Kitz Forum
Broadband Related => Router Monitoring Software => Topic started by: roseway on October 02, 2012, 10:42:15 AM
-
- corrected SNR tab being visible when SNR per tone is included with bitloading
- corrected auto snapshots of separate SNR per tone graph
- corrected incorrect colouring of separate SNR per tone graph on VDSL2 connections
- corrected failure to save some settings on exit
This is a V1 release candidate.
https://docs.google.com/folder/d/0BxbUtOYVZ_SCZ1BnY3RKYnN2b1U/edit
-
Testing now ....
-
Thank you, Eric.
The 64-bit rs-ux RC1 package has now been downloaded to The Cattery.
-
All seems to be working well using a HG612 modem on my VDSL2 connection.
The snapshot timing does appear to creep 2 minutes sooner every 2 hours though.
-
All seems to be working well using a HG612 modem on my VDSL2 connection.
The snapshot timing does appear to creep 2 minutes sooner every 2 hours though.
Thanks for the report. The 2-minute overlap was deliberate, to ensure that no data was missed. But this was done when I was using a rather inaccurate timer. As I'm now using an accurate timer, I guess the overlap is no longer necessary. I'll remove it.
-
Eric firstly I have just started rs-ux again since reconnecting my D-Link and low and behold the system up time is correct and the AS time now shows to be correct since i have only just done a reconnect (I was testing my HG612 which is another story). So it looks like the AS time does not work well when the router has been up for weeks, or that's my guess. I'll keep an eye on it.
Now to my HG612 testing - all went well using my W7 laptop to unlock and configure it, I then plugged in the phone lead and again it connected to my ADSL service OK however I did not have telnet available as a separate program so I thought I'd check out rs-ux for the stats data and here is my problem. Although I correctly entered the user and password (which works OK from a browser) it kept getting login errors. I did have the sh box ticked as it shows that pop up message about the HG612 needing it. No matter how many times I tried it would not connect and gave telnet errors.
Have you any ideas as to why this might be? Are there any other rs-ux users with an HG612 on ADSL trying this out?
Stuart
Edit: Having just read the help it would seem that the web password and the telnet password are not the same on the HG612 so will try that next time I test this...
-
So it looks like the AS time does not work well when the router has been up for weeks, or that's my guess.
I could be wrong, but I don't think it's to do with router uptime. I think that, with some routers such as the DSL-2740B, the AS value always gives the time since the first connection after a reboot. So if you get a re-sync while the router remains running, the AS reference time doesn't change.
Have you any ideas as to why this might be? Are there any other rs-ux users with an HG612 on ADSL trying this out?
Yes, I use an HG612. You need to change the "Stats command prefix" to "xdslcmd info" for this router, so perhaps you haven't done that.
-
Eric this was at least the second or third re-connect after reboot so system up time is 5+ days and AS time is showing 1 hr 59 mins which is correct frm when I re-plugged the phone cable after testing the HG612.
Stuart
Forgot to add that the popup messages over Uptime, AS Time and Downstream speed dont fit in the boxes correctly. In the case of the two time ones the box is wider than the text but only first half of each line shows up, in the last case the 'd' of speed is cut off slightly.
-
Eric this was at least the second or third re-connect after reboot so system up time is 5+ days and AS time is showing 1 hr 59 mins which is correct frm when I re-plugged the phone cable after testing the HG612.
I see. This may be a bug in the router firmware then. I don't try to do anything clever here - I just read read the AS value reported by the router, and convert it to days, hours, minutes and seconds.
Forgot to add that the popup messages over Uptime, AS Time and Downstream speed dont fit in the boxes correctly. In the case of the two time ones the box is wider than the text but only first half of each line shows up, in the last case the 'd' of speed is cut off slightly.
That's an odd one, and I don't see it here. Do you have unusual default fonts on your system, or anything like that? I'll have a look and see if I can do anything about this.
-
I see. This may be a bug in the router firmware then. I don't try to do anything clever here - I just read read the AS value reported by the router, and convert it to days, hours, minutes and seconds.
I'll see what happens over the next few days.
That's an odd one, and I don't see it here. Do you have unusual default fonts on your system, or anything like that? I'll have a look and see if I can do anything about this.
AFAIK yes I do have the standard Fedora ones (this was noticed on Linux) plus a whole load more and I've not checked W7 to see what happens there.
Stuart
-
Eric your help info is of course accurate and the web interface does not have the same password as the telnet one on the HG612. Annoying as so far I've not found any way of changing it, although I can restrict acces for telnet from specific local lan addresses. So I now have rs-ux working fine.
The popup boxes do work fine on Window7 by the way.
Stuart
-
Thanks Stuart.
-
Eric re the pop up box I have managed a snapshot of what happens on my machine attached here.
Stuart
-
That's really odd. It looks as though the box is the right size for the text, but the text gets cut off part of the way across. I might suspect a window redrawing issue, but probably not if it's repeatable. I really don't know about this one.
-
It's repeatable Eric everytime I hover over the messages. Actually it's not repeatable ....... grrrrr I just tried it and it is perfect, only difference is that it took two snapshots a few minutes ago of SNRM and Connection Speed, dont know if that helps. I dont want to shut it down now as I am trying to monitor the HG612 but later tonight I will close it before I shutdown the PC so I'll see what happens when I close the program and then start it anew.
Stuart
-
I can't think of any way that the popup hints would be affected by taking a couple of snapshots. But the fact that it doesn't happen every time is bringing me back to thinking about a window redrawing issue. Is the PC in question possibly a rather low power model?
-
Is the PC in question possibly a rather low power model?
No its an AMD Phenom quad core so it's quite fast ;)
Stuart
-
Eric may I make a suggestion? I have been using rs-ux to see how my HG612 behaves and one thing I think might be useful is somewhere to record the time and level of the highest and lowest snrm values say in any 1 day. This would mean that I could go directly to the relevant snapshot to see the actual graph containing the values recorded. Does anyone else see any value in this?
Stuart
-
That's a useful suggestion Stuart. It's been in my mind that scrolling the graphs becomes a bit unwieldy after the program has been recording for some time, and looking through lots of snapshots is just as unwieldy. I could display this extra data on the Stats page, and add buttons on the graph pages to jump directly to the relevant positions on the graphs.
-
Eric ref the pop up problem it happens on a fresh start of the program every time. Not always exactly the same symptom as this morning the pop up is blank in each case. I suspect it cures itself later but not sure when exactly - maybe when the graph fills and starts scrolling but I've not been around to check yet!
Stuart
-
I suspect it cures itself later but not sure when exactly - maybe when the graph fills and starts scrolling but I've not been around to check yet!
The popup hints are completely independent of anything the program is doing (except insofar as it's all using the same system resources). So it's hard to see any way it would 'cure' itself, unless the clue is in the way Linux caches program data in memory, or something like that.
-
Strange that today nothing has changed, they are still blank!
Stuart
-
Stuart, could you let me know what distro this system uses, and what desktop? I've raised the issue on the Lazarus (development system) forum, and that question has been asked.
-
I am using Fedora 17 with KDE 4.9.2 all fully up to date. I also use an ATI gfx card although not a high end gaming one it's an hd4350.
Stuart
PS. just to say I get pop up boxes in many other applications of the exact same colour and font and they all work fine.
-
Stuart, I've uploaded Linux 32-bit and 64-bit versions which have been built with Qt instead of GTK2 which I've normally used for the Linux versions. The Qt versions will be suitable for use on KDE systems only. I would appreciate it if you could try this version to see if it fixes the popup issue.
https://docs.google.com/folder/d/0BxbUtOYVZ_SCZ1BnY3RKYnN2b1U/edit
For explanation, I think that the issue may be the result of a small amount of incompatibility between GTK2 applications and the KDE desktop. I've noticed on my own KDE systems that there are occasional slight redraw problems with GTK2 programs, and these don't occur with Qt programs.
-
Eric it wont run because it cannot find libQt4Pas.so.5 so I obviously need another package installed but not yet found which one I need.
Stuart
-
Sorry about that. That library is part of the development system and shouldn't be needed for running the compiled program. I'll find out what's gone wrong.
-
Would it be possible to implement a 'lowest SNRM' and 'highest SNRM' feature similar to RouterStats?
-
Would it be possible to implement a 'lowest SNRM' and 'highest SNRM' feature similar to RouterStats?
I'm not familiar with that feature in Routerstats. What does it do - just keep a record of the highest and lowest SNRM values over a period? If so, it will be easy to implement.
-
Would it be possible to implement a 'lowest SNRM' and 'highest SNRM' feature similar to RouterStats?
I'm not familiar with that feature in Routerstats. What does it do - just keep a record of the highest and lowest SNRM values over a period? If so, it will be easy to implement.
When one right clicks on the SNRM graph, it gives one a 'lowest SNRM' and 'highest SNRM'.
-
Thanks, in that case I should be able to do it before releasing v 1.0. I don't want to add any major new features at this stage, because I want to concentrate on clearing the last couple of issues with the RC1 version, but this is a small feature.
-
Sorry about that. That library is part of the development system and shouldn't be needed for running the compiled program. I'll find out what's gone wrong.
Not sure what you have found but I did take a look around and it seems that solving this does not look easy from what I have read. I would not worry too much about the issue of the pop ups, it's an annoyance rather then affecting the programs use.
Stuart
-
Not sure what you have found but I did take a look around and it seems that solving this does not look easy from what I have read. I would not worry too much about the issue of the pop ups, it's an annoyance rather then affecting the programs use.
Stuart
Yes, thanks. There must be a solution (apart from including a copy of the qt4pas library with rs-ux) because I just did a brand new installation of Debian Testing 64-bit KDE, and the Qt version of rs-ux worked fine on this, even though the qt4pas library wasn't installed. But another Qt program I built produces the same error as you reported. So it's not at all clear to me why some Qt programs run on some systems, but not all. I think I'll do what you say, and shelve this for the time being.
-
Eric I have had another thought about a possible addition to the program. I am obviously running rs-ux while playing with the router. This afternoon it did a re-sync while I was watching however had I not been here it would not have been as obvious, I happen to notice the AS uptime reset. I was thinking that you could have an optional log file which could be used to record things like the time AS uptime resets, highest and lowest snrm etc.
Stuart
-
Yes Stuart, an event log sounds like a good idea. I'll put it on the todo list.
-
Just had a funny with this version. Came back from shopping and looked at the main screen only to see it had stopped graphing about 2 hours previously, thought it might have hung completely so clicked stop and that worked, followed by a pop about corruption to which I said OK and it then allowed me to start graphing again without seemingly any problems. The program was the only thing left running (other than system processes) while I was out for a couple of hours. The router was still working normally as my wife was using the laptop on the net OK. AS uptime indicates the router has not done any re-syncs during this period, the last one being when I set the snrm back to default. There is no way she could have attempted to login to the router as she has no knowledge of these things!
Stuart
-
Did you take a note of the actual wording of the error message?
-
Did you take a note of the actual wording of the error message?
No sorry, it's one I've seen before about OK to continue or Cancel to end but sorry I did not capture the wording about what the error was. Will do better next time :-[
Stuart
-
Never mind, I think I know the message you're referring to. This probably has the same cause as the freeze reported by Bald_Eagle1, and as it's fairly rare it's the devil to diagnose.
-
Just had an error this morning, no hang or anything just an Access Violation. Clicked cancel to end program and restarted OK.
Stuart
-
Just had an error this morning, no hang or anything just an Access Violation. Clicked cancel to end program and restarted OK.
Just a thought Stuart,
Are you also running any derivatives of my harvesting/graphing scripts at the same time as rs-ux/rs-w?
I found that one or the other can freeze, presumably when both attempt to access the modem/router at the same time.
Even staggering the timing of the data harvesting doesn't quite resolve the issue as it appears that eventually they catch up with each other.
My comments may be a red herring though.
As mentioned by Eric, it is quite rare & the devil to diagnose, thus cannot yet be definitely attributed to a timing clash.
Do you maybe have any other software running that may just attempt to access the modem/router at the same time?
-
No nothing else runs accessing the router, rs-ux is the only thing, apart from normal internet access ofcourse.
Stuart
-
roseway, running rs-w with A NETGEAR DG834GT, the application doesn't graph any FEC data on the FEC tab.
-
roseway, running rs-w with A NETGEAR DG834GT, the application doesn't graph any FEC data on the FEC tab.
Is your connection interleaved?
-
Is your connection interleaved?
Yep. Interleaving depth of 64 at the moment.
-
OK, could you please copy the result of the command "adslctl info --stats" so I can see what's missing. Thanks.
-
Here you go:
# adslctl info --stats
adslctl: ADSL driver and PHY status
Status: ShowtimeRetrain Reason: 8000
Channel: FAST, Upstream rate = 1076 Kbps, Downstream rate = 7752 Kbps
Link Power State: L0
Mode: ADSL2+
Channel: Fast
Trellis: U:ON /D:ON
Line Status: No Defect
Training Status: Showtime
Down Up
SNR (dB): 3.6 5.3
Attn(dB): 43.5 22.9
Pwr(dBm): 0.0 13.0
Max(Kbps): 8244 1084
Rate (Kbps): 7752 1076
G.dmt framing
K: 121(0) 154
R: 10 0
S: 1 1
D: 64 1
ADSL2 framing
MSGc: 59 10
B: 120 153
M: 1 1
T: 2 1
R: 10 0
S: 0.4973 4.5461
L: 2107 271
D: 64 1
Counters
SF: 1166855 1047853
SFErr(CRCErr): 71 0
RS: 151691208 0
RSCorr: 359924 0
RSUnCorr: 891 0
HEC: 64 0
OCD: 0 0
LCD: 0 0
Total Cells: 344877997 2278051
Data Cells: 114336912 8027
Drop Cells: 0
Bit Errors: 0 0
ES: 47 0
SES: 0 0
UAS: 38 0
AS(Uptime): 18862
INP: 1.21 0.00
PER: 16.16 18.18
delay: 7.95 1.13
OR: 32.16 7.03
Bitswap: 2942 0
Total uptime: 18862
Total time = 5 hours 15 min 24 sec
SF = 1166855
CRC = 71
LOS = 0
LOF = 0
ES = 47
Latest 1 day time = 5 hours 15 min 24 sec
SF = 1166855
CRC = 71
LOS = 0
LOF = 0
ES = 47
Latest 15 minutes time = 24 sec
SF = 1551
CRC = 0
LOS = 0
LOF = 0
ES = 0
Previous 15 minutes time = 15 min 0 sec
SF = 55656
CRC = 1
LOS = 0
LOF = 0
ES = 1
Previous 1 day time = 0 sec
SF = 0
CRC = 0
LOS = 0
LOF = 0
ES = 0
15 minutes interval [-30 min to -15 min] time = 15 min 0 sec
SF = 55657
CRC = 5
LOS = 0
LOF = 0
ES = 3
15 minutes interval [-45 min to -30 min] time = 15 min 0 sec
SF = 55718
CRC = 1
LOS = 0
LOF = 0
ES = 1
15 minutes interval [-60 min to -45 min] time = 15 min 0 sec
SF = 55718
CRC = 3
LOS = 0
LOF = 0
ES = 3
Showtime Drop Reason: 8000
Last Retrain Reason: 8000
#
-
There's the explanation then. The router isn't reporting any FECs. But it does report RSCorr, which on my HG612 is the same as the number of FECs, so I'll have to look at getting the FEC data from there. I won't be able to do it immediately, because I'm not too well at the moment, and don't feel like doing anything which taxes the brain.
-
If you are desperate to see your DS & US FEC/RSCorr graphs separately in the meantime, you could always use the other graphing scripts, either for the HG612 or the DG834GT.
(as per the attached examples from when your connection was ADSL1 (G.DMT))
-
Just had another hang, nothing on the rs-ux window - it had gone balnk and was using 100% or 1 core of my quad cpu. System did kill it when I clicked on the window x at top right. No way of knowing what it was doing.
Stuart
-
I'm struggling with this one, I'm afraid. I've had it happen once myself, but I couldn't identify any cause.
-
Eric I've had a couple of crashes with the usual OK or Cancel box but not much else to go on. Without me having to install your compile environment is there any way I can run it in some kind of debug more, perhaps gdb or the like? Might be able to catch something then.
Stuart
-
Stuart, thanks for the offer. I can build a debug version, and in fact I should have done so earlier, but real life has got in the way. :)
Let me know what version you use, and I'll do it ASAP.
-
Eric I run the 32-bit Linux version on Fedora 17. I don't mind installing some extras if needed so let me know and I'll give it a go.
Stuart
-
Thanks Stuart. As it happens, I've just received some helpful comments on the source code from another person, and I'll try to incorporate his suggestions first, then build a debug version to test.
-
Eric that's great. I will be unable to do any testing next week (a week in N Devon :) ), hopefully others will be able to start testing if you are ready before I am back.
Stuart