--- Log opened Mon Jun 18 00:00:02 2018
00:52 < codyrutscher> I have a question
00:52 < drac_boy> hi
00:53 < drac_boy> sorry if this is silly question but is there ever such thing as outdoor cat-5/-6 location or not so much? like I mean waterproof rj45 connections and wind/rain-tolerable ports
00:54 < RJ45> drac_boy: wow, that is such a dumbshit question
00:55 < RJ45> drac_boy: you trully are a neanderthal bone-headed retard
00:55 < RJ45> such a tragic waste of a human brain
00:55 < RJ45> (JK)
00:56 < RJ45> drac_boy: funny enough though, I actually stumbled upon one of those yesterday on AliExpress
00:56 < RJ45> yeah, they're a thing, not really standardised or anything, but a thing for sure
00:57 < codyrutscher> can i ask a question
00:57 < codyrutscher> in all seriousness
00:58 < RJ45> https://www.aliexpress.com/item/IP68-RJ45/32750619132.html
00:58 < RJ45> codyrutscher: no
01:00 < drac_boy> rj45 hm yeah interesting aliexpress source. anyway I was just wondering .. I know that usually remote network towers have a sealed waterproof box that you put all the normal cables&hardware in then effectively use a lot of rubber to seal the front access door and any grommet holes .. right? :)
01:00 < RJ45> drac_boy: you f****** idiot
01:01 < RJ45> I mean, sure, lol
01:01 < RJ45> (´･ω･`)
01:01  * drac_boy smacks rj45 back stoogies-style for saying that :P
01:01 < drac_boy> heh
01:01  * RJ45 drops a few packets
01:01 < drac_boy> anyay rj45 doing anything atm or just a quiet time now?
01:02 < RJ45> ah jus chillin, watchin a cockumentary about Japan orphans
01:02 < scientes> my moca network isn't working but internet is. What am I doing wrong
01:03 < RJ45> scientes: if the mocca isn't working, try a latte or espresso instead
01:04 < drac_boy> rj45 or java if you don't mind having to deal with oracles afterward :P
01:04 < drac_boy> heheh
01:05 < RJ45> funny thing, IRL I can't stand hot drinks, at-all, I only drink cold soft drinks
01:06 < drac_boy> rj45 not just you .. health-wise I only have water, milk, or certain soft drinks
01:07 < RJ45> I don't avoid hot drinks for a health reason though, I just can't stand them
01:08 < RJ45> but I'm really weird with food and drink, I also hate cheese and cream, despite not being lactose intolerant .
01:08 < codyrutscher> security question
01:08 < RJ45> yup, I'm that one fucko who orders a damn cheeseless pizza
01:33 < drac_boy> rj45 I've had a few pizzas with no tomato sauce used :)
01:34 < RJ45> no tomato sauce?, not even barbecue sauce??
01:36 < drac_boy> basil sauce twice and I forgot about the others :)
02:36 < BenderRodriguez> it should be a federal crime to eat pizza that's lacking tomato sauce
02:41 <+pppingme> alfredo sauce is a big thing around here..
03:45 < strixdio> If I use a raspberry pi as a wifi -> ethernet bridge so my pfSense box can use wlan as a 2ndary WAN, would that perhaps make the rpi vulnerable, or perhaps "more" vulnerable, if I'm connecting to my ISPs wifi?
04:40 < darkmeson> strixdio: thanks to issues like hole 196, sharing wifi with untrusted third parties DOES bring with it far more potential insecurities, but that's more to do with your "ISP's wifi" than the device that's connecting to it
05:03 < Nitter> hey have you used httpbin.org?
05:03 < Nitter> I'm doing some silly tests, but I'm surely missing something
05:03 < Nitter> $ nc httpbin.org 80
05:03 < Nitter> then I send this:
05:03 < Nitter> GET /delay/5 HTTP/1.1
05:03 < Nitter> Host: httpbin.org
05:03 < Nitter> Connection: keep-alive
05:03 < Nitter> Accept: application/json
05:03 < Nitter> Accept-Encoding: gzip
05:03 < Nitter> and I get this: HTTP/1.1 505 HTTP Version Not Supported
05:05 < myxenovia> hi
05:05 < myxenovia> im trying to mixed sound through voip
05:05 < myxenovia>  is it possible to mix pcm raw data with wav file data?
06:11  * rE-BoOt wonders why this place is so silent...ಠ_ಠ
06:16 < winsoff> net's down, rE-BoOt
06:21 <+pppingme> rE-BoOt start talking, then it won't be silent
06:23 < rE-BoOt> HELLO Darkness, my old friend..
06:23 <+pppingme> E1ephant around?
06:27 < majculluh> water
06:28 < adleff> anyone know if 802.3ae/D3.2 is mandatory in the standard?
06:28 < adleff> in other words, do I really need to turn on udld on 10G links?
06:30 < majculluh> that requires equipment
06:30 < majculluh> tonjo
06:30 < adleff> what?
06:31 < majculluh> you must pay to know these answers
06:31 < majculluh> your looking at fallout crystals
06:31 < adleff> are you referring to the ieee documents? if so I am aware
06:31 < majculluh> they are nearly all unique
06:32 < majculluh> the hardware
06:32 < majculluh> pay for tests
06:32  * adleff casts hurtmore at majculluh 
06:33 < majculluh> beautiful lie, huh?
06:35 < majculluh> they might have life
06:35 < majculluh> or maybe?
06:35 < majculluh> keep locking memories, in search of a companion
06:36 < majculluh> "in search of" with leonard nimoy
06:37 < majculluh> bro. frank can say "clean water"
06:37 < melissa666> what would be equivalent of this command using ip instead of ifconfig? --> `ifconfig wlan0:avahi netmask 255.255.255.255`
06:37 < majculluh> mel
06:37 < majculluh> why you rw rw rw
06:37 < majculluh> you need work for me mel
06:38 < majculluh> 700 club
06:38 < majculluh> might have life
06:39 < melissa666> a life with you as a boss wouldn't be worth living.
06:39 < majculluh> maybe, huh? how deep can you go
06:39 < majculluh> deep as I?
06:39 < melissa666> majculluh, You read my mind. Exactly.
06:39 < majculluh> what if you met the Father of I
06:40 < melissa666> majculluh, Well, generally the fathers of people such as yourself are even worse.
06:40 < majculluh> life can have a much deeper meaning
06:41 < majculluh> how deep and perfect and crystalline can your memory go
06:41 < majculluh> and then what do with it
06:41 < melissa666> My mind is a shimmering crystal majculluh. It's very shiny.
06:41 < melissa666> That's what shimmering means.
06:41 < majculluh> but rw,rw,rw
06:42 < majculluh> what is a free agent
06:42 < melissa666> None of us have free will. We are all controlled by Jeebus' daddy.
06:43 < majculluh> what if life does have free will
06:44 < melissa666> majculluh, I know you're lonely and craving attention, and I was taking pity on you. But I have to go now. I suggest you take up roller derby if you want to make some friends.
06:44 < majculluh> might have life?
06:44 < melissa666> majculluh, probably not. but afterwards, there is eternal sleep.
06:44 < melissa666> cya
06:45 < majculluh> melissa
06:45 < majculluh> light me
06:45 < majculluh> me light
06:46 < majculluh> nonina
06:46 < majculluh> started a war broke the rules
06:47 < majculluh> meliza!
06:49 < majculluh> the publinux is working
07:10 < senaps> i have my centos 7 machine  reciving ICMP echo reply(tcpdump) but i don't have the ping command showing any data. how can i check why is that?
07:11 <+pppingme> senaps what are you expecting to see?
07:12 < senaps> ping responses. the ping command doesnt return anything. just blank screen unticl i ctrl+c it. but tcpdump shows that the data is comming through the interface!
07:12 < senaps> i am testing some routing configs.
07:13 < senaps> pppingme ^^^
07:13 <+pppingme> sounds like firewall
07:14 < senaps> firewall's are clear. i have disabled it. i just got one reply back after about 20-30 seconds, no other replies has come yet
07:15 <+pppingme> whats the time on that reply?
07:16 < senaps> 64 bytes from 192.168.30.30: icmp_seq=186 ttl=127 time=0.549 ms
07:16 < senaps> 463 packets transmitted, 1 received, 99% packet loss, time 461998ms      |    rtt min/avg/max/mdev = 0.549/0.549/0.549/0.000 ms
07:17 < linux_probe> lol
07:18 < linux_probe> how r to firewall
07:18 < senaps> we have traceroute disabled on firewall, and another machine in my subnet can ping the machine im pinging without problem. whatever the problem is, it's mine.
07:47 < senaps> https://pastebin.com/c80Ju9L1  this is my route table, `.30.X` and `.50.X` subnets have routes, so two machines on these subnets can ping and request/reply each other. but my machine with 2 interfaces on both subnets can't. i have huge packet lost! and can't ping machines on other subnet(i ping from specific interface and a dst from another subnet)
07:53 < detha> senaps: tcpdump, and see what is happening. maybe redirects, maybe asymmetrical routing, who knows
07:58 < majculluh> i.e. "no dude i dont think that"
07:59 < adleff> majculluh, hey
07:59 < adleff> did you drink too much asshole juice
07:59 < adleff> always be sure to follow the directions on the label
08:00 < senaps> detha i did, packets come in, they get lost in the os! and arent delivered to application layer. i watched the packets come through the interface.
08:02 < detha> senaps: if you see the packets on the interface, but not further up the stack, the next most likely culprit is the firewall (and specifically established/related rules with asymmetric routing)
08:03 < bane5000> anyone here have any experience setting using pfsense's openvpn client to route traffic from one particular interface through a vpn?
08:13 < spaces> morning chicks!
08:13 < spaces> linux_probe are you late or early ?
08:17 < linux_probe> both spaces
08:18 < linux_probe> at one point I rolled out the door by 10 after 5am
08:19 < linux_probe> and was wprking like a slave by 6:30 latest  long drive!
08:19 < linux_probe> worked 14 to 17 hours and drove home, then went out partying
08:20 < linux_probe> hard on a man roling in at 3am and waking up for work at 5am again ;)
08:20 < spaces> hehe
08:20 < spaces> so you are screwed up ?
08:21 < linux_probe> that was 20 years ;) https://www.youtube.com/watch?v=ZeAM1vwEcFg
08:22 < linux_probe> Mel Blanc, The Man of 1000 Voices [1981] - AMAZING TALENT !!
08:23 < linux_probe> real actor there ;)
08:23 < linux_probe> not he darwin idiots that are flaunted over today
08:24 < spaces> linux_probe you need sleep it seems
08:26 < linux_probe> so does ur mum @ spaces
08:27 < spaces> linux_probe no she did some others after you passed away after 2 min
08:28 < spaces> and she needs her sleep, keeps her sexy
08:31 < senaps> if any of you guys able to help me with this problem in centos/networking/routing! https://pastebin.com/MhPcnzZE
08:41 < spaces> linux_probe but are we sexy ?
08:46 < detha> senaps: haven't looked at it in great detail, but your problem appears to be the "some of these subnets have routes to each other(connected via a router)". Asymmetric routing, and things arriving at interfaces the machine doesn't expect it on
10:01 < huwjr> hey, I have some servers co-located to host a few websites.. they are reasonably powerful servers and have ssds with gigabit NICs. my switch is gigabit but my uplink is only 100Mbps. the DC have some graphs showing that I use hardly any of the available bandwidth, but I was wondering whether having such a small pipe might impact page load times? I’m finding that ‘time to first byte’ is always slow(er) on my kit.
10:05 < b0bbytables> huwjr: have you run anything on the server(s) to check it's render time? and a traceroute both ways between your desktop and the server may give some hints if there are any other points of lag
10:29 <+catphish> morning!
10:31 < Apachez> huwjr: not reqally
10:31 < huwjr> sorry guys. sidetracked as usual!
10:31 < Apachez> Im running a site on a p3 733 mhz thats like 15 year old installation
10:31 < Apachez> check coding of your site
10:31 < Apachez> and use good settings
10:31 < huwjr> nope, I suppose I haven’t /strictly or completely/ ruled out previous servers being faster :) but on some occassions it’s been a huge upgrade in specification
10:31 < Apachez> deflating is good to lower bandwidth impact
11:01 < Nothing4You> is there an "easy" way to use nfs4 with custom user mapping? i have user foo@host1 which i'd like to map to bar@host2, different uids
11:02 < Nothing4You> on linux
11:03 < Atro> wrong channel
11:04 < Atro> try the linux channel
11:33 < majcull56> why did you move?
11:34 < majcull56> some networks need a reboot out of the vm
11:44 < majcull56> we need to be getting Sphinx available
12:07 < djph> o/
12:16 < tya99> would an unmanaged switch strip VLAN tags?
12:16 < djph> it might
12:17 < Atro> it does
12:17 < tya99> right.
12:17 < djph> *might just simply drop the frames
12:17 < tya99> i thought that might be the case
12:17 < Atro> djph: but im curious as to why, actually
12:17 < Atro> for an unmanaged switch does no packet validation
12:18 < Atro> im thinking it cannot just be about  the length
12:18 < tya99> so i have a bunch of these on my network https://www.cisco.com/c/en/us/support/switches/sg-100d-08-8-port-gigabit-switch/model.html
12:18 < Atro> the switch being a grill where only specific length packets can go through
12:18 < djph> Atro: dot1q frames are (IIRC) 1522 bytes, instead of 1500 (1508?)
12:18 < tya99> i wanted to put a host in that (packets leave host tagged) pass through unmanaged switch into tagged port on the managed switch
12:18 < tya99> and then leave through to the router on port 1 (also tagged)
12:19 < Atro> jesus christ why does that switch exis
12:19 < djph> Atro: too big = it's not gonna get thru
12:19 < Atro> djph: are you implying only managed switches support jumbo ?
12:19 < tya99> lol
12:20 < Atro> and, the switch being stupid, it'll drop the packet instead of fragmenting it, i guess
12:20 < djph> Atro: I'm implying "old enough" switches may only pass 1500 byte frames, and nothing more.  *newer* dumb-as-a-rock switches *may* pass through dot1q frames unmolested.  But then, you have fucking unmanaged switches on your network (ewwww)
12:21 < Atro> djph: yeah i was thinking about those ancient unmanaged switches, close to 100mb ports lol
12:21 < tya99> and wait that's not the one i have
12:22 < tya99> because it said something about 802.1q support
12:22 < regdude> dumb switches should ignore VLAN tags and forward packets only based on MAC addresses, anything past the DST-MAC should be ignored. Of course, expect everything, I wouldn't be surprised if there exists a weird switch
12:22 < Atro> regdude: yes but that implies it actually looks at the packet content and doesn't just drop it off based on length
12:23 < regdude> true, if the hardware MTU on a switch is set to 1500, then FCS is stripped off and a switch can drop the packet since it not a Ethernet packet
12:24 < regdude> but... again, I wouldn't be surprised if such switch exists
12:25 < djph> regdude: "such a switch" ... in terms of?
12:26 < regdude> I imagine if they label that dot1q is supported, then hardware MTU is set to 1504 at least
12:26 < djph> 1508, IIRC
12:26 < regdude> 4 bytes
12:26 < tya99> oh wait
12:27 < djph> the other 4 are for something else that came out around then too
12:27 < Atro> wiki says tagged is 1518
12:27 < tya99> hmm
12:27 < tya99> The Cisco SG 100D-08 delivers Quality of Service (QoS); all received packets are examined for QoS priority encoding. The switch reads the priority level and forwards the packet based on that priority level. For example, during heavy loads voice and video traffic are given priority over data traffic. (The switch is configured to comply with 802.1p, VLAN tagged frames.) This ensures that time-sensitive
12:27 < tya99> traffic gets the highest level of service.
12:27 < Atro> tya99: ye u can do tagged on that shit
12:27 < tya99> so that gave me the impression it would not touch the VLAN tags.
12:27 < djph> Atro: yeah, that's the one ...
12:27 < tya99> if they exist coming in it will leave them there
12:28 < regdude> 1500 is L3 MTU, 14  bytes for mac src+dst, 4 bytes the VLAN tag
12:29 < Atro> tya99: naw
12:30 < SwedeMike> one always has to know how each platform calculates MTU, some do L3 (1500), some count ethernet header minus CRC (14 bytes), then some do or do not count the vlan tag (4 bytes each), and for rate calculations some needs to add the IFG and other on-wire things that happen per-packet.
12:30 < Atro> Yeah, so we can just assume the switch has a fixed MTU slapped on it, and drops intead of fragments
12:34 < tya99> those cisco switches yay they don't strip the packets
12:35 < tya99> i think what happened when i clicked apply in the web ui it didn't get committed because i cut myself off :)
12:35 < SwedeMike> Atro: yes, if the MRU (maximum receive unit) is lower than the packet size, then the packet will be silently dropped.
12:46 <+catphish> MRU has always seemed a bit pointless to me, if you received a packet why throw it away
12:47 <+catphish> only throw it away when it won't fit in a buffer somewhere
13:07 < screwsss> whaddup gang
14:38 < laddite> I have a issue where 2 hosts share the same failover IP,  the problem is that when the request in made from within any of the two hosts to itself (to the DNS with A record pointing to failover ip), it will immediately think it is the IP and the traffic will "resolve" to the current host, which breaks the failover
14:39 < laddite> is there any way to prevent the host from considering the IP connected to the interface
14:45 < mawk> maybe don't assign the ip to the host laddite
14:45 < mawk> else it is possible to not consider the local ip, but it's pretty tricky
14:45 < mawk> you need to prevent the loop that will inevitabily happen
14:46 < mawk> so you need to differentiate packets coming from loopback and packets coming from ethX
14:46 < mawk> but it can be done
14:46 < mawk> albeit extremely dirty
14:46 < mawk> so you take a look at the list of policy routing rules with ip rule, you see "0:      from all lookup local" at the top
14:46 < mawk> that means the local addresses have the highest priority for routing
14:47 < laddite> mawk: so, probably a better approach would be to use a script to add/remove the entry from interface based on the current state?
14:47 < mawk> you'd need to move that rule and add a new rule before
14:47 < mawk> that doesn't seem too right either
14:47 < mawk> who's doing the failover ?
14:47 < mawk> a box before the two machines ?
14:47 < light> keepalived
14:47 < TotallyNotKim> or any sort of virtual ip manager
14:48 < laddite> I was thinking of keepalived, currently it's all manual :/
14:49 < laddite> but it seems that having it attached for the both them is the wrong approach, right?
14:50 < mawk> sometimes some derivative of that approach is unavoidable, but here it can be avoided for sure
14:57 < laddite> thanks, mawk
14:58 < screwsss> does this net limiter shit actually work?
14:58 < heller_> hey guys
14:59 < djph> screwsss: what net limiter shit?
15:00 < heller_> What does it mean if i get alot of STP port X stp port state is set to disabled or forwarding?
15:00 < heller_> someone is creating a loop?
15:00 < djph> probably
15:00 < djph> and/or the switches are fighting with each other
15:00 < heller_> There should be no other switches
15:00 < heller_> only routers
15:01 < screwsss> https://www.netlimiter.com/
15:01 < djph> spanning tree is a switch thing ...
15:01 < heller_> its a main switch, where clients get one port to their area
15:02 < heller_> they basically cannot do a loop
15:02 < heller_> they only have acces to one port
15:02 < regdude> a bridge has STP, initially only bridges had STP, then switches started to have STP as well
15:02 < regdude> what kind of routers and switch are you using?
15:03 < heller_> I dont know a client is connecting to the swith
15:03 < heller_> but the swith is a planet branded
15:04 < heller_> But it seems to trigger STP immediately, because i'm not getting the MAC address to the mac table
15:04 < regdude> port can go into a disabled state if the physical port is flapping or it received its own BPDU (loop)
15:04 < regdude> BPDU will work a bit faster in most cases than MAC learning
15:04 < Reventlov> Hi
15:05 < heller_> BPDU guardi is disabled
15:06 < Reventlov> Is there a faster way to count packaets in a pcap (and do some filtering) in python than using tshark?
15:06 < Reventlov> like, counting wifi retransmissions
15:06 < screwsss> djph supposed to be a program whereby you can throttle your upload / download speeds
15:06 < Reventlov> (iterating with pyshark over the packets takes something like 180 seconds, while just using a read filter using tshark is a matter of seconds)
15:06 < screwsss> handy if you're on an adsl connection and your uploads interrupt your download speeds
15:07 < screwsss> in my case, it almost totally chocked it. so therefore i cannot upload a large file to youtube
15:08 < djph> ouch, upload crushing down ...
15:08 < regdude> heller_: BPDU guard is a feature on top of STP, BPDU are the packets that STP uses to determine the netowork's topology and to detect loops
15:08 < screwsss> yep. exactly.
15:08 < djph> sounds like a local-pc thing to do QoS maybe?
15:10 < heller_> ah okay
16:14 < acpowerr> Hello all. My isp is set up in such a way that multiple people can be behind the same IP at the same time. This prevents the outside world to directly connect to a home pc. Is there any way to enable access to my pc from the outside?
16:14 < Atro> acpowerr: you must have a static WAN ip or a dyndns name
16:14 < acpowerr> I do have ddns
16:15 < acpowerr> It just doesn't work
16:15 < acpowerr> Because it points to the external IP
16:15 < Atro> okay, then you need to port-forward
16:15 < acpowerr> I do not have access to the external IP
16:15 < Atro> but be advised, without security, you're at risk
16:15 < acpowerr> It's controlled by the isp
16:15 < acpowerr> And it only forwards packets
16:15 < acpowerr> To my pc
16:15 < Atro> Okay, buy how do you connect on your router?
16:15 < Atro> *but
16:16 < acpowerr> Router - >nanostation
16:16 < acpowerr> Then another nanostation
16:16 < acpowerr> And then I don't really know
16:16 < Atro> i assume the nanostations are the ISP's ?
16:16 < acpowerr> Yeah, thought mine is on the roof
16:16 < acpowerr> It connects to another one a few miles away
16:17 < Atro> if you only control the router, you need to find out what WAN ip it has
16:17 < Atro> if it's an internal or external one
16:24 < acpowerr> I control the router only
16:24 < acpowerr> But the Wan IP is internal
16:24 < acpowerr> As in something like a nat
16:25 < acpowerr> So I am behind probably a quadruple nat
16:25 < acpowerr> Or something of the sort
16:27 < acpowerr> OK so it is like this router - nanostation-another nanostation miles away which connects to the router that is visible to the world
16:27 < acpowerr> So i am behind three devices
16:28 < acpowerr> Two of which I do not control
16:28 < acpowerr> Well, three I suppose
16:28 < acpowerr> So ddns doesn't appear to work for me
16:29 < acpowerr> And even if it points to the external IP
16:29 < acpowerr> That router cannot forward the packet to me
16:31 < djph> gotta config that router.  if you can't, then you're SOL
16:34 < lasdam> hey, I have a medium long network programming question about epoll and stuff. it might be more than 1 paragraphs long. do I post it here, pastebin it, or post elsewhere like ##programming ?
16:36 < acpowerr> djph that is bad
16:37 < djph> acpowerr: what is?
16:46 < simbalion> Is there a "easy" way to determine what local processes are initiating a TCP connection on 127.0.0.1:25?
16:56 <+catphish> simbalion: netstat -anp
16:56 <+catphish> or netstat -anpl to include listening sockets
16:56 <+catphish> if you run those as root you will see which process is using each socket
16:57 < Roq> lsof -i  also works
16:58 < spaces> all networks sexy and rocking ?
16:59 < fantus> what would be the reverse mx record/ ptr be used for ? is it important ?
17:00 < spaces> fantus sure it is, if the IP/domain you are sending from is a valid combination
17:00 < spaces> server also
17:00 < spaces> hostname
17:03 < fantus> spaces: i get this : ERROR: No reverse DNS (PTR) entries. no reverse (PTR) detected on my domain...the domain in question has google mx records setup and its working but https://intodns.com is giving me the result like an error
17:03 < spaces> fantus dns cache, try mxtoolbox.com
17:04 < fantus> spaces: https://mxtoolbox.com/ReverseLookup.aspx do i enter my cPanel/website IP here?
17:04 < fantus> it routes back to my NameServers if its website ip
17:05 < fantus> all is green
17:05 < fantus> seems good
17:05 < fantus> intodns.com still gives me the error on PTR tho
17:06 < spaces> then your DNS servers are not working properly
17:06 < spaces> do you have a reverse DNS zone ?
17:06 < spaces> ZONE!
17:07 < fantus> i guess i'll have to ask my service provider
17:07 < fantus> unless its smth i can do in WHM spaces ?
17:09 < spaces> you need to indeed
17:09 < fantus> and last question, how can this impact my email system? will i still be able to send/receive emails ? spaces
17:11 < spaces> yes but they probably will be blocked tagged as spam
17:11 < fantus> ok tks
17:12 < spaces> np
17:19 < simbalion> catphish: Thanks!
17:25 < lasdam> does it make sense to combine epoll and select calls in an application, epoll for "outer" calls, select for "inner" ones?
17:31 < ash_worksi> when I dig ANY my server I get 'HINFO ... ANY obsoleted'; if that's true, shouldn't that also be the case if I dig google.com ANY ?
17:33 < ||cw> ash_worksi: no?  servers can choose to support obsoleted command if they want
17:33 < ash_worksi> hmm
17:33 < ash_worksi> thanks
17:34 < tpr> I suppose cloudflare is one of the few who say it's obsoleted.
17:34 < ||cw> ANY works against local 2012 level AD too
17:35 < ||cw> and a somewhat recent bind9
17:38 < Apachez> most want to block ANY due to possible size of reply
17:40 < ||cw> I can see how that would add up.  HE doens't block it either though
17:40 < Apachez> most arent
17:41 < Apachez> its similar to blocking ping echo requests (and replies)
17:41 < Apachez> the proper way is to throttle like max 10/s/ip or so
17:41 < Apachez> but with more being behind CGNAT that will fuck things up too
17:41 < Apachez> and ANY isnt really used by any real world application
17:41 < ash_worksi> why would my server not get an answer from dig ANY while my localhost can?
17:41 < Apachez> I mean most knows if its A/AAAA or PTR they ask for
17:41 < Apachez> or MX
17:41 < Apachez> or some other
17:42 < Apachez> ANY is handy for troubleshooting only
17:42 < ash_worksi> specifically ANY
17:42 < Apachez> your server = ?
17:42 < Apachez> is it isp maintained server?
17:42 < Apachez> or one you are admin for?
17:42 < Apachez> the localhost dns client cache is dnsmasq or something
17:42 < Apachez> I doubt that will filter anything except for srcip asking
17:43 < ash_worksi> it's in a data center; I have permission on it
17:44 < ash_worksi> I find it weird that I get an answer for A but not ANY
17:44 < ash_worksi> even if it's an HINFO obsoleted answer
17:44 < epaphus> Hello all. If I set 3 Wifi APs with same SSID, different channel (1,6,11) but instead each gives out its own IPs in different subnet (192.168.1.0/24, 192.168.2.0/24, 192.168.3.0/24) would I still be able to achieve roaming? I understand that the practice is to eave a single AP do the dhcp.
17:44 < Apachez> if its your server then check its config
17:45 < TotallyNotKim> Apachez: I think he _wants_ the different subnets
17:46 < ash_worksi> Apachez: what am I looking for exactly?
17:47 < TotallyNotKim> epaphus: I'd say yes, if whatever is behind the ap's is able to route the packets to the appropriate destinations
17:47 < TotallyNotKim> why-ever one would want this
17:47 < epaphus> TotallyNotKim, why is it that the recommendation is to have a single dhcp server? what difference does it make?
17:48 < TotallyNotKim> epaphus: it's not about what's done usually or what's recommended but what you need and how you get there
17:52 < ash_worksi> Apachez: I mean, I guess the server I'm digging from is blocking ANY like you said, but I don't know where/how to configure that... can you give me a push in the right direction?
17:56 < DocScrutinizer05> off topic: please sign https://www.change.org/p/european-parliament-stop-the-censorship-machinery-save-the-internet?utm_source=share_petition&utm_medium=copylink&utm_campaign=share_petition - read https://blog.github.com/2018-03-14-eu-proposal-upload-filters-code/ for more context.
17:59 < epaphus> TotallyNotKim, ok thanks...
18:02 < epaphus> Btw, I have an issue with a AP that I put as a test. When there is only 3 users on it,  get good performance. When the rest of the office users hop on it (12) latency starts increasing a lot, some packet loss occurs. These are all cellphone users and their consumption is 1-2 mbits out of 25mbits of my WAN. I suspect that one wifi user is mafunctioning and putting noise into the network or something like that. How can I be sure?
18:03 < compdoc> turn off each users pc one at a time?
18:04 < compdoc> maybe the AP doesnt have the horsepower needed?
18:04 < djph> ^ either of those
18:18 < TandyUK> hmm.. 14x 4tb sas disks, on a hp p420.  I can decide what raid layout to go for..  14x4 r5, 2 sets of 7x4 in raid 50, 14x4 raid 6, ...
18:19 < TandyUK> this is going to literally expose the resulting array to the os, for it to be lvm'd and have lv's exported via iscsi
18:19 < UncleDrax> HBA mode, ZFS !
18:19 < UncleDrax> ;]
18:19 < TandyUK> im leaning towards raid50 for speed/safety balance
18:19 < TandyUK> my homelab, so i dont have room to back it up :P
18:20 < UncleDrax> sounds reasonable tome.. but i'm not a storage guru (and my arrays usually stop around 8 disks per.. so way easier MTBF gap to combat)
18:21 < heller_> hey guys
18:21 < UncleDrax> .. and smaller disks
18:21 < UncleDrax> the larged prod disks I have are still 1TB
18:21 < heller_> can you explain why edgeswitch has option to set ports as trunk OR tagged
18:21 < heller_> but tagged works also as trunk?
18:21 < TandyUK> yeah thats why raid50 appeals too, raid 60 id just be losing too much capacity
18:22 < TandyUK> yeah but at home ive got lots of meadia to store
18:22 < UncleDrax> and people won't not get paid if the whole thing dies
18:22 < TandyUK> you may have just convinced me to do 2 arrays though... 8x10tb or whatever the biggest i can get it, in raid 6, and 6x 1tb 15k disks for speed
18:23 < TandyUK> well not entirely lol, its also hte iscsi store for my homelab esxi cluster
18:23 < TandyUK> hence i do want _some_ speed
18:23 < UncleDrax> ah ok, so you do need performance.. ya
18:23 < TandyUK> i think 2 arrays is the way to go
18:23 < TandyUK> one for bulk, one for speed
18:23 < UncleDrax> heller_: sorry, not a EdgeSwitch user
18:24 < UncleDrax> yeap, you could config your 'speed' as 10 or whatever if you thought that little extra on write would matter
18:26 < UncleDrax> huh that P420 is suprisingly inexpensive for a 'real' controller
18:26 < gef3233> Hello! I was doing some tests and I found a sawtooth pattern on the ping RTT, apparently is a known problem when the network is saturated but I couldn't find the reason (is it buffers?, polling?) can anyone help me?
18:26 < gef3233> This is an image of what I mean: https://www.pingplotter.com/themes/pingman/images/manual/interpret_bandwidth_sat_1.png
18:30 < UncleDrax> I'm not a network queuing expert, but could be a queuing issue. You have the ability to change your queuing method?
18:30 < UncleDrax> also see: Bufferbloat (maybe)
18:31 < UncleDrax> but I guess the P420 becomes 'real' once you slap on that flash cache module
18:32 < UncleDrax> ok that's enough HPE talk for me today.. any more and I'll have to update mine
18:33 < gef3233> UncleDrax, but we have tried with a slower datarate and according to our findings it only depends on time and not on the throughput
18:34 < gef3233> We have ruled out application buffers at least
18:34 < gef3233> Is by no means an issue to us, but I am incredibly curious as to why this happens
18:35 < UncleDrax> time as in 'time of day' or 'duration of test'
18:36 < gef3233> As duration of test and as packets/unit of time
18:36 < UncleDrax> well keep in mind if it is a buffer issue, it could be with any device along that path.
18:36 < gef3233> If it were a queing problem we would have seen it go worse with a higher datarate
18:37 < UncleDrax> rgr
18:37 < gef3233> It only happens on the wireless part of the connection, pinging within the lan gives us stable ping rtt
18:38 < UncleDrax> ah, well you said the magic word
18:39 < UncleDrax> (which is also a queue for me to excuse myself and take care of some other stuff.. I'm trying very hard to purposely not delve into Wireless troubleshooting ;] )
18:39 < gef3233> Sure :) The only thing that worries me is the cyclic nature of this behaviour
18:41 < Atro> someone help me, im stupid with IPv6
18:41 < UncleDrax> ya, well clearly at least you isolated it to a single segment and pair of devices, so that helps the t-shoot
18:43 < gef3233> It helps indeed, hopefully I will be able to understand it better :)
18:44 < UncleDrax> Atro: was there a question, or just a statement?
18:46 < Atro> UncleDrax:  yeah but i ran outta internet
18:47 < UncleDrax> Atro: fair enough. the Internet is a fad anyway
18:47 < Atro> If i get an ipv6 /48 from an ISP, i can toss it directly to my clients(I have a home router) ?
18:47 < tds> well if that's legacy internet you've run out of, v6 is much bigger ;)
18:47 < Atro> If so,how does the dns work?
18:48 < Atro> I got via dhcp client over pppoe, a /48 that i can relay(?) To clients?
18:48 < tds> you'd typically have a /64 out of that /48 on-link per VLAN
18:48 < tds> so if that's a "normal" home network, just a single /64 on the lan network
18:49 < tds> then you can enable sending router advertisments on your router, and if you set the autoconfiguration flag then devices should generate themselves addresses within that /64
18:51 < Atro> Oh.
18:52 < Atro> And the flag should be like "take dis ip and dns and im defaults gw"
18:52 < tds> that's what the RA contains, the flag is just something you can add to the RA to tell devices to automatically generate addresses
18:52 < Atro> I probably messed up the lan side
18:52 < tds> (otherwise they'll point a default route at the router, but not generate an address)
18:53 < tds> what router is this?
18:53 < Atro> A mikrotik
18:53 < tds> ah, I'm not familiar with their stuff, sorry
18:53 < Atro> It's alright, i only spent 2 hours on the wan stuff
18:53 < tds> i'm guessing you have some kind of dhcpv6 client doing prefix delegation to get that /48 routed to you?
18:54 < Atro> I believe so, i can put a pool on what i get from the isp
19:20 < luke-jr> Is there a modern bus-topology Ethernet?
19:21 < BullHorn> i set up an OpenVPN server on Windows 10 and im able to connect to it from a client - however the traffic doesnt actually go through the server
19:21 < BullHorn> then i added the following line to the server configuration to enforce ALL traffic from clients goes through the server: push "redirect-gateway def1"
19:22 < BullHorn> what happens now is that clients connect to the server but are unable to send or receive any traffic whatsoever... what am i missing ._.
19:25 < ALowther_> Am I understanding correctly that passive PoE Out will just blindly send power through as soon as anything is plugged in, regardless of it's compatibility. Same for passive PoE In, which will just start receiving power regardless if it's too much or too little?...Whereas active PoE Out/In will only start sending/drawing power, respectively, one it has done a "hand-shake" to verify compatibility?
19:26 < turtle> yep
19:26 < turtle> you can always test passive poe by putting the cable in your mouth
19:26 < ALowther_> turtle: ;).
19:27 < ALowther_> turtle: So it's sending power even if nothing is plugged in?
19:32 < detha> ALowther_: 'passive' PoE is just 24V or 48V between two pairs.
19:34 < ALowther_> detha: & what is active?
19:35 < UncleDrax> turtle: I like to do that with the high-power stuff we run up cell towers.. good times
19:36 < detha> a small probe voltage that checks for a 250 ohm resistor, if it sees one applies a little bit of power, and starts negotiating about how much to apply for full operation
19:37 < detha> UncleDrax: if it doesn't light up a fluorescent tube, it's not high power ;)
20:00 < sovereignentity> I have an Arris SB8200 Modem. when I plug a laptop into the Modem I can connect to the internet. But if I disconnect the laptop and connect the roku I have no connection. what could be the reason?
20:01 < Mr_Midnight> sovereignentity: Are you getting an external IP from the modem?  It could be your provider only allows one mac address to pull a public IP
20:02 < obcecado> sovereignentity: it might need a reset
20:02 < S_SubZero> they may have differing IP setups
20:02 < sovereignentity> the roku is the only thing connected
20:03 < sovereignentity> I have not reset the modem
20:03 < Mr_Midnight> but you connected the laptop at some point and it gave the laptop an IP... could be it doesn't hand out a second until the other IP has expired which could be up to a week sometimes in ISP DHCP pools
20:03 < Apachez> detha: its not the amount of V that kills you, its the amount of A
20:04 < sovereignentity> Mr_Midnight, and everyone thanks I'll reset the modem
20:15 < Johnjay> 400 Mhz is a really weak frequency
20:15 < Johnjay> it barely seem to penetrate a wall
20:16 < fattredd> Hey guys, I'm ubable to consistantly connect to my web server. I think my ISP is blocking protocols. Does this make sense? I'm able to connect via OpenVPN, but not via ssh, http, https, etc. consistantly. Every once in a while it will work again. Any ideas?
20:17 <+pppingme> fattredd where are you tryinig to connect from?
20:17 <+pppingme> Johnjay I wouldn't say thats true
20:17 < fattredd> From my work to my home
20:18 <+pppingme> fattredd you sure your ip isn't just changing?
20:18 < Johnjay> pppingme: i'm using a key finder from amazon that uses 2 AAA batteries
20:18 < Johnjay> it can't work through 2 walls
20:18 < fattredd> Yes. I've got a dynamic dns, and on top of that I've double checked the current IP through teamviewer
20:18 <+pppingme> Johnjay its probably not because its 400mhz, but just an incredibly weak signal to start with, you'd have the same issue at any frequency with such a weak signal
20:18 < Johnjay> fattredd: did you double check all router and firewall settings? try multiple devices with the server?
20:19 < Johnjay> pppingme: i was a bit disappointed because i thought this device could double as a car finder if i left the unit on my car
20:19 < Johnjay> but unfortunately it's too weak for even that
20:19 <+pppingme> fattredd isp's don't tend to randomly block, its going to be way more consistent
20:20 < fattredd> Johnjay: I've done both. The wierd part is the inconsistancy. Sometimes I can get the http fine, other times I can't
20:20 < fattredd> I can always get openvpn though
20:20 < Johnjay> sounds like your router is dying
20:20 < fattredd> Oh shit no
20:20 <+pppingme> fattredd whats your firewall/router/nat device?
20:20 < fattredd> edgeRouter X
20:20 < jason85> How does sslstrip work under the hood? Does it simple strip https from links or rewrite 302 redirects?
20:21 <+pppingme> fattredd whens last time it was rebooted?
20:22 < fattredd> About 5 days ago
20:23 < mnemon> jason85: yes, just http link and redirect rewriting.
20:24 < fattredd> Is there a way to check the edgerouter's logs?
20:24 < jason85> mnemon, Ok, thanks for confirming
20:29 < mnemon> np
20:31 < fattredd> Found the logs. Looks like I'm not getting the http request at all?
20:31 < Johnjay> fattredd: is your internet all line or is there a wireless part?
20:32 < Johnjay> i know when i reboot my router and antenna i get less packet loss from my company
20:32 < fattredd> I've got both. I'll try the reboot
20:33 < fattredd> Thanks Johnjay
20:44 < Apachez> interresting evidence https://mitti.se/nyheter/biljettkontroll-vagen-domstol/?omrade=hela-stockholm  use google translate if you dont know swedish
20:45 < tds> Atro: I only just thought to ask, did you get v6 working in the end?
20:47 < Apachez> Johnjay: the lower the freq the more likely to penetrate a wall
20:47 < Apachez> if it doesnt in your case then its most likely due to too small tx power
20:47 < Johnjay> no the opposite
20:47 < Johnjay> higher freq means high energy
20:47 < Apachez> and too bad antenna on receiving side, use directional antennas instead of omnidirectional
20:47 < Johnjay> which would penetrate more right?
20:47 < Apachez> not really
20:48 < Apachez> for a given amount of tx power, lets say 1 watt, a lower freq will penetrate deeper into/through a wall than a higher freq
20:48 < Apachez> for a higher freq to penetrate as deep as a lower freq the higher freq must use far more tx power
20:49 < Apachez> thats why 10 meters of water is enough to shield against x and gammarays from a nuclear reactor
20:49 < Apachez> while it wont stop a submarine from communicating with its surroundings while being submerged
20:49 < kamura> also lower hz can diffract around obsticales much easier
20:50 < Johnjay> interesting
20:50 < Johnjay> ok i was wrong then
20:53 < ALowther_> detha: Okay, sorry. I was away for a bit. Just for some clarification. passive PoE is either sending 24V or 48V and hoping for the best, whereas active PoE can negotiate a range of applied voltages?
20:54 < Atro> tds:im drunk at ripe
20:54 < tds> ah, have fun :)
20:55 < UncleDrax> sometimes 'hoping' for the best with passive PoE  means 'frying the device on the other end'.
20:56 < ALowther_> UncleDrax: Exactly my thoughts, which is what I am trying to understand.
20:56 < UncleDrax> which is really just a warning to make sure you match the kit appropriately
20:57 < ALowther_> UncleDrax: But if you know your device requirements does active PoE offer more options are far as voltage? Or it is just a safety mechanism to make sure a device is compliant with the standard and if not it doesn't work.
20:59 < UncleDrax> well afaik, 'active' means 'a standard that was designed and published' so it complies to 802.3af or 802.3at - which also happen to have handshaking and stuff in addition to voltage/wattage requirements.
21:00 < UncleDrax> so you can go read those documents.. I am not a PoE expert
21:00 < UncleDrax> (todays trend, for those following along, is I am not an expert in anything)
21:01 < UncleDrax> tbh as an operator, so long as the devices at each end agree, I don't care. I prefer activePoE since it handshakes and I can't accidently set 56vDC on a 24vDC radio
21:01 < tds> since passive poe also just uses one of the pairs for power, you'll normally be limited to 10/100 with most equipment
21:02 < tds> I think there are some passive poe implementations that can do gigabit, they're less common though
21:03 < UncleDrax> for standard PoE, there's a chart: https://en.wikipedia.org/wiki/Power_over_Ethernet#Standard_implementation
21:18 < ALowther_> tds, UncleDrax: Thanks for info & links
21:36 < Capprentice> Hi! Im trying to advertise the default route from Core Switch to Server Switch using default-infomation originate always in OSPF. All interfaces as shown in the picture - https://imgur.com/a/XHuM4fN are in area 0. When I look for defult route I only see one in AGGREGATION_SWITCH. Why the server switch does not get the default route?
21:43 < rcr_> does anyone know why the table of contents line for rfc2459 looks the way it does? https://www.ietf.org/rfc/rfc2459.txt
21:43 < rcr_> there are ^H (backspace) characters in between the extra chars too
21:43 < rcr_> strange
21:44 < UncleDrax> formatting or charset issue. *shrug*
21:49 <+pppingme> Capprentice does "server switch" already have a default route that you're expecting the learned one to over-ride?
23:02 < BenderRodriguez> pppingme: teach me VXLAN
23:04 <+pppingme> did you google it?
23:29 < LonGrangeAntenna> i have question
23:29 < fattredd> Okay
23:29 < tds> assuming it's related to networking, just ask :)
23:29 < LonGrangeAntenna> so my laptop has antenna for wifi, im trying to plug the wifi to my dekstop via ethernet (this works) but the question is
23:30 < LonGrangeAntenna> will it give me latency problems
23:30 < LonGrangeAntenna> how siginificant
23:30 < LonGrangeAntenna> there are reasons why im not just plugging the antenna adaptor into the desktop
23:30 < fattredd> Should be roughly the same as the wifi was
23:30 < LonGrangeAntenna> ok that all
23:31 < Harlock> wat?
23:32 < Harlock> you mean you are bridging the wifi to eithernet on the laptop right
23:32 < tds> are you actually bridging, or doing routing + nat?
23:32 < LonGrangeAntenna> wifi signal goes to antenna that is plugged into laptop
23:33 < LonGrangeAntenna> pipe the connection from laptop through ethernet cable into my desktop
23:33 < LonGrangeAntenna> im using the basic windows10 connection sharing feature on laptop
23:33 < Harlock> is your pipe abs or pvc
23:33 < LonGrangeAntenna> just wasnt sure if it would give known issues or delays that i didnt know about
23:33 < Harlock> oh it's nat then
23:34 < LonGrangeAntenna> yea by pipe i just meant ethernet cable
23:34 < LonGrangeAntenna> from my understanding piping means to connect to system in some manner
23:34 < Harlock> i meant inside the laptop
23:34 < tds> you might have more issues with nat traversal (and other related nat mess) if you're doing double nat, otherwise you're probably fine
23:35 < Harlock> the most latent part of the path is the wifi anyway
23:35 < LonGrangeAntenna> yeah
23:35 < Harlock> but you won't be able to access the desktop from the rest of the nework without port forwarding if that matters
23:36 < Harlock> you can "bridge" 2 interfaces instead
23:36 < tds> if you're able to set static routes on the first router, that's a much nicer solution
--- Log closed Tue Jun 19 00:00:04 2018