53 < djph> Apachez: that was right around the time of queen obama fucking with school lunches, wasn't it?
12:53 < Apachez> that LA county politicians made everythign they could to fuck things up for him
12:54 < Apachez> banned him from schoolarea etc
12:54 < Apachez> because oh behold it turned out that the kids actually liked his food
12:54 < Apachez> and that food didnt have like shitloads of sugar etc stuffed in it
12:54 < djph> funny how that works
12:55 < shtrb> He could start giving away cheese
12:55 < myrat> who can help me with network
12:55 < djph> got into that argument with some teachers around then
12:55 < shtrb> There is a cheese reserve :D
12:55 < djph> myrat: ask a real question.
12:55 < myrat> djph i lost ospf route
12:55 < Apachez> hmm perhaps it was that second attempt that went in LA?
12:55 < Apachez> seems like the first season was west virgina https://www.youtube.com/watch?v=oLgmk323H6k
12:55 < myrat> some vlans down
12:56 < djph> Apachez: something about "I couldn't make a healthy lunch for under x00 kcal".  I had 50kcal left over.
12:56 < shtrb> From petro dollar to Cheese dollar http://www.slate.com/blogs/moneybox/2016/05/02/the_united_states_has_stockpiled_a_grotesque_surplus_of_american_cheese.html :D
12:56 < myrat> djph
12:56 < djph> myrat: so fix your ospf.  Or, give a better description of the network & everything so that we can actually help you.
12:56 < Apachez> https://www.youtube.com/watch?v=v6NLmdT6rTg
12:56 < Apachez> ahh right, season 2 was in LA
12:57 < myrat> dfph okay wait a minute
13:03 < myrat> dfph problem with hello packages
13:03 < myrat> dfph: the ospf cant see the other ospf
13:05 < djph> fix your firewalls?
13:06 < myrat> djph: the problem can be in firewall?
13:06 < djph> sure
13:07 < grawity> or in the switches if they block multicast
13:07 < djph> if you're accidentally dropping the packets from the other router ...
13:07 < djph> ^ or that
13:07 < myrat> djph: the ospf cant see they neighbors
13:09 < myrat> grawity: how i can see if it blocks
13:10 < regdude> if your switch has IGMP snooping, then most switches don't make an exception for local multicast addresses, which are used by OSPF
13:11 < grawity> aren't they required to?
13:11 < regdude> not really, RFC is only a guide line
13:11 < regdude> like your all mighty Cisco ignores it as well
13:13 < myrat> the switch is snr
13:17 < regdude> well it does have IGMP Snooping, but the manual is in Russian, don't want to look for the local multicast support, try disabling IGMP snooping
13:17 < shangul> Hi. What is Route IP? It is written in my ISP's panel: "Route IP: Doesn't have Route IP". for my account.
13:19 < djph> perhaps they mean gateway?  who knows ... ISPs do weird shit
13:21 < shangul> djph, what gateway?
13:21 < djph> shangul: I would imagine the gateway you're supposed to connect to.  ISP should've provided you instructions for setup (or have DHCP / pppoe / etc. handling giving you all the relevant information)
13:22 < myrat> regdude: okay i'll try it
13:25 < zeldafan78> "From" can be "John Doe <bgates@google.ru>" or "bgates@google.ru", but the SMTP "MAIL FROM" can only be an e-mail address, right? No name or other formatting?
13:28 < shangul> when I setup port forwarding/PVC's Virtual Server, does my router inform some other server(NAT server?) that I'm opening port X?
13:29 < zeldafan78> https://documentation.mailgun.com/en/latest/api-sending.html#retrieving-stored-messages <-- Why is it called "recipients" and not "recipient"? The description says "recipient of the message as reported by MAIL TO during SMTP chat."
13:31 < sinatrablue> ususally you have to manually setup the secondary port to listen
13:34 < TandyUK> you can send messages to multiple recipients in one SMTP transaction
13:34 < TandyUK> eg, a message to: bob@hotmail.com, joe@hotmail.com, alex@hotmail.com, all 3 messages can be delivered in one SMTP connection
13:38 < shangul> sinatrablue, me?
13:38 < zeldafan78> TandyUK: The description says "recipient of the message as reported by MAIL TO during SMTP chat."
13:38 < zeldafan78> I have never been able to decipher WTF they mean by multiple addresses in "to", Cc and Bcc. These are utterly confusing concepts to me which completely ruin the "cleanness" of e-mail.
13:39 < zeldafan78> Is there a limit of how many you can "to"-send to? And will they be invisible to each other? Etc.
13:39 < djph> no, no.
13:39 < djph> well, maybe there is a finite number of bytes in a 'to' field ...
13:39 < TandyUK> cc and bcc are different headers
13:40 < djph> but if you don't want people to see each other's addresses, you bcc them
13:40 < TandyUK> TO: is a specific one, and can have multiple addresses
13:40 < zeldafan78> I would never want to send the *identical* e-mail to several people. I'd send one to each person as a separate "entity".
13:40 < TandyUK> well yeah there is a limit somewhere, but its not 1 SMTP connection == 1 message to 1 person
13:40 < djph> I take it you've never wanted to send a report to your boss and co-workers at the same time?
13:41 < zeldafan78> Never had any such, so no.
13:41 < zeldafan78> But even if it may be useful, it confuses me.
13:41 < sinatrablue>  why wouldnt you want an identical email? batch sending should be fine
13:42 < djph> There's nothing confusing about sending the same message to a handful of people ... it'd be daft to type the same message N times to send to N individuals
13:42 < zeldafan78> I'm trying to have a clear "from" e-mail adress in my local database. It gets fucked up if they can send as "Full Name <e-mail>" and not just a clean e-mail address, or multiple ones listed with commas.
13:42 < djph> then you're doing something wrong
13:42 < ^7heo> more news at 11.
13:43 < ^7heo> seriously we should put that in the /topic
13:43 < ^7heo> "you're doing something wrong"
13:43 < zeldafan78> Mailgun documentation = written by monkeys.
13:43 < sinatrablue>  its your fault and you should feel bad
13:43 < ^7heo> Documentation is statically written by monkeys.
13:43 < djph> apes, technically.
13:44 < ^7heo> Someone once told a manager: "given infinite time, an army of monkeys could rewrite the works of shakespeare"
13:44 < zeldafan78> Maybe it's important information that I should be saving if an incoming e-mail was sent not just to my e-mail address, but also to a bunch of others which I may or may not be seeing due to the confusing ways you apparently can send e-mails.
13:45 < ^7heo> They took it literally and went with "all we have to get written is documentation, so I guess 10 months will be enough then"
13:45 < sinatrablue> documentation is like sex, even when its bad, its still better than nothing
13:45 < ^7heo> if documentation is like sex, that would explain why most of us don't get any,
13:45 < myrat> regdude
13:45 < sinatrablue> wait there arent any programming porn stars?
13:46 < ^7heo> because people won't pay to see fat male blobs sweating on each other while having gay sex.
13:46 < djph> ^7heo: actually, rule 34 ...
13:46 < myrat> regdude: igmp in agr is disabled
13:46 < zeldafan78> I swear, I have spent so many months of my life (probably years at this point) just reading and re-reading the same sentence or several sentences in various "documentations" and just cannot figure out what the hell they are saying. It's so ambiguous and vague that it makes me want to scream.
13:46 < ^7heo> djph: it doesn't mean people would pay ;)
13:47 < sinatrablue> id watch that if they had documentation on it
13:47 < myrat> djph: no the problem ain't in firewall
13:48 < djph> ^7heo: fair enough
13:48 < djph> zeldafan78: perhaps the common factor then is you.
13:49 < djph> zeldafan78: email, barring the fact that you kinda see everything in one place, is very much like paper-mail.
13:50 < djph> you write a letter, address it to someone (or multiple someones), and send it on its way ...
13:51 < sinatrablue> way too easy
13:51 < sinatrablue> gotta write a program to auto copy paste the contents first
13:52 < zeldafan78> "recipients 	string 	recipient of the message as reported by MAIL TO during SMTP chat."
13:52 < zeldafan78> You decipher, please.
13:52 < zeldafan78> It contradicts itself.
13:52 < zeldafan78> Is it recipientS or recipient?
13:53 < sinatrablue> :zelda maybe its in spanish
14:14 < myrat> djph: are you here
14:16 < djph> no
14:17 < myrat> haha
14:17 < myrat> then who write no
14:17 < djph> "recipients" (type string) - recipient of the message, provided in a MAIL TO command during an SMTP session
14:19 < djph> zeldafan78: you can issue multiple "MAIL TO" commands in a single SMTP session.
14:22 < dogbert_2> hey djph
14:22 < djph> 'sup dogbert_2
14:23 < dogbert_2> not much...
14:23 < djph> another day in paradise, amirite?
14:26 < penguin_> Hey for fun i am making a network cable killswitch, what cable on a RJ45 plug should i connect to the switch?
14:28 < djph> you do realize that's gonna break the spec, and as such, is likely to result in subpar performance (if any at all)
14:29 < penguin_> djph, Yea i know. I am just making it for fun
14:31 < djph> o...kay then. I find stranded conductors easier to solder to switches, but solid isn't that difficult
14:31 < djph> use whatever you feel like, I guess
14:32 < penguin_> djph, But what colour would kill the conneciton?
14:32 < djph> OH, you meant to ask which *conductor* would break it, not which cable ...
14:33 < penguin_> yes
14:33 < djph> pins 1,2,3,6 would be the best bet, as they're tx/rx for fast-ethernet.  anything else would just let gbit fall back to fe
14:33 < djph> *any of pins [...]
14:33 < dogbert_2> hmmmmm!
14:34 < djph> as for colors, it depends on which wiring scheme you use (568A or B)
14:35 < Kingrat> although usually if you have a break in a single wire 4/5 or 7/8 it will fail to connect all together from what i have seen even if 1/2 and 3/6 are good
14:35 < djph> you've got some weird shit networking cards then Kingrat
14:35 < Kingrat> try it sometime
14:35 < djph> I've never not seen things fail back to FE if 4/5, 7/8 were missing
14:36 < Kingrat> if all 4 are missing, yes, it works fine
14:36 < djph> ... although, maybe it was that an entire pair (e.g. 4/5) was out
14:36 < Kingrat> im saying ive seen plenty of cables with bad connections in one of those pins fail to connect on numerous occasions
14:37 < djph> now, if it's a PSE / PD, oh yeah, you'd better have all 4 pairs or it won't work
14:37 < djph> *it's between a [...]
14:38 < penguin_> okey thanks
14:43 < zeldafan78> djph: I don't understand what you meant.
14:43 < zeldafan78> "recipients  string  recipient of the message as reported by MAIL TO during SMTP chat." <-- So is it recipientS or recipient?
14:48 < djph> zeldafan78: it depends - are you sending to one or more recipients?
14:49 < djph> if one recipient, then "recipient".  If many, then "recipients" (note that I've not read the RFC in some time, you may need a separate "MAIL TO" invocation for every intended recipient
14:52 < djph> oops, it's RCPT TO, not MAIL TO
15:26 < Miguel2013> why looking up hostnames takes time? like 5 seconds or more why is not instant like pinging a local ip
15:27 < bezaban> slow dns server
15:27 < dogbert_2> DNS can be time consuming, if it's a slow server
15:29 < bezaban> multiple levels of recursive lookups may add to that
15:30 < detha> or a resolver library that implements happy eyeballs and no real ipv6
15:31 < detha> s/real/working/
15:33 < bezaban> mmm, broken ip6 is bad
15:38 < zeldafan78> "recipients  string  recipient of the message as reported by MAIL TO during SMTP chat." <-- So is it recipientS or recipient?
15:38 < zeldafan78> I want to know if I'm supposed to take into account multiple addresses or what.
15:38 < zeldafan78> The fucking retards have zero examples and nonsensical descriptions for everything.
15:38 < zeldafan78> So implementing this takes 5 months instead of 5 minutes.
15:39 < zeldafan78> And if you ask them by e-mail, they barely read what you type and harass you for days afterwards with "your feedback is wanted" and "your ticket has been closed" etc.
15:40 < djph> zeldafan78: of course you have to account for someone sending a message "to" multiple people at one time.
15:40 < djph> zeldafan78: see RFC5321 for full details of SMTP.
15:43 < djph> zeldafan78: although, that being said, why on earth are you writing your own SMTP server?
15:49 < zeldafan78> djph: I'm not. I'm trying to use Mailgun specifically for *not* running or making my own SMTP server.
15:49 < zeldafan78> djph: When the e-mail is received by any of those people it's addressed to, do *they* see the list of other people it was sent to as well?
15:50 < zeldafan78> The JSON data returned when asking for stored messages has the parameter called "recipients", but the description says "recipient of the message as reported by MAIL TO during SMTP chat."
15:51 < zeldafan78> So... why is it named wrong? Or can it contain multiple addresses?
15:55 < Miguel2013> are phone hotspots dns servers slower than comcast dns
15:55 < djph> zeldafan78: of course they do. that's how emails work
15:56 < Apachez> zeldafan78: unless you sent with BCC
15:56 < Apachez> Blinded Carbon Copy
15:56 < Miguel2013> anybody here uses chrome
15:57 < djph> ohh, this is some shitty hipster psuedo-email API god-knows-what
15:58 < Miguel2013> anybody knows why my wireless n card v1 from 2009 works on usb2 ports but not on 3
15:59 < Miguel2013> on usb3 port or a hub it says something about unable to run driver
16:00 < djph> zeldafan78: it's not wrong, it's "their API" so they can do whatever they want
16:02 < djph> Miguel2013: your USB3 ports are fubar?
16:02 < Miguel2013> insignia
16:02 < djph> "FUBAR"
16:03 < Miguel2013> it looks very good and strong
16:03 < Miguel2013> like is tight
16:03 < Miguel2013> my usb2hub is from linksys
16:03 < djph> that doesn't mean that the controller or anything else is actually "good"
16:03 < Miguel2013> u think there's a bad usb hub
16:04 < Miguel2013> I don't think there is much to accuse
16:04 < Miguel2013> it works fine from the usb2hub
16:04 < djph> or the driver is fubar, or any number of things that can go south.  But since it works in the USB2.0 port, call it a day.
16:05 < Miguel2013> my usb2hub is kinda unstable at least for connecting sata to usb
16:05 < Miguel2013> the usb3hub did great
16:05 < Miguel2013> tho
16:05 < djph> so then don't use it for that...
16:05 < Miguel2013> k k friend
16:06 < Miguel2013> just thought there is a easy fix it's a tplink n usb card and oficialy there is not win10 driver for it
16:06 < Miguel2013> is an older version
16:06 < drac_boy> hi
16:07 < djph> there is. USE. THE. FUCKING. USB2. PORT.
16:07 < Miguel2013> but eventualy I must know why it doesn't work on the usb3.0 hub
16:08 < Miguel2013> I wanna know the reason
16:08 < djph> because it's fucking old, and TP-Link doesn't fucking care to release a newer driver
16:09 < djph> ... will you be needing us to tell you why you have to eat, sleep, and breathe next?
16:09 < Miguel2013> but usb3.0 suppose to be backward compatible
16:09 < djph> good god, you're a dense motherfucker, aren't you?
16:09 < Miguel2013> I seem stuff like this happen between the 2 with bootable usb drivers too
16:10 < Miguel2013> I'm not trying to make u angry
16:10 < Miguel2013> I don't think I'm anoying
16:10 < djph> you're asking the same fucking question in as many ways as you can, because you don't like the answer.
16:10 < djph> answer is - it doesn't work, because the manufacturer doesn't care.
16:11 < Miguel2013> I thought maybe microsoft will patch it?
16:11 < Miguel2013> they're friendly
16:11 < djph> don't hold your breath
16:11 < UncleDrax> best results are when you email Bill Gates and ask nicely.. and attached a $1million USD check to your request
16:12 < djph> UncleDrax: all I ever get back is goats.... ohhhhh, maybe it's because I never send the cheque
16:13 < zeldafan78> Sigh...
16:14 < Miguel2013> yea thinking twice it doesn't seem posible
16:14 < Miguel2013> they sometimes make drivers tho for generic or make it generic driver
16:15 < zeldafan78> Is there a mail sending company that isn't run by morons who cannot explain the way to the exit if their life depended on it?
16:15 < zeldafan78> I'm sick of guessing what the fuck Mailgun means.
16:16 < drac_boy> mailgun?
16:16 < zeldafan78> Also, different e-mail question: if I need to include a "unique security code" in e-mails, can I send that as some kind of header instead of making it part of the actual text contents (to not look so messy)?
16:16 < Miguel2013> time to disconect from the internetr
16:17 < djph> zeldafan78: "probably", although sticking it in the body is perfectly fine.
16:17 < djph> zeldafan78: read their manual then ... their manual for their service is the only thing that matters.
16:18 < zeldafan78> Where do you think I'm looking? Obviously I'm reading and quoting from their "manual".
16:18 < zeldafan78> If you can call that word salad a "manual".
16:18 < djph> I mean, even just looking at their site, the thing is stupid simple --> curl user 'apikey' \ -option "value" \ -option "value" \ -option "value" \ -whatever
16:19 < derpingit> hi giys. dumb 101 question here, but please bear with me. i have a 200/20mbit connection at home, and i'm looking to stream netflix through a vpn provider that uses openvpn, alongside my dmz traffic (for guest wifi). my current hardware (er-x) does no more than 10mbit on openvpn performance, so i was thinking pfsense. does my 20mbit upload mean i should only be looking at hardware that does 20mbit openvpn since that's the most i can get? or 200mit sinc
16:19 < jvdmr> zeldafan78: have you tried just testing the service? like sending an e-mail to your software and checking what you actually receive?
16:19 < zeldafan78> djph: ... what?
16:19 < djph> derpingit: that might work
16:19 < GodOfsea> Hello
16:19 < zeldafan78> jvdmr: This is about receiving e-mails and it's ambiguous whether it can include multiple recipients and what that means.
16:20 < djph> zeldafan78: checked their website ... the frontpage has a dead simple example of how ther headers work (if you use curl)
16:20 < UncleDrax> derpingit: well the 20mbit is your upload.. so are you intending to upload through your VPN beyond 20mbit? that said, most netflix streams I think are in the 3-5mbps range (outside of '4K' streams.. no idea what those are).
16:20 < zeldafan78> ..
16:20 < derpingit> djph what might work. sorry for not framing the question correctly: do i need 20mbit openvpn hardware? or 200mbit hardware
16:20 < zeldafan78> djph: I really don't even know how to respond to that.
16:20 < jvdmr> zeldafan78: I know, I saw the conversation here
16:20 < djph> derpingit: more is more etter
16:20 < GodOfsea> Failed password for invalid user ubuntu from 102.177.165.71 port 56976 ssh2      I am seeing this in journalctl-xe pretty sure thats not my ip , is someone trying to log in ?
16:20 < zeldafan78> Seems to have nothing to do with what I said. Too tired to try to explain everything again.
16:21 < zeldafan78> jvdmr: Then that makes no sense.
16:21 < jvdmr> zeldafan78: I'm just saying, if the documentation isn't clear, just try actually using it and maybe it'll work, or maybe you'll get an error that's more clear than the documentation
16:21 < derpingit> UncleDrax mostly consuming bandwith trhough streaming and browsing
16:21 < zeldafan78> jvdmr: My project isn't anywhere close to being at the test stage.
16:21 < detha> derpingit: running streams from some vps to your network, or pulling streams into your network and pushing them somewhere else?
16:22 < derpingit> no pushing anywhere
16:22 < djph> zeldafan78: curl -s --user 'yourkey' \ \n theirurl \ \n -F from="from" \ \n -F to="first person" \ \n -F to="second person" [...[
16:22 < djph> zeldafan78: granted, there are probably other options, but hey that's right there in the doc under "sending via API"
16:22 < derpingit> i'm thinking of getting a seedbox, and that traffic will be download as well (i think?)
16:23 < derpingit> thank you for your input
16:23 < jvdmr> zeldafan78: I'm just not sure how you're stuck on this, just work past it in a way that you can easily change to the other option later if need be?
16:23 < jvdmr> like, implement it using a list (for multiple recipients), then if it turns out it's not a list and always just 1 recipient, change the implementation to not use a list
16:24 < jvdmr> can't be too hard...
16:24 < detha> derpingit: if you have 200Mb download, and you are downloading things through a vpn, look for something that does 200Mb vpn performance
16:24 < djph> jvdmr: hell, the "list" is just multiple invocations of '-F to="emailaddress"'
16:24 < djph> jvdmr: at least with curl
16:26 < derpingit> thank you, detha
16:27 < djph> zeldafan78: you have a link to the specific page you're having trouble with?
16:43 < DocScrutinizer05> slightly OT: can DNS or a HTTP server reply forward a (HTTP) connection to another *port*?
16:44 < DocScrutinizer05> forward/redirect
16:46 <+catphish> http certainly can with a 30x redirect
16:46 < TandyUK> erm sort of, look at mod_proxy for example
16:46 < DocScrutinizer05> DNS not?
16:46 < TandyUK> or a redirect yeah
16:46 <+catphish> no, dns can't
16:47 <+catphish> a browser will only ever start by going to 80 or 443 (unless the user enters a port manually)
16:47 <+catphish> DNS can't change that
16:47 < DocScrutinizer05> purpose: https://foo.dyndns.bar  -> <some-IP>:<some-port>
16:47 <+catphish> yeah, you need an http redirect for that
16:47 < DocScrutinizer05> ta!
16:47 < TandyUK> your webserver however can do this by proxying the request
16:47 <+catphish> or an http proxy as TandyUK says
16:48 < TandyUK> so either client requests http://foo.bar.dns,  server responds with a 301 HTTP redirect to https://ip:port
16:48 < TandyUK> or, client connects to server, server connects to ip:port, and proxies the request back
16:48 < DocScrutinizer05> yup, that's it then
16:48 < DocScrutinizer05> first one
18:08 <+catphish> quiet in here today
18:09 < djph> shhhh
18:13 < skyroveRR> Let them sleep!
19:05 < jenia> hello everyone
19:08 < jenia> I don't understand completely why the need 4 IMS. In the book I'm reading it says: the real strength of IMS is that it creates a service environment where any service can access any aspect of the session. This allows service provided to create far richer services... wtf does that mean? Programs allows others programs to "access any aspect of the session". What do networks have to do with it?
19:08 < jenia> This allows service providers*
19:12 < BlacksmithMetash> hi
19:12 < BlacksmithMetash> i need some help
19:12 < BlacksmithMetash> i want traceroute my broadband to see which ip is stealing most of the traffic, slowing the internet. i can see this with wireshark?
19:13 < Drzacek> Hi
19:13 < Drzacek> I need a free domain
19:13 < Drzacek> dot.fscking.tk is a no go since I lost one domain there already
19:14 < Drzacek> it can be weirdest, longest domain there is
19:14 < UncleDrax> jenia: you're gonna have to define what IMS means. you talking about the IBM Product?
19:15 < jenia> IP Multimedia Subsystem
19:15 < detha> jenia: sounds like sales droids were involved in creating that text
19:15 < jenia> detha: lol. that's what I think too
19:15 < jenia> :)
19:16 < UncleDrax> ah.. radio network stuff.. definately a thing I Know Nuth Zing on.
19:17 < jenia> :)
19:18 < UncleDrax> BlacksmithMetash: you want to see what other host in your network is consuming all your bandwidth? If so - that's not traceroute. that would be a tool like Netflow/sFlow/IPFIX you enable on your router (if it supports it, or a similar reporting mechanism to show per-host utilization)
19:23 < BlacksmithMetash> UncleDrax:
19:24 < BlacksmithMetash> i want to see which ip's is using most of my bandwidth
19:24 < BlacksmithMetash> netflow/sflow/ipfix is tools for linux?
19:25 < tpr> they are protocols for reporting about network flows, you are supposed to run one on your router
19:25 < UncleDrax> they are reporting .. ya that
19:25 < BlacksmithMetash> okay
19:25 < tpr> pmacct is something that's available for linux too, but I haven't really delved into it that much
19:25 < BlacksmithMetash> i have cisco switches
19:25 < BlacksmithMetash> can i enter in them?
19:25 < Apachez> BlacksmithMetash: our deepest appologies
19:26 < Apachez> or how you say that in english
19:26 < tpr> BlacksmithMetash: https://en.wikipedia.org/wiki/NetFlow
19:27 < BlacksmithMetash> thanks tpr
19:27 < tpr> none of it is really not targeted for end users though :P
19:27 < BlacksmithMetash> now i have to see if i can see this traffic on my computer
19:28 < tpr> if you have a capture wireshark can also show some statistics about what it is seeing
19:28 < BlacksmithMetash> based on my gateway
19:28 < tpr> https://wiki.wireshark.org/Statistics
19:28 < BlacksmithMetash> tpr... i can do both of theses things, ya?
19:28 < tpr> sure, whatever floats your boat :-)
19:28 < tpr> wireshark may be easier for testing though
19:28 < BlacksmithMetash> network flow see who is using the most and see which packs are been read and sent
19:29 < Apachez> tpr: tears from snowflakes?
19:29 < BlacksmithMetash> i think so
19:29 < tpr> Apachez: pardon me?
19:29 < Apachez> <tpr> sure, whatever floats your boat :-)
19:29 < BlacksmithMetash> i still gonna do the course
19:29 < tpr> ummh, I'm not getting it.
19:29 < BlacksmithMetash> excuse me
19:29 < BlacksmithMetash> i'm here away
19:30 < BlacksmithMetash> thanks everybody!
19:30 < Apachez> tears from snowflakes have a negative boiance
19:55 < alabaster> Hello all I am so lost. Would it be to much to ask for small insight into using GNS3??
19:57 < alabaster> I guess that's to much of a loaded question
20:04 < detha> alabaster: don't ask to ask, ask the question
20:07 < alabaster> I don't even know where to begin
20:07 < alabaster> Detha have you ever used GNS3?
20:08 < detha> alabaster: long ago. vague memories of 'get the timeslice thing right'
20:09 < alabaster> I guess it could go in so many ways. like the correct host binding
20:24 < SpeakerToMeat> Hello
20:26 < SpeakerToMeat> Very very stupid question... I'm trying to see if there's some sort of networking problem in a lan i have no control, the behaviour could be explained by networking. So far I'm trying mtr runs and longish (1 minute) iperf runs, so far in tcp I'm gonna try in udp, mtr has a few rare response time spikes in the first jump, it's a two jump setup it seems, (3 really, machine->switch->switch->machine), if I
20:26 < SpeakerToMeat> do believe it's a sporadic network problem, comming from switch overload or troubles and not wiring, and if I discover the switches both support jumbo packets, do you think setting jumbo packets, to send fewer, larger packets, might help, or do nothing?
20:27 < SpeakerToMeat> I can't get much help form the network admins in the site for this.
20:29 < ||cw> bigger packets can help if the data suits it and a it's a PPS issue and not just a bandwidth issue
20:30 < SpeakerToMeat> Nod
20:30 < ||cw> but unless yo can get switch metrics, it's all a guessing game.  maybe see if SNMP public is enabled, assuming you can find the switch's management IP
20:30 < SpeakerToMeat> I'll test if the jumbo frames are on, and if its a pps issue.
20:31 < SpeakerToMeat> Hmmm not a bad idea... these guys should have a dashboard in theory, but they're doing ip assigment tracking via excel files so.
20:31 < ||cw> at least they arne't using hosts files
20:33 < SpeakerToMeat> Are firewalld questions ot here?
20:36 < SpeakerToMeat> ||cw: iperf3 in tcp for a 60 second run gives me 654 to 702 (two runs) Mbit transfer, but it gave me 382 and 276 retrs in each run respectively... that's not good, is it?
20:36 < SpeakerToMeat> eh sorry to target you
20:36  * SpeakerToMeat opens the question to the room at large
21:02 < ppf> can anyone explain DNS to me?
21:02 < ppf> i have a domain foo.com
21:03 < ppf> given i have a server at 1.2.3.4, i can of course set an A record to point there
21:03 < ppf> but what if i want subdomains?
21:03 < ppf> do i add an NS record for the same address?
21:04 < ppf> actually, NS needs to be a name, right? so can i set NS to ns.foo.com?
21:04 < ppf> who resolves that name then?
21:04 < grawity> if the NS target is below the domain it handles, then the parent domain has a 'glue record'
21:04 < grawity> i.e. a copy of the A/AAAA records
21:05 < grawity> and if you want subdomains, no, you don't need to NS every one of thme
21:05 < grawity> unless you intend to host them on different DNS servers, that is
21:07 < ppf> i meant, if i want to run the nameserver for the subdomains on the same ip
21:09 < grawity> I still don't get what you mean
21:27 < ppf> yeah me neither :)(
22:16 < tyrsen> is this a place to ask about setting up webservers and such or is it only for lans
22:18 < ||cw> tyrsen: as it relates to IPs and ports it's on topic here. but a channel for your specific server software is likely better
22:18 < tyrsen> i don't have a server yet
22:19 < tyrsen> i'm looking at buying a domain to set up my own mailserver etc
22:19 < tyrsen> just wanted to ask some questions about the process surrounding that
22:19 < Apachez> sure
22:19 < Apachez> dont ask to ask
22:22 < tyrsen> well i'm looking at some very short domain names, like a couple letters
22:22 < tyrsen> most of them end up of course being from other countries
22:22 < tyrsen> am i likely to have any problems as a result of that? wikipedia says the ones i'm looking at have no restrictions
22:45 <+pppingme> tyrsen people like ".com" and such, they don't like visiting country based domains, unless they are in that country
22:46 <+pppingme> most people would say they feel scammy
22:46 < manny_> Hi. Does anybody know the standard prices in europe for a 10G fiber uplink (MPLS or similar) ?
22:46 <+catphish> tyrsen: it's extremely unlikely you'll get a 2-character domain name anywhere
22:46 <+pppingme> even when you print the domain name on advertising, letterhead, etc..  people still like to add .com to it..
22:46 < tyrsen> there's loads available
22:46 <+catphish> tyrsen: example?
22:46 <+pppingme> catphish I think he means "histname.cc"
22:46 < tyrsen> i mean like x.yz
22:47 < tyrsen> one letter and a 2 letter tld
22:47 < tyrsen> $10-30/year
22:47 <+catphish> tyrsen: where are you looking?
22:47 < tyrsen> namecheap.com
22:47 <+pppingme> you aren't going to find a single letter domain like "x", regardless of tld
22:47 <+catphish> that seems unlikely
22:47 < tyrsen> i'm literally sitting on the checkout page
22:47 < tyrsen> you can find like 2 dozen at least for any letter
22:47 <+pppingme> for a single letter domain??
22:47 <+catphish> namecheap is a domain seller, what domain are you looking at?
22:47 < tyrsen> yes
22:48 < tds> what tld is this for?
22:48 < tyrsen> i'm not sure it's wise to tell you :D
22:48 <+pppingme> I'd bet the transaction either doesn't complete, or gets revoked later..  single letter names are generally reserved
22:48 <+catphish> well if you can really buy a 1 character domain anywhere, buy it
22:48 <+catphish> it will be worth serious money imo
22:48 < tyrsen> that particular country's tld on wikipedia says it's free to register for anyone, you don't have to have local ties or residency or anything
22:48  * pppingme is betting it will get revoked right after he buys it
22:49 <+catphish> more likely it simply isn't valid
22:49 < tyrsen> well this is why i'm asking
22:49 <+catphish> well try it if you like
22:49 < tyrsen> say it does  get revoked after i pay for it, i'm assuming i can get a refund?
22:50 < WeirdTolkienishF> you guys ever play around with multicasting?
22:50 < tds> if you don't want to tell us the tld, then you should be able to find their regulations yourself
22:50 <+pppingme> WeirdTolkienishF describe your problem and question..  thats too generic
22:50 <+catphish> anyway, TLDs from obscure countries are less valuable than .com but lots of people use them because they can be short and cool
22:50 <+pppingme> tyrsen generally, I'd expect that..
22:50 <+catphish> no way you're getting *1* character though
22:50 < WeirdTolkienishF> pppingme, lol, no problem
22:50 < WeirdTolkienishF> just wondering
22:51 <+catphish> WeirdTolkienishF: many people here have used multicast
22:51 < WeirdTolkienishF> ive never used it
22:51 < tds> the only time I'd expect to see a 1 character domain available with a registrar is if they have a partnership with sedo/similar :P
22:51 <+pppingme> you probably have and don't even realize it
22:51 < tds> and idk if there are even many 1 char domains for sale via that kind of platform
22:51 <+pppingme> WeirdTolkienishF  you probably have and don't even realize it
22:51 < WeirdTolkienishF> it's possible, never set it up though
22:52 <+catphish> https://en.wikipedia.org/wiki/Single-letter_second-level_domain
22:52 <+pppingme> WeirdTolkienishF thats half the idea, it typically doesn't need much setup depending on what you're doing
22:52 < tyrsen> catphish: i don't think you understand. just go at https://tld-list.com/ and put in a single letter., and set the tld length to 2
22:52 < WeirdTolkienishF> well i've done udp broadcast
22:52 < WeirdTolkienishF> i guess that's multicast
22:52 < tyrsen> you can scroll through the list, 95% are gone but out of 2000 tld's that still dozens for any letter
22:52 < WeirdTolkienishF> just reading wikipedia
22:52 <+pppingme> not quite the same thing
22:52 < ||cw> WeirdTolkienishF: there's not really much special about using it
22:52 <+catphish> tyrsen: i understand, i just bet its a mistake
22:53 <+catphish> The list of Country code top-level domains which have been identified to allow single-letter domains are : .ac .af .ag .ai .am .asia .biz .bo .by .bz .cm .cn .co .com.ar .com.br .co.at .co.il .co.uk .cr .cz .cx .dj .de .dk .fm .gd .gg .gl .gp .gs .gt .gy .hn .ht .ie.[2] im .io .je .kg .ki .kw .la .lb .lc .ly .md .mg .mk .mp .ms .mW .mx .mu .nf .nz .pe .ph .pk .pl .pn .pr .pw .ro .sh .st .tc .tl .tt .to .tv .travel .ua .ws .vc .vg .vn and .vu
22:53 <+catphish> so if one in that list is showing as available, cool i guess
22:53 <+pppingme> WeirdTolkienishF do you have any "network printers" especially consumer class?  have you ever zapped a youtube video from your phone to your tv?
22:53 < tyrsen> yep, it's in that list
22:53 <+catphish> well, if you want it, buy it
22:53 <+catphish> you'll be getting a great deal IMO
22:53 < WeirdTolkienishF> that uses multicast?
22:54 <+pppingme> both those use it, as well as 100's of other things
22:54 < ||cw> WeirdTolkienishF: windows printer discovery does, yeah
22:54 < tds> it seems pretty unlikely you'll lose out on anything from attempting to buy it, so why not :)
22:54 <+pppingme> tyrsen how much to register?
22:54 < ||cw> open wireshark, set filter to multicast, listen away
22:54 < tyrsen> like 30 bucks a year
22:54 <+pppingme> thats not a big hit if it goes south
22:54 <+catphish> tyrsen: you're not going to lose anything by trying, and i'd say a single character domain is always cool
22:55 < tyrsen> yea, i'm thinking i may as well buy a 10 year registration immediately, in case they try to hike the price later
22:56 <+catphish> that's not very common, but sure
22:57 <+pppingme> tyrsen you aren't going through go-daddy are you?
22:57 < tyrsen> no, i'm using namecheap to buy
22:57 <+pppingme> they have a bad rep about buying at one price, and renewing for a higher price
22:58 <+pppingme> but overall, thats not common at all
22:59 < tyrsen> eh, screw it, may as well give it a go
22:59 < tyrsen> as you said, in a decade who knows, this might be a better investment than bitcoin :D
22:59 < tds> well please let us know if you manage to get it :)
23:00 <+catphish> domains were a great investment, in 1992 :)
23:01 < tds> there are lots of high value .co.uk/similar domains, I'll be interested if any of those people forget to buy the .uk ones before the deadline (next year iirc)
23:02 < tyrsen> yea, next june
23:03 <+catphish> tds: yeah those UK ones will sell fast
23:10 < tds> catphish: yeah, I suspect lots of people will automate getting them though :/
23:10 <+catphish> tds: the high value ones, definitely
23:11 < tds> it's almost tempting to try and script it, that seems like a stupid game to play though
23:12 <+catphish> interestingly, my company has not taken up our .uk trademark
23:12 <+catphish>     Right of registration:
23:12 <+catphish>         atechmedia.co.uk
23:12 <+catphish> we never use UK domains, so meh
23:13 < tds> oh, you're using one of the fancy new gtlds, those seem to be used pretty rarely
23:13 <+catphish> if we ever wanted it, i imagine it would be pretty hard for anyone else to ever claim they were using it in good faith
23:14 <+catphish> well in our case, it was a pretty obvious choice :)
23:14 < tds> yeah, I'm sure you could get it back pretty easily, you'd have to go through the little nominet fake court process :P
23:16 < tyrsen> ah drat, it didn't let me get it
23:16 < tds> what a surprise ;)
23:16 < tyrsen> support contacted me and said, that even though it shows as available, they don't let anyone buy a domain shorter than 3 letters from them
23:17 < tyrsen> which is really stupid
23:17 < tds> what was it?
23:19 < tyrsen> d.mx
23:19 < tyrsen> a guy can dream, eh?
23:19 < tyrsen> i quite like the .mx ending as 'mail exchange' if i'm just going to use it for personal email
23:19 < tyrsen> might go hunting a .io one instead though
23:21 < tyrsen> there's a lotta really snazzy ones available for practically free if you don't mind a longer tld
23:21 < tyrsen> e.g. i can have my initials .cloud or .space or .tech etc etc etc
23:21 < tyrsen> but that's kinda dumb
23:21 <+catphish> there are 26 letters in the alphabet, assuming there are also about 26 TLDs that offer 1-character registrations, at $20 a year each, it would cost approx $13,000 to register all of them
23:21 <+catphish> that's basically no money to a speculative domain investor
23:22 < tyrsen> yea, but a guy can dream
23:22 < qman__> One hot covers the investment easily
23:22 < qman__> hit*
23:22 < tyrsen> funny thing is they were quite happy to charge my paypal before saying the registration failed
23:22 < tyrsen> i've immediately filed a claim, but that's extra dumb
23:23 <+catphish> tyrsen: you could try tweeting @elonmusk and asking him if you can have x.com
23:23 <+catphish> he's a pretty cool guy
23:23 < tyrsen> no, he's really not
23:23 < qman__> What, you think they're going to register it before being paid?
23:23 < tyrsen> i think they'd feckin check
23:23 < tyrsen> if their system would even allow it
23:24 <+catphish> you literally can't get any cooler than the guy who sold paypal to start an electric supercar company and a space company :)
23:24 < tyrsen> he's literally the real life version of a supervillain e.g. lex luthor
23:25 <+catphish> i always thought he was like Hank Scorpio
23:25 < tyrsen> why do you think he is putting so much money into high density energy storage... obviously for the deathrays
23:25 <+catphish> although Hank is based on Richard Branson
23:25 < tyrsen> why do you think he is putting so much money into rocketry... obviously for the moon base
23:25 < tyrsen> i could go on but i shan't
23:27 < tyrsen> so then what do you reckon is the least dumb of the various .word tld's? .space? .life? .solar? .world? i'm leaning towards .cloud ...
23:27 <+catphish> depends on context
23:27 < qman__> .cloud is super buzzwordy and will date it
23:27 <+catphish> if your company is called atech media then atech.media makes sense :)
23:27 < tyrsen> it's not for a company
23:27 < tyrsen> it's for my own personal email address
23:28 <+catphish> if you're gay and called tyrsen then tyrsen.gay might work
23:28 < tyrsen> i'm not cat, and if i was i wouldn't broadcast it like that
23:28 < tds> I've heard of some people having email sent from the new gtlds blacklisted by default, so I'd be slightly wary of that
23:28 < tyrsen> ah, hmm
23:28 < qman__> I haven't had any issues
23:29 < tyrsen> currently i use hushmail as my email provider
23:29 <+catphish> tds: the problem is actually more one of validations not accepting new TLDs in email addresses
23:29 < tyrsen> it's a very short ###@hush.ai
23:29 < tyrsen> which is quite nice
23:29 < qman__> Just set your shit up correctly, use spf and dkim+dmarc
23:29 < tyrsen> and they're a decent canadian company
23:29 < tyrsen> which i trust more than most places tbh
23:29 <+catphish> tds: i got refused car insurance at ta car dealer last year, they couldn't say why, i asked them to try applying again with a .com emil address and it worked :|
23:29 < tyrsen> i just thought i'd go in on something that's uniquely me
23:29 < tds> catphish: that's annoying :/
23:30 <+catphish> *at a
23:30 < qman__> Heh
23:30 <+catphish> and yes, it can be annoying, fortunately i have a .com backup :)
23:30 < tyrsen> yea i have a gmail backup
23:30 < tyrsen> which i use for registering unimportant things
23:30 < qman__> I've got a .net and a couple .orgs
23:30 <+catphish> the stupid thing is that .longword has ALWAYS been a technically valid TLD
23:31 <+catphish> so it never made sense to reject it
23:31 < tyrsen> i can have my initials .org for only £19k
23:31 < tyrsen> it's.... tempting? llol no
23:31 < qman__> Hah
23:31 <+catphish> and you're sure you're not gay?
23:31 < tyrsen> are you?
23:31 < tyrsen> you seem rather obsessed
23:31 < tds> if the last letters of your last name are a valid tld then that's a decent idea
23:31 < qman__> and I complain about $9 a year
23:31 < tds> ie ty@rs.en
23:32 < tyrsen> tyrsen is not my real name
23:32 < tyrsen> who the heck uses their real name on the internet ?
23:32 < tds> I know, and I don't think en is a real tld either :P
23:34 <+catphish> most people use their real name on the internet
23:34 <+catphish> though it baffles me that they do
23:34 < tyrsen> maybe i'm just paranoid
23:34 < qman__> Yeah, that changed when most people using the internet stopped being tech savvy folks and started being regular people
23:34 <+catphish> i'm not particularly precious about it, but you wouldn't believe the number of people who are simultaneously paranoid about their privacy / anonymity but also use facebook with their real identity
23:35 < tyrsen> like, i literally have no social media accounts except a 'john smith' kinda facebook for those few things which need a facebook registration
23:35 < tyrsen> i'm quite proud of the fact that if you google my name you get literally no results
23:35 < tyrsen> that actually helped me get my last job, too, cause the head was super impressed by that
23:35 <+catphish> of course, my real identity is top secret and i would never eveal it here
23:35 <+catphish> *reveal
23:37 < S_SubZero> well that means your name clearly isn't John Smith,  Narrowing it down!!
23:37 < tyrsen> that was not the right butten
23:37 <+catphish> my real name isn't Cat Phish
23:38 < tds> oh no, now you can't use the domain cat@phi.sh
23:38 < qman__> my name does show up in a google search, but mist of it is other people by the same namw, and even the ones that are me aren't very accurate
23:38 < qman__> I don't use any social media
23:39 < tyrsen> i'm entirely certain i'm the only person in the history of persons to have ever had my particular name
23:39 < tyrsen> given the aforementioned lack of search results
23:39 <+catphish> https://www.google.co.uk/search?q=charlie+smurthwaite
23:39 < tyrsen> given that i'm a multicultural kinda fellow and have a double barrel surname
23:39 < tyrsen> it's entirely unique
23:39 <+catphish> whoever that is, they get a lot of hits
23:40 < tds> in general I find it too much effort to stay anonymous on irc/whatever, so I'd rather just not bother
23:41 < tyrsen> it depends what you use it for
23:41 < qman__> Trying to stay completely anonymous is a losing battle, but at the same time I have nothing to gain by advertising my identity
23:41 < tyrsen> if you are involved in programming projects and have to link repos and such, sure
23:41 < tyrsen> if you're just here for some q&a that's different
23:44 <+catphish> people's identities rarely come up here
23:45 <+catphish> so it rarely matters
23:51 < tyrsen> on another note, just got my refund through paypal, case closed :D
23:51 < tyrsen> obviously they knew they weren't winning that one
--- Log closed Wed Jun 06 00:00:46 2018