I have noticed a glitch. On a local LAN Flex 6500 running V2. Sometimes when shutting down the software and restarting it there is no activity on the spectrum or waterfall but the S meter is showing signal and the audio is coming out of the speakers. I have seen this a few times so far today and will continue to see if it has a common cause to report.
Posted 3 years ago
Tim - W4TME, Customer Experience Manager
The Smartlink Network Test shows GREEN. I just went to McDonald's and my laptop found my 6500 easy, connected easy. BUT, no spectrum or waterfall. Audio good, freq display ok, changes bands, ATU works. Came home. Connected my home PC to the 6500, full display and waterfall. Something not quite right.
Tim - W4TME, Customer Experience Manager
If the free wifi provider is blocking the streaming of large packets, this will result in a lack of waterfall data because the large VITA-49 packets are being blocked.
It works just fine at Burger King, unless you order the fish sandwich, then the screen goes dark...
I experience the same at my QTH connected to a nearby Xfinity hotspot. I believe that data is being blocked because I see the same on some streaming activities like Netflix. I am going to the clubhouse next to try it on our internet there.
I wonder if you could try turning waterfall completely off, dragging the divider to the bottom to make it disappear, set the rate to 0 and display FPS to about 5 and see if the bandwidth drops enough that the McRouter stops blocking the panadapter display.
(Or Tim, does this still get sent via large packets, only fewer of them?)
I can live without waterfall, but it is tough without the panadapter trace!
(Or Tim, does this still get sent via large packets, only fewer of them?)
I can live without waterfall, but it is tough without the panadapter trace!
it's not just at McDonald's the issue is happening Ken. I'm trying to connect to a flex 6300 in Maryland from my QTH in Newfoundland. I can tx, rx, atu works, great audio, but no waterall (which we want) and the pan trace seems to time out right after it starts.
I am on a hardwired fibre connection 320 MB/s down, 115 MB/s up and the site of the equipment in Maryland is also hardwired and gets 50 down, 15 or so up which is plenty, but neither of us are using wifi.
I did packet traces and keep getting malformed packets on the vita 49 protocol, which is from my understanding the protocol the waterfall and pan trace would depend on.
I am very surprised at the battle we're having to get this going. It seems so straight forward. What are we missing?
I am on a hardwired fibre connection 320 MB/s down, 115 MB/s up and the site of the equipment in Maryland is also hardwired and gets 50 down, 15 or so up which is plenty, but neither of us are using wifi.
I did packet traces and keep getting malformed packets on the vita 49 protocol, which is from my understanding the protocol the waterfall and pan trace would depend on.
I am very surprised at the battle we're having to get this going. It seems so straight forward. What are we missing?
Hmmm. (Scratches head...)
Well, that was my one good idea.
You certainly have enough bandwidth up and down at both ends of the connection. For sure, There is a clog in the drain somewhere.....
Serious Networking troubleshooting is out of my league. The guys at FRS will sort it out eventually.
Well, that was my one good idea.
You certainly have enough bandwidth up and down at both ends of the connection. For sure, There is a clog in the drain somewhere.....
Serious Networking troubleshooting is out of my league. The guys at FRS will sort it out eventually.
Eventually I'm sure they will. But for a new potential customer like me it's not a good first impression.
Peter K1PGV, Elmer
@Chris Hillier: That's understandable. Just please keep in mind that real-time streaming across the WAN becomes a *very* difficult problem, once "it works almost all the time for about 90% of the people." There are countless practical issues that need to be dealt with, and those can only be discovered and dealt with in actual use. This is the very first release of this product, and such things are bound to crop up.
Note that streaming with NetFlix (for one example) is a *far* simpler problem than implementing SmartLink. It's unfortunate that you happen to be one of the folks who are experiencing one of the issues.
That there are some issues showing up here and there speaks to the difficulty of the problem, not to any deficiency in the engineering team.
Peter
K1PGV
(a network protocol architect in a former life, and implementer of some non-Flex related pseduo-real-time streaming code)
Note that streaming with NetFlix (for one example) is a *far* simpler problem than implementing SmartLink. It's unfortunate that you happen to be one of the folks who are experiencing one of the issues.
That there are some issues showing up here and there speaks to the difficulty of the problem, not to any deficiency in the engineering team.
Peter
K1PGV
(a network protocol architect in a former life, and implementer of some non-Flex related pseduo-real-time streaming code)
Peter, that sounds like tech-speak for "streaming video game designer" or "video conferencing system specialist."
I understand your frustration, Chris. Even more frustrating is that by all initial measures you are WAY above minimum specs for system needs. So your situation is not typical by any means. It *should* be working with plenty of room to spare.
There is a wrench in the works somewhere in your network, obviously. Purely guessing here, to kill time while the experts look into it......perhaps some obscure networking parameter configured on one end or the other of your network connection (router or firewall settings.). Or perhaps even CAT5/6 patch cable problems. Or something the internet provider on either end of the connection is doing. Or some strange deal in the handoff between one provider and the other as it crosses the border?
There is a wrench in the works somewhere in your network, obviously. Purely guessing here, to kill time while the experts look into it......perhaps some obscure networking parameter configured on one end or the other of your network connection (router or firewall settings.). Or perhaps even CAT5/6 patch cable problems. Or something the internet provider on either end of the connection is doing. Or some strange deal in the handoff between one provider and the other as it crosses the border?
Peter K1PGV, Elmer
>that sounds like tech-speak for
Yeah... no. Nothing nearly so elegant or interesting, I'm afraid.
Several years back, I wrote a program called ScannerCast that streams "live" scanner audio over the Internet from user PCs to the Radio Reference web site. I learned a lot about Internet streaming on that project.. which was definitely only pseudo-real-time as it buffered several seconds of audio before transmission.
Peter
K1PGV
Yeah... no. Nothing nearly so elegant or interesting, I'm afraid.
Several years back, I wrote a program called ScannerCast that streams "live" scanner audio over the Internet from user PCs to the Radio Reference web site. I learned a lot about Internet streaming on that project.. which was definitely only pseudo-real-time as it buffered several seconds of audio before transmission.
Peter
K1PGV
Tim - SmartSDR for Windows.
Tim - W4TME, Customer Experience Manager
Try cold booting the radio (power down, remove power for several seconds, replace power and boot it up)
Tim, cold boot does not seem to fix the issue from occuring again. It seems to happen if I use SmartLink to log into a friends account from my LAN and access one of his Flex radios over the internet and then come back to access my local Flex while still logged in to his SmartLink account. Or if I have logged into the friends SmartLink account, Log out of his SmartLink account and then Log into my SmartLink account and try to access my local Flex.
IF I reboot the local PC or my Flex 6500 and then restart SmartLink it seems to work again.
IF I reboot the local PC or my Flex 6500 and then restart SmartLink it seems to work again.
(Edited)
Tim - W4TME, Customer Experience Manager
SmartLink is not enabled when you are on the local LAN. Are you making a SmartLink connection through the Internet or using the local LAN connection?
OK, I’m going to start this from the start. I have a 6500, hard wired via ethernet cable to a router. My desk PC, Windows 10, i7, is also hard wired via ethernet cable to the same router. I downloaded 2.0 and installed it on my desk PC. Then booted the 6500. Bought the license, installed 2.0 on the 6500, rebooted all. Started SmartSDR on my desk PC and it works fine. Spectrum scope and water fall working fine. I can unlink the meters on the right and move them to another monitor, real neat.
I then installed 2.0 on my Maestro. The Maestro is connected to WiFi that is on the other (outside) of the router. The Maestro could not see the 6500 direct. Download speeds are over 60mps and upload is around 10mps. When I logged in with SMARTLINK on the Maestro, the Maestro saw my 6500 ready. I connected to it. So, my Maestro is going “over the internet” to come back and connect to the 6500. It connects, shows freq, meters, changes freqs, changes bands, ATU works, get audio, I can key the radio, look at menus etc, EXCEPT no spectrum scope, other than about 1 inch. And no waterfall at all. Rebooted all a couple of times. Same story.
So, I linked my Maestro to my Android phone hot spot. Same story, can connect fine, no spectrum or waterfall.
I then installed SmartSDR on my MS Surface Book, Windows 10, i7. Connected to my home wifi the laptop saw the 6500 and I connected. No waterfall or spectrum scope. Again this was thru the internet and SMARTLINK.
Next I went to McDonalds with the MS Surface Book and connected to McDonalds Wifi. Had a good wifi connection. Started SmartSDR than logged in via SMARTLINK and the laptop saw my 6500 at home... YEA. Connected. Everything seemed to work but the spectrum and waterfall.
An observation. The waterfall area has about a 1 inch square in the middle that is totally black/empty. Odd. I have photos of the screen and the network numbers. below
David Decoons wo2x, Elmer
If the waterfall and panadapter work fine when on the local network but do not work via SmartLink then I would check your SmartLink port settings. If you are using automatic in SmartSDR and Maestro then the router must support uPNP.
Connect to the router and ensure uPNP is enabled (what model router?).
If that does not work then you might need to use manual port forwarding..
Dave wo2x
Connect to the router and ensure uPNP is enabled (what model router?).
If that does not work then you might need to use manual port forwarding..
Dave wo2x
Kent, your picture of the screen looks exactly like mine except I am using SmartSDR on a windows computer. I even have the exact same black box in a portion of the waterfall. Dave, in my case I am connected to the Flex 6500 over the local LAN and it does the same as in Kent's pictures.
(Edited)
uPNP is working
Tim - W4TME, Customer Experience Manager
Kent - this is a SmartLink connected Maestro that is showing the problem. it isn't on the local LAN with the radio. The problem is with the Internet provider that is providing the Internet access. I suspect they are blocking large packets as they are interpreting them as streaming data. The panadapter display is contained in these large packets.
Eric - KE5DTO, Official Rep
If I read the description correctly, a SmartLink connection is not going to work properly when the public IP of both the radio and the client are the same (and I think that is what is being described here).
I have Time Warner now Spectrum and do not see waterfall on local PC with my 6600 at Remote site on Charter network. MTU is set to 1500 and not blocking fragmented packets. I can see same on IPhone if on WiFi off the Canle router but looks fine if go to Verizon over the air data and turn off WI-FI.
Call in tomorrow to see if blocking by waterfall traffic. Display is same as others here.
73, Russ K5OA
Call in tomorrow to see if blocking by waterfall traffic. Display is same as others here.
73, Russ K5OA
Tim, no the local Flex 6500 is not showing SmartLink as the connection. However other Flex radios that are Smartlink connections are available and show on the SmartSDR screen. The same problem happen regardless if I am logged into another stations SmartLink account or my own SmartLink account. Because I am working on the computer that is on my local LAN along with my Flex 6500 the local (non-SmartLink) Flex 6500 shows as available. It is when I launch this local Flex 6500 that I get the issue.
If you wish me to demo to you via Teamviewer I can do so. Give me your direct line # and I will call you.
If you wish me to demo to you via Teamviewer I can do so. Give me your direct line # and I will call you.
(Edited)
Tim - W4TME, Customer Experience Manager
Please open a HelpDesk ticket so I can look into this in more detail.
I had. #18926
I will piggy back on Kent's ticket.
WELL.. I have a friend in town with a 6500. He gave me his login and password. I changed my login and password to his and log in. I saw HIS 6500 and connected to it. I could control all his 6500 AND I had his spectrum and water fall... yea... sorta. Then I gave him my login and password, and he connected to my 6500 and had full control and spectrum and waterfall... yea... sorta. He and I have the same high speed internet provider.
So, a couple of things. One. my internet does pass the large packets. Because I could use my wifi and internet to connect and operate the others 6500, and he could connect and operate mine.
Two. But, I cannot get spectrum and water fall with my Maestro or laptop connected to the same wifi and internet provider. How can this be???
So, a couple of things. One. my internet does pass the large packets. Because I could use my wifi and internet to connect and operate the others 6500, and he could connect and operate mine.
Two. But, I cannot get spectrum and water fall with my Maestro or laptop connected to the same wifi and internet provider. How can this be???
Tim - W4TME, Customer Experience Manager
Have you taken your Maestro to a different location and used it to access your radio, not using the same Internet provider?
Yes. I have a ATT Samsung Android phone with Hot spot service. I turned on my phone hot spot service, then connected my Maestro to it and saw my 6500. I connected to it. Same thing, everything works but spectrum scope and waterfall.
I even tried using the phone hot spot and Maestro connecting to my friends 6500 across town. EVERYTHING worked, including the scope and waterfall.
Then I tried my MS Surface Pro book thru the phone hot spot, could not see the 6500 spectrum or waterfall. Connect to my friends 6500 and everything worked.
So, my home wifi was not in use for this. My 6500 is hard wired to the router. And my friend can connect to my 6500 and everything works.
I even tried using the phone hot spot and Maestro connecting to my friends 6500 across town. EVERYTHING worked, including the scope and waterfall.
Then I tried my MS Surface Pro book thru the phone hot spot, could not see the 6500 spectrum or waterfall. Connect to my friends 6500 and everything worked.
So, my home wifi was not in use for this. My 6500 is hard wired to the router. And my friend can connect to my 6500 and everything works.
Tim - W4TME, Customer Experience Manager
Understood. Have you tried the Maestro on an Internet connection that wasn't your hot spot?
Yes, I used MY Maestro connected to my internet to connect to the friends 6500 and it worked. I did try my laptop from Mcdonnalds and it did not work
First post ever: Earlier Tim mentioned "large packet length" for v2.0 VITA-49 packets. The question, I think, is ... do these packets exceed one of the ISP's Maximum Transmission Unit (MTU) limit. Can you tell us what the packet length is, so we can ask a meaningful question of our ISP? Along those lines, Some VTC hardware that uses UDP (and I'm just guessing that waterfall data uses UDP) allows the user to set the MTU. If we could do that, we could duck under an ISP's limits and be good to go. I'd be willing to accept any losses that come with that. I bought a 6300 (primary radio is a 6700) just for remote operation and am now trying to operate the remote backwards (Newfoundland to Maryland) and we are experiencing the "no waterfall" issue. Otherwise it works wonderfully!
Thanks!
Sam
Thanks!
Sam
(Edited)
Tim - W4TME, Customer Experience Manager
... do these packets exceed one of the ISP's Maximum Transmission Unit (MTU) limit.
That is one possibility. The VITA-49 spectrum data (yes, it is UDP) can be larger than the 1500 byte payload of an Ethernet frame, but this is OK because the network stack in the radio is responsible for fragmenting the data into 1500 byte payloads and the network stack on the client end does the re-assembly. This is a tried and true feature of TCP/IP because not all data quanta fit nicely in 1500 byte chunks. So we are sending full sized Ethernet frames (1518 bytes).
I have run into this MTU problem before in a previous life where an MTU on a piece of equipment (I'll refrain from rating out the offending manufacturer) was set for 1500. But a "full" Ethernet frame is actually 1518 bytes (payload + MAC header+ CRC). Most devices interpret a 1500 MTU as 1538 bytes. However, in this situation, this was not the case. Needless to say, when we started streaming full-size data frames from one host to the other, nothing got through. Packets that had data payloads of 1482 bytes made it through.
That is one possibility. The VITA-49 spectrum data (yes, it is UDP) can be larger than the 1500 byte payload of an Ethernet frame, but this is OK because the network stack in the radio is responsible for fragmenting the data into 1500 byte payloads and the network stack on the client end does the re-assembly. This is a tried and true feature of TCP/IP because not all data quanta fit nicely in 1500 byte chunks. So we are sending full sized Ethernet frames (1518 bytes).
I have run into this MTU problem before in a previous life where an MTU on a piece of equipment (I'll refrain from rating out the offending manufacturer) was set for 1500. But a "full" Ethernet frame is actually 1518 bytes (payload + MAC header+ CRC). Most devices interpret a 1500 MTU as 1538 bytes. However, in this situation, this was not the case. Needless to say, when we started streaming full-size data frames from one host to the other, nothing got through. Packets that had data payloads of 1482 bytes made it through.
Thanks Tim,
Assuming that's the issue, If we are very close to a limit (that is imposed by some significant number of networks out there), could the packet size be reduced slightly to accommodate this?
If there are other troubleshooting actions that you think might address this issue (no Panafall at remote user site), I would greatly appreciate a summary of them.
Thanks
Sam
Assuming that's the issue, If we are very close to a limit (that is imposed by some significant number of networks out there), could the packet size be reduced slightly to accommodate this?
If there are other troubleshooting actions that you think might address this issue (no Panafall at remote user site), I would greatly appreciate a summary of them.
Thanks
Sam
Tim - W4TME, Customer Experience Manager
We are investigating options. The engineering team is on a tour of McDonald's and Taco Bells in Austin to use their wifi so we can reproduce the issue and take a network capture to validate our assumptions.
Peter K1PGV, Elmer
That's interesting, Tim.
What's the (max) Interface MTU on the radio? Can we gain some perf in the local LAN case by enabling Jumbograms on our LAN and end system? Increasing our local MTU to, say, 9K?
Peter
K1PGV
What's the (max) Interface MTU on the radio? Can we gain some perf in the local LAN case by enabling Jumbograms on our LAN and end system? Increasing our local MTU to, say, 9K?
Peter
K1PGV
Eric - KE5DTO, Official Rep
Tim wanted us to make sure that everyone knew to try the new shoe-shaped Spicy Double Chalupa. Mmmmmm.
Eric - KE5DTO, Official Rep
On a more serious note, we will have to experiment. The MTU is just one theory at this point. Wifi networks are quite varied from one place to another.
Peter K1PGV, Elmer
I guess the relevant question would be: Does the radio, or somebody, do path MTU discovery to determine the max size packet end-to-end that can be sent? Or does it "dumb down" to 576 as a result of... something?
(To be clear: For me, this is really of academic interest. I don't see myself playing SmartLink anytime soon, due to the lack of decent IP infrastructure where I live.)
Peter
K1PGV
(To be clear: For me, this is really of academic interest. I don't see myself playing SmartLink anytime soon, due to the lack of decent IP infrastructure where I live.)
Peter
K1PGV
Let me add that ... in this case of no panafall, both ends are wired. One with fiber (300 Meg down and a flock of speed uplink) the other is coaxial system testing at 90 down and ~15 up. So it is not a WiFi issue for us.
sl
Tim - W4TME, Customer Experience Manager
Understood. I do not suspect this is a wifi related issue, but a network configuration somewhere in the path between the radio and the client, regardless of how they are connected.
Another note: With both ends turning in very good speed check numbers, we are getting only three yellow bars for network quality and ~ 310 MS latency. When pinging test server latency is far less 30 and 20 MS respectively. Latency numbers are essentially identical from local restaurant (802.11g of continuously varying quality) and from 1500 miles hard wired (Newfoundland).
(Edited)
Tim - W4TME, Customer Experience Manager
Network quality indicator uses the loss of VITA-49 packets the validator of link quality. The client knows what VITA-49 packets to expect and if it starts missing them, that triggers an indication of line quality issues.
Here's a Wireshark capture from the Newfoundland end showing a malformed VITA-49 packet. May be of use in hunting this down. All wired ... no wireless.
(Edited)
Verified this morning (with Chris VO1IDX) that the VITA-49 packets were arriving malformed consistently. Tim has explained (above) that the smartlink network quality indicator reflects dropped V-49 packets. The presence of modest network quality indications when operating on fairly high quality network makes sense then, if something is maligning those specific packets and not others. We do full motion (arms waving) video conferencing over the same connection with no difficulty.
Update: Good packets arriving at VO1IDX this morning (per Wireshark capture) and still no panafall. I agree with IDX, we've expended our network troubleshooting options. Will keep a keen ear to the wire.
I just did packet capturing this morning for an hour Sam. Last night my captures were full of malformed packets but this morning I could barely find one. I thought we might have been onto something, but alas that was foiled. Back to the drawing board as they say. -vo1idx
(Edited)
Tim - W4TME, Customer Experience Manager
Sam - please open a HelPDesk support ticket so we can work on this issue using our standard support processes. This information is helpful and I do not want to allow it to fall through the cracks with all of the other community traffic coming in.
David Decoons wo2x, Elmer
I wonder if one of your routers have Jumbo Frames enabled. One of the other people in this thread checked his router at radio end and found it enabled. He disabled it and will be testing today to see if it helps.
Just trying to think outside the. Oz.
Dave wo2x
Just trying to think outside the. Oz.
Dave wo2x
Peter K1PGV, Elmer
>I wonder if one of your routers have Jumbo Frames enabled
IF the radio generates JumboGrams (we don't know if it does, I actually doubt it) then having a router on the LAN that handles these without fragmentation would only be a good thing, wouldn't it?
Peter
K1PGV
IF the radio generates JumboGrams (we don't know if it does, I actually doubt it) then having a router on the LAN that handles these without fragmentation would only be a good thing, wouldn't it?
Peter
K1PGV
Tim - W4TME, Customer Experience Manager
Jumbo frames are turned off.
I have had "permit jumbo frames? Both on and off on my router and it didn't seem to make a difference at my location. But I have kept it turned off mostly for the past six months.
I tested again a few hours ago and it still doesn't make any difference here that I have noticed. Perhaps the router and modem have different settings?
Anyway, it is back off again for good.
I tested again a few hours ago and it still doesn't make any difference here that I have noticed. Perhaps the router and modem have different settings?
Anyway, it is back off again for good.
Tim - W4TME, Customer Experience Manager
The radio does not use jumbo frames.
I am having the same issue. Loaded V2 on my work computer and connect to my 6700 at home, no problem. All was working well. I closed out SmartSDR to do some other work, came back and reloaded and NO spectrum display or waterfall. Audio was fine. I could transmit no problem.
I notice that DAX is NOT streaming the audio. I have channel 2 selected on the Slice for DAX but nothing streaming. I've logged out of Smart CAT and logged back in and it seems connected, but DAX still shows nothing streaming. Going to to play with it some more and see. Will try resetting the hardware at home when I get back and maybe try it again.
..............Bob W4PG
I notice that DAX is NOT streaming the audio. I have channel 2 selected on the Slice for DAX but nothing streaming. I've logged out of Smart CAT and logged back in and it seems connected, but DAX still shows nothing streaming. Going to to play with it some more and see. Will try resetting the hardware at home when I get back and maybe try it again.
..............Bob W4PG
So I just ran SmartSDR WITHOUT DAX or SmartCAT running first and the display WORKS! Generally both CAT and DAX are running and logged in since those start at bootup. Not sure what this means . . . going to try some more.
Tim - W4TME, Customer Experience Manager
I suspect it is the same issue as the ones reported above. DAX streams large VITA-49 packets for audio.
What does this mean (from my router log):
Jul 28 15:28:59 Whole System ACK Flood Attack from WAN Rule:Default deny<br>
D-Link router 3 or 4 years old.
Jul 28 15:28:59 Whole System ACK Flood Attack from WAN Rule:Default deny<br>
D-Link router 3 or 4 years old.
David Decoons wo2x, Elmer
Ken. I was grasping at straws since some people are having issues with the UDP packets and potentially due to the size of the packets.
Dave wo2x
Dave wo2x
Problem resolved on our circuit ... our fix may help others
In an early post to this thread, Tim mentioned fragmentation ... that was the answer for me and VO1IDX. My cable modem firewall had an option to block fragmented packets. I unchecked it and the panafall works fine in Newfoundland. We re-enabled the blocking feature and the panafall goes away. So, you need to be allowing fragmented packets through any firewall involved.
73
Sam K3KLC
and
Chris VO1IDX
In an early post to this thread, Tim mentioned fragmentation ... that was the answer for me and VO1IDX. My cable modem firewall had an option to block fragmented packets. I unchecked it and the panafall works fine in Newfoundland. We re-enabled the blocking feature and the panafall goes away. So, you need to be allowing fragmented packets through any firewall involved.
73
Sam K3KLC
and
Chris VO1IDX
WOW...Nice catch!
The fun starts now!
Enjoy...
Ken - NM9P
The fun starts now!
Enjoy...
Ken - NM9P
Peter K1PGV, Elmer
Excellent find. Good thinking! BRAVO! Yay! Congratulations!!
This WILL help MANY people.
>My cable modem firewall had an option to block fragmented packets
There are third-party firewall/AV products that also block fragmented packets... so knowing that you must DISable this blocking will be very helpful to folks going forward.
I very pleased to hear you found the solution.
(Having SAID that, it would be nice if the MTU was discovered by SmartLink and this was not necessary... but, maybe that will happen in some future release).
Peter
K1PGV
This WILL help MANY people.
>My cable modem firewall had an option to block fragmented packets
There are third-party firewall/AV products that also block fragmented packets... so knowing that you must DISable this blocking will be very helpful to folks going forward.
I very pleased to hear you found the solution.
(Having SAID that, it would be nice if the MTU was discovered by SmartLink and this was not necessary... but, maybe that will happen in some future release).
Peter
K1PGV
Maybe Tim could comment on the firewall thing as well. I wondering if by chance it's the size of the packets formed, in that the size is 1518 vs the max packet size is at 1500? therefore the packet is fragmanted and the firewall is stopping it...until it is changed.
(Edited)
Peter K1PGV, Elmer
(1518 is the max size with Ethernet header/checksum... 1500 is the max "message" size)
It doesn't matter: If the radio is generating 1500 byte packets... or even 1200 byte packets... it is *extremely* likely that on the path from SSDR to the radio over the Internet there will be SOME intermediate system that can't handle a packet that large.
You can check this out by looking at the WireShark/Ethereal trace and seeing the largest IP packet you receive.
There's really very little risk to passing fragmented IP packets, and allowing fragmented packets to pass is extremely common. Anybody doing gaming will need to allow fragmented IP packets, for example. There *are* some very basic IP fragmentation exploits, but they are pretty mundane and easily countered by any good system or piece of network equipment.
Peter
K1PGV
It doesn't matter: If the radio is generating 1500 byte packets... or even 1200 byte packets... it is *extremely* likely that on the path from SSDR to the radio over the Internet there will be SOME intermediate system that can't handle a packet that large.
You can check this out by looking at the WireShark/Ethereal trace and seeing the largest IP packet you receive.
There's really very little risk to passing fragmented IP packets, and allowing fragmented packets to pass is extremely common. Anybody doing gaming will need to allow fragmented IP packets, for example. There *are* some very basic IP fragmentation exploits, but they are pretty mundane and easily countered by any good system or piece of network equipment.
Peter
K1PGV
I just looked at my router and the maximum MTU I can set is 1500. Me-thinks I need to upgrade as this router is several years old!! :-)
Peter K1PGV, Elmer
1500 is perfectly fine. No need to upgrade because of that.
Peter
K1PGV
Peter
K1PGV
Thanks, Peter. I don't know much about packet size but figured that might be an issue. The D-Link router I have at work does not have an option for accepting fragmented packets. My home router (Arris) does. So I'm guessing there is some setting in the D-Link that does something similar but not sure what that would be. As I mentioned, the D-Link router I'm using is several years old.
Tim - W4TME, Customer Experience Manager
I suspect the "malformed" VITA-49 packet was one of the smaller fragments that got through.
David Decoons wo2x, Elmer
Tim, if jumbo frames were enabled in the router but the radio and PC were not set to use jumbo frames, would this cause issues? Also if the router and PC support jumbo frames but the radio doesn't would that be an issue?
I never have jumbo frames enabled on any of my networking equipment but figures I'd ask so I understand if it would affect other users who may have it enabled in the router..
73
Dave wo2x
I never have jumbo frames enabled on any of my networking equipment but figures I'd ask so I understand if it would affect other users who may have it enabled in the router..
73
Dave wo2x
I worked the issue yesterday and found three units running Smartlink did not display the waterfall but all else worked fine. Expected the ISP blocking some packets, At first light today, all three units are ok and displaying the waterfall without a change in router setting or ISP! SmartLink issue?
Thanks, Russ K5OA
Thanks, Russ K5OA
Tim - W4TME, Customer Experience Manager
Count yourself lucky and be thankful. I suspect your ISP did something overnight or something rebooted and all is well in Beantown again.
