Kitz Forum
Broadband Related => Router Monitoring Software => Topic started by: roseway on May 08, 2013, 11:41:29 AM
-
I've got tired of referring to the program by two different names, so I've decided to give it a more meaningful name, DSLstats. When you first run this version it should pick up your previous rs-ux/w configuration if it exists, and copy it across to the new name. You may need to check that the router login password is still correct. The snapshot directory will remain the same, so you may want to edit it.
Full list of changes in this version:
- changed program name to DSLstats (previously rs-x/rs-w)
- added option for logarithmic Y axis on CRC graph
- added option to disable popup hints
- passwords are now encrypted in the .ini file
- added option to save snapshots in daily folders
- added page of advanced tweaks
- the SNRM tweak value can now be changed in increments of 0.1 dB
- added option to hide upstream or downstream data on 'per minute' graphs
- added some extra data to the Stats page
- added options to start minimised and to start recording on program launch
- event log is now displayed by default
- fixed issue with some systems in which autosizing of text items didn't work correctly
- fixed issue whereby the bitswaps per minute graph would plot continuous zeroes after a resync
- fixed issue whereby the interleaving values were read from the wrong section on some systems
- fixed misbehaviour of max sustainable speed graph after a re-sync
- corrected absence of DSLAM/MSAN type in Stats page with some systems
http://dslstats.plainroad.me.uk
-
Eric I cant run the 32-bit linux version without it crashing with an access violation. Initially I tried using my rsuxc settings but then deleted them and started afresh but still no joy it crashes. It happens right at the end of sampling.
Stuart
-
I'll check that out as quickly as I can, Stuart.
-
I've checked it out and it runs with no problems here. I've downloaded the 32-bit Linux version from the website, and that also runs with no problem. So I'm a bit mystified.
Could you try running it from a terminal, to see if any error messages are returned?
-
Eric
./dslstats
TLResourceList.Sort 53 DUPLICATE RESOURCE FOUND: TBitchart:PNG
TLResourceList.Sort 54 DUPLICATE RESOURCE FOUND: TBitchart:PNG
TLResourceList.Sort 57 DUPLICATE RESOURCE FOUND: TLinechart:PNG
TLResourceList.Sort 58 DUPLICATE RESOURCE FOUND: TLinechart:PNG
TLResourceList.Sort 60 DUPLICATE RESOURCE FOUND: TLinediagchart:PNG
TLResourceList.Sort 61 DUPLICATE RESOURCE FOUND: TLinediagchart:PNG
TApplication.HandleException Access violation
Stack trace:
$08305597
$08315645
$08313BE3
$0813AA8E
$08154F21
$08156930
$08306FE6
$081387A8
$0812B51B
$0812CC46
$081D065B
$0819FA49
$081AE04B
$081AE1C1
$4E6BEED2
WARNING: TLCLComponent.Destroy with LCLRefCount>0. Hint: Maybe the component is processing an event?
is what shows up in terminal at end of sampling.
Stuart
-
Does sampling seem to take the normal time, or is it curtailed?
-
Seems to me that sampling takes about the normal time. I'll try turning off some of the options to see if I can see any difference.
Stuart
Edit: just tested with only SNR and connection speed and still crashes
-
I've just had a thought - what router are you currently using? I did make a change to the way I extract the connection speed and max attainable speeds, so it's just possible that something has gone astray there. If you're not using the HG612, could you please copy the output of the --stats command to me.
-
Eric it's an HG622.
# xdslcmd info --stats
xdslcmd: ADSL driver and PHY status
Status: Showtime
Retrain Reason: 8000
Max: Upstream rate = 1120 Kbps, Downstream rate = 9724 Kbps
Channel: INTR, Upstream rate = 1020 Kbps, Downstream rate = 8758 Kbps
Link Power State: L0
Mode: ADSL2+
TPS-TC: ATM Mode
Trellis: U:ON /D:ON
Line Status: No Defect
Training Status: Showtime
Down Up
SNR (dB): 6.5 9.2
Attn(dB): 42.0 21.3
Pwr(dBm): 0.0 12.7
ADSL2 framing
Bearer 0
MSGc: 42 10
B: 122 145
M: 2 1
T: 3 1
R: 8 0
S: 0.8963 4.5447
L: 2267 257
D: 64 1
Counters
Bearer 0
SF: 477402234 922991
SFErr: 183973 3522
RS: 13222377 4293983
RSCorr: 69606993 0
RSUnCorr: 1273825 0
Bearer 0
HEC: 855901 1955
OCD: 474 0
LCD: 474 0
Total Cells: 196119830 1334086945
Data Cells: 469939540 64859014
Drop Cells: 0
Bit Errors: 60302306 167193
ES: 50713 3011
SES: 1480 0
UAS: 104 104
AS: 7701776
Bearer 0
INP: 0.50 0.00
INPRein: 0 0
delay: 14 1
PER: 16.13 18.17
OR: 23.80 7.04
Bitswap: 3802847/3805744 52838/52838
Total time = 1 days 8 hours 38 min 50 sec
FEC: 74168886 0
CRC: 188295 3732
ES: 50713 3011
SES: 1480 0
UAS: 104 104
LOS: 0 0
LOF: 0 0
Latest 15 minutes time = 8 min 50 sec
FEC: 24019 0
CRC: 335 0
ES: 83 0
SES: 1 0
UAS: 0 0
LOS: 0 0
LOF: 0 0
Previous 15 minutes time = 15 min 0 sec
FEC: 52181 0
CRC: 750 0
ES: 160 0
SES: 1 0
UAS: 0 0
LOS: 0 0
LOF: 0 0
Latest 1 day time = 8 hours 38 min 50 sec
FEC: 931621 0
CRC: 15054 38
ES: 2517 37
SES: 97 0
UAS: 0 0
LOS: 0 0
LOF: 0 0
Previous 1 day time = 24 hours 0 sec
FEC: 550676 0
CRC: 5835 74
ES: 1753 38
SES: 17 0
UAS: 0 0
LOS: 0 0
LOF: 0 0
Since Link time = 49 days 17 hours 2 min 46 sec
FEC: 69606993 0
CRC: 183973 3522
ES: 49028 2835
SES: 1469 0
UAS: 0 0
LOS: 0 0
LOF: 0 0
#
Stuart
-
Thanks Stuart, I think that may explain the problem.
I've got some computer problems at the moment, and I'm using a different PC for this, but I'll get back to the problem as soon as I can. :)
-
I've fixed the problem now. Different versions of the Broadcom CLI report the connection speeds in several different ways, and in making a change in this area to fix another problem I'd inadvertently messed up the processing of the data from the HG622. I've got it running properly now with an HG622, and I'll upload the corrected version shortly.
-
Thanks Eric, superb support as always.
May I ask one thing, so far I've not seen an option to get to the new version screen while the program is running, if I'm right and there is not an option can you add an option for manually checking for a new version, and perhaps an option for checking every n hours or days? Would be useful for anyone running the program 24x7.
Stuart
-
Corrected version 3.1 is now available for download: http://dslstats.plainroad.me.uk
Stuart, the program should check every 12 hours for a new version. Perhaps this is too long a period, and I should reduce it to (say) every three hours? And I could certainly add an option to manually check for an upgrade.
-
Corrected version 3.1 is now available for download: http://dslstats.plainroad.me.uk
Stuart, the program should check every 12 hours for a new version. Perhaps this is too long a period, and I should reduce it to (say) every three hours? And I could certainly add an option to manually check for an upgrade.
I'll give the 3.1 version a go now.
As to the checking for new versions I'd make it default to 24 hours and provide an option to do it manually, personally I think 12 hours is a bit short. Maybe you could provide an option for the user to decide how long between checks, I'd go for a minimum of 1 day upwards to say 30 days and give a Check Now option as well. Just my view on it.
Stuart
Edit: Success Eric it is running fine.
As an aside just noticed that my line did a re-sync today after 90 days of uptime. Just wonder if TT force a re-sync after 90 days? Could not see a reason for it otherwise as the SNR was good.
-
Thanks for letting me know, Stuart. I'll have a think about your suggestion for upgrade checking, but it sounds reasonable.
90 days of uptime, eh? At the present time I'm pleased if I get one day. Earlier this afternoon it gave up altogether (still in sync, but wouldn't connect to TT). So I switched the router off for 45 minutes, and at the moment it's OK. :fingers:
-
soon as I run this new version (3.1) i had lots of strange behaviour on my computer, this is what happened.
the DSL Stats app opened straight to tray, i could not open it from the tray, i opened task manager and i could not kill the dslstats.exe *32 process, then all of a sudden it looked like it was logging me out as my screen blended to black (like does on shutdown) and it made the "disconnection" noise..the screen came back on and said a USB device had failed? also internet connected was lost but came back on, Task Manager was showing only half the stuff in task manager too..
Iam using Windows 7 x64 on a Gigabyte H77 mobo with 8Gb DDR3 ram and 120Gb SSD - I have running Avast Internet Security, JD S Auto Speedtester, Share-X, GreedyTorrent, Sky Drive and Via HD Audio Deck, oh and my mobile phone Sony Xperia Miro was charging via USB...the only other USB connection is a JoyPad and USB wireless Keyboard and Mouse transceiver.
I will unplug the phone, reboot and try to run it again and see what happens.
-
There's just one thing I can think of - I've realised that, if your internet connection is down when you start up the program, there is quite a long timeout before it gives up checking for an updated version and displays normally. I suppose it's possible that things could have misbehaved in the way you describe as a result of the long timeout, but it certainly puzzles me.
I suggest that you delete rsw.ini (the ini file for rs-w) and dslstats.ini if it exists to start it afresh. I have reduced the timeout to 5 seconds now, but I've done that since v3.1.
-
...well I reboot and this time its working a bit better (had some freezing when flicking between configuration pages but only once for a few seconds) also, this time Avast was all over it checking it saying it may be a suspicious file but didnt block it, previously it did not ...strange?
I deleted all RS-W files, there was only DSL-Stats files in the folder.
anyway this version looks brilliant and thanks for adding the suggestions I had, really appreciate it :) :) :thumbs: :clap:
but there are some problems: (im using Netgear DG834Nv1 with DGTeam 1018)
1) AS Uptime isnt reported in the bottom window
2) Nothing is logging in BitSwaps per tone (it is checked for monitoring)
3) The BitSwaps Per minute graph is showing zero (which I dont believe is correct)
(https://forum.kitz.co.uk/proxy.php?request=http%3A%2F%2Fi.imgur.com%2FamHFjge.jpg&hash=280162c5992a5ec050f6160d150eda7838ff5de6)
(https://forum.kitz.co.uk/proxy.php?request=http%3A%2F%2Fi.imgur.com%2FBSWHqSO.jpg&hash=297ecd2cea1aa448043567c05afb22c687456107)
(https://forum.kitz.co.uk/proxy.php?request=http%3A%2F%2Fi.imgur.com%2FC6a3zAm.jpg&hash=c4da8484861c13ffbb87c05eefd1562ddc5795bc)
the missus is calling on me to strip beds and do house work so u may be a while hehe
thanks again
-
Could you please copy the output of the --stats command to me - hopefully that will reveal the reason for the missing data.
-
since that 'crashing/funny' business I had before my mobile phone no longer connects too the computer, it says in DevMgr that the software required is corrupt in the registry? - this happened as soon as I launched the DSL-Stats program, ive tried uninstalling and re-installing but it just says the same thing, so Im gunna restore the computer and retry - last night I installed Avast Internet Security on 20 day trial and Im wondering if its related to that? --thing is everything was working fine after installing it, and using the comp today was fine until the very moment I executed the DSLstats app.
after leaving for a while there was some data in the the bitswaps graph but no where near what should have been? - ive since restarted the comp and no graphs have been saved on exit, when I run DSLstats it didnt start minimized and is asking for my login data again? when ive checked all the data/settings has been reset/forgot? perhaps its because I restarted windows without closing the program first? is there anyway to get around that? so when i shutdown windows it automatically asks DSLstats nicely to shutdown (so it saves the graphs)..? I think I used some tool to set aggressive shutdown times etc and forcibly shutdown unresponsive apps.
--stats log:
adslctl: ADSL driver and PHY status
Status: ShowtimeRetrain Reason: 8000
Channel: FAST, Upstream rate = 1204 Kbps, Downstream rate = 17600 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): 7.1 8.6
Attn(dB): 27.5 13.6
Pwr(dBm): 18.6 12.3
Max(Kbps): 21277 1260
Rate (Kbps): 17600 1204
G.dmt framing
K: 145(0) 16
R: 12 16
S: 1 8
D: 96 8
ADSL2 framing
MSGc: 51 15
B: 144 15
M: 1 8
T: 4 7
R: 12 16
S: 0.2846 3.6455
L: 4412 316
D: 96 8
Counters
SF: 36263606 36777614
SFErr(CRCErr): 280 8
RS: 3973135088 1275598
RSCorr: 448179 3832
RSUnCorr: 13446 0
HEC: 237 14
OCD: 0 0
LCD: 0 0
Total Cells: 2949229205 1896513748
Data Cells: 218702120 19742787
Drop Cells: 0
Bit Errors: 0 1639
ES: 306 105
SES: 2 367
UAS: 152 0
AS(Uptime): 588442
INP: 1.04 1.62
PER: 16.22 16.74
delay: 6.83 7.29
OR: 28.10 10.03
Bitswap: 42971 305
Total uptime: 795525
Total time = 1 days 5 hours 2 min 20 sec
SF = 49123081
CRC = 1360
LOS = 2
LOF = 18
ES = 306
Latest 1 day time = 5 hours 2 min 20 sec
SF = 1117929
CRC = 9
LOS = 0
LOF = 0
ES = 8
Latest 15 minutes time = 2 min 20 sec
SF = 8653
CRC = 0
LOS = 0
LOF = 0
ES = 0
Previous 15 minutes time = 15 min 0 sec
SF = 55507
CRC = 1
LOS = 0
LOF = 0
ES = 1
Previous 1 day time = 24 hours 0 sec
SF = 5324537
CRC = 22
LOS = 0
LOF = 0
ES = 17
15 minutes interval [-30 min to -15 min] time = 15 min 0 sec
SF = 55446
CRC = 0
LOS = 0
LOF = 0
ES = 0
15 minutes interval [-45 min to -30 min] time = 15 min 0 sec
SF = 55445
CRC = 0
LOS = 0
LOF = 0
ES = 0
15 minutes interval [-60 min to -45 min] time = 15 min 0 sec
SF = 55507
CRC = 2
LOS = 0
LOF = 0
ES = 2
Showtime Drop Reason: 8000
Last Retrain Reason: 8000
#
-
Thanks for the --stats info.
DSLstats doesn't save the configuration settings until it shuts down, so presumably your forced shutdown stopped that happening. I could arrange to save the settings periodically, or perhaps just when the "Start recording" button is pressed.
As far as I know, the bitswap graphs accurately reflect the reported data, but it's hard to check without sitting over the program, taking regular notes of the reported bitswap values, and seeing if they match up with the graph.
-
Im just popping in to update you, I currently have no AV installed at all.
too keep it short about my USB issues, running DSL Stats has cause a crash on my system thats done something to windows USB device drivers or the regsitry (as I was getting corrupt registry error in devmgr for my mobile phone), Ive managed to get my Xperia Mobile working again (by removing USB, uninstalling, removing power from system and turning on and reinstalling) but can not get my Joypad working again using same method...well, I did once but had to change USB port using method I mentioned..but shutdown the computer & start back up it was as if it was uninstalled and again driver installed failed.?? just cant get my head around it? error 43
anyway, back to DSL Stats :- before with old 2.91 version I would have loads of bitswaps, now I havent? I can show you graphs before and after (see below_, just doesnt look right? unless my router is suddenly having a short period of no bit-swappping? - also the AS Uptime is missing.
EDIT: Iam getting very small amounts of bitswap recording in Bit-Swaps Per Minute - but nothing reflected in Bit-Swaps Per Tone? (see images below) - EDIT 1.5: I lie .. now there is upload bitswaps showing in BS per tone but no download BS's showing?
EDIT 2: It seems to have issues running alongside DMT Tool (which RS-W didnt) - ive been getting "no data in last response" and "no response from router" when DMT Tool was running only for few minutes.
when I ran it just now it said no response from the router, so i closed it, deleted all the files (again about 4th time ive done this) then extracted them again from the zip file and try again and set it up again, closed it to save the settings ??? and its working. 8) :)
it would be great if it saved settings as they were made...or had SAVE/APPLY button somewhere? because if one shuts down windows without first exiting the program it wont remember their settings or save images etc... I dont recall having any problems with RS-W in this respect? it seemed to me that the settings were saved as amended? as i never shut down RS-W I just shut down windows.
also, I think it should start all graphs up in compressed view... expanded view should only be chosen by the user for a more refined inspection, the save to folder works great but any "expanded view" images (QLN, HLOG) dont save as a viewable image, it says INVALID IMAGE, when i change to compressed view and exit the program and re-check its ok..
another suggestion is combine the SNRM Tweak and Additional Tweaks in a manner so an SNRM Tweak command can be execute along with the others.. this is because your current setup means you can have either and not both, so if im manually using 3db SNRM and want to enable i24k or PHyRe then my SNRM will go back to sky's default of 7db (with i24k and PHyRe on) - I used to set commands in DMT Tool to execute whatever I wanted with the SNRM tweak included so I'd get both.
here you can see DSL Stats Bit-Swaps Per minute:
(https://forum.kitz.co.uk/proxy.php?request=http%3A%2F%2Fi.imgur.com%2FuhR6u9a.jpg&hash=bf9323dd00cf4c8b422d81104d8c9f861e0cf397)
here you see nothing reflected in Bit Swaps Per Tone:
(https://forum.kitz.co.uk/proxy.php?request=http%3A%2F%2Fi.imgur.com%2FMoDbU0S.jpg&hash=93b862a829ec9f28e1f90a5c027f4a0cea374a3c)
here is what my bit-swaps per minute usually look like:
(https://forum.kitz.co.uk/proxy.php?request=http%3A%2F%2Fi.imgur.com%2F9T0NPzN.jpg&hash=4b4e386da414761d9de09ed4490a4849a93872bb)
heres what my bitswap per tone would normally look like after a bit run:
(https://forum.kitz.co.uk/proxy.php?request=http%3A%2F%2Fi.imgur.com%2Fr1koF2j.jpg&hash=359118409ca550e6c971e73bb445b477bb5b2fa3)
Iam going for my tea now and walk the dog... when i get back I need to try and sort this USB Joypad issue out and have a closer inspection on these bitswaps
thanks again for your efforts Eric... sorry to be moaning lol...
-
There's a lot there to absorb, but just a couple of random thoughts:
Bitswaps per tone: this graph is cumulative, so the longer you run it the more data is shown. So you can't really compare the new with the old. If you've continued to run v3.1 I expect that this graph will have accumulated a lot more data.
Concerning running the program alongside DMT-tool, I would be surprised if the new version is really different from the earlier version, because the code which manages telnet access and collecting the data didn't change between the two versions. Having two different programs accessing the router's telnet interface at the same time will always have the possibility of conflict, and my first reaction is that this is just a case of random timing causing the conflict. But I could be wrong. :)
-
I used to have both running all the time and never had any problems, if memory serves me right so did router stats app have issues running with Dmt.. Or was it the graphing scripts? Can't remember
Dmt only polls once then once in blue moon while it's open so shouldn't affect dslstats from polling the router..??
EDIT: I may ask the bald eagle to get involved as he knows summit about Dmt tool access vs other apps I'm sure it's something to do with login
She is using the comp just now, I will test it when I get on, also it would have been running for hours so should have some decent bit swap stats..
Surely if there's been at least 1 bit swap that shows in the graph then there should be something in the swaps per tone graph? As you can see from the graph there's some downstream bit swaps but nothing in the per tone graph? Oh well I will find out soon, think I will run rs-w too and compare... Then I can be certain
Sent from my Sony Xperia Miro on Tapatalk
-
...OK it seems to be working alright now... apart from the uptime being missing of course, Iam running DMT Tool and its working alongside it still - the thing is though...what was causing it to fail before? its in the log:
09 May 2013 19:08:49 Recording started
09 May 2013 19:08:53 Recording started
09 May 2013 19:08:53 Previous sample still being processed
09 May 2013 19:15:52 Previous sample still being processed
09 May 2013 19:16:23 Exception during telnet login
09 May 2013 19:16:23 Exception during telnet login
09 May 2013 19:16:52 Router failed to respond to telnet request
09 May 2013 19:17:52 Previous sample still being processed
09 May 2013 19:18:22 Exception during telnet login
09 May 2013 19:18:22 Exception during telnet login
09 May 2013 19:18:52 No stats received
09 May 2013 19:19:22 Router failed to respond to telnet request
09 May 2013 19:19:52 Router failed to respond to telnet request
09 May 2013 21:09:21 Auto snapshots taken
09 May 2013 23:09:41 Auto snapshots taken
10 May 2013 01:10:01 Auto snapshots taken
Next time I restart DSL-Stats I will monitor the bitswaps per minute and per tone graphs... am I right in saying that if there is ANY activity logged in the per minute graphs that there should be something in the per tone graphs? (or do the per tone graphs generate every X minutes?)
I noticed you put the RS Percentage in for me :) :) - well chuffed - oh and the DLSAM info works too :) :)
cheers
EDIT: bug report - when changing colors for download...everything but the "bit swaps per minute" change color, they remain red...even if i change the 'other data' color to blue (like I did with download just messing about)
EDIT2: bug report - SNR Graph sometimes doesnt save correctly, well hasnt for last 3 saves, I have since disabled it in hope that it will be included together in the Bit-loading graph rather than seperate graphs...all other graphs are fine
-
I used to have both running all the time and never had any problems, if memory serves me right so did router stats app have issues running with Dmt.. Or was it the graphing scripts? Can't remember
Dmt only polls once then once in blue moon while it's open so shouldn't affect dslstats from polling the router..??
The issue (I think) is whether the two programs happen to be collecting data at the same time. If DMT only polls the router occasionally, then the conflict will occur only rarely. In fact I suspect that it's even more rare than this, and that it depends on what stage the data collection has got to for each program. In the course of development I frequently have two copies of DSLstats (or rs-ux/w) running at the same time, and it's a long time since I experienced any clashes.
In the case of Routerstats, I understand that it keeps the telnet interface open continuously and only disconnects when the user stops it running. So there would seem to be more possibility of conflict with other programs in this case.
-
EDIT: bug report - when changing colors for download...everything but the "bit swaps per minute" change color, they remain red...even if i change the 'other data' color to blue (like I did with download just messing about)
A small omission on my part. I'll correct it.
-
well been running all night with no apparant issues...
is the RScorr from when DSL stats stared and then 24 hours back? or is it from the total available in the router (sinc link time)
thanks
-
Bitswapping:
Not a bug but a feature/consequence
My router a Dlink 2640B does not report bitswaps as a specific line in the ADSL2 stats. By I mean the line is not there at all (nor the INP, PER, Delay, OR lines either).
Therefore your bitswaps per time graph is zero, as this is where you are taking the data from.
Somewhat confusingly to those not in the know, the bitswaps per tone graph IS there. This will be because you are presumably collecting this data by comparing the change in the bit loading data’s at each tone from sample to sample.
I presume that Routerstats which does show the bitswaps/time is collecting the bitswaps total data by summarising the changes in the D/S and U/S bit loading data obtained for each tone rather than relying on the bitswaps total figure actually being provided as a specific line entry on the stats.
I confirm that the program routerstats does keep the telnet session open continuously. There is a terminal screen/tab in the program where you can type commands as if you were at the telnet prompt.
-
@c6em: You're correct about the source of the bitswaps per tone data. I did look at the possibility of summing up that data to get the total bitswaps, but it resulted in a figure which was much higher than the figure reported by the router in the Bitswap: line. This is why I added a note on the bitswaps per tone graph to say that it wasn't accurate in absolute terms.
The other difficulty is that I don't know how often the router swaps bits. It's quite possible that several bits could be swapped back and forth between samples, but the bitloading ends up the same in one sample as it was in the previous sample.
What I'm saying in a rather long-winded fashion is that I don't really know how best to proceed with routers which don't report the total bitswap values. I don't want to report data of dubious accuracy if I can help it, but I take your point about the confusion of a blank graph.
-
is the RScorr from when DSL stats stared and then 24 hours back? or is it from the total available in the router (sinc link time)
thanks
The figures on the Stats page are the total figures as reported by the router, so they count from the last re-sync.
-
Can't thank you enough for this program and adding my requests to it... THANK You :-)
Before I has to use Dmt and just counted back the decimal places to get a rough idea of what it was in percent and I never bothered with rsuncorr so nice to have those.. Also the i24k etc.... Just for fun: the SNRM tweak can not be applied when used with the i24k tweak... Never works for me... Think the noise margin has to be running at tsnrm for i24k to work..
Sent from my Sony Xperia Miro on Tapatalk
-
My bitswaps per tone graph hasn't worked for around a month now, but I have around 50 Bitswaps per minute so I'm assuming this is something to do at my end?
-
My bitswaps per tone graph hasn't worked for around a month now, but I have around 50 Bitswaps per minute so I'm assuming this is something to do at my end?
There are some routers which don't refresh the bitloading except when there's a re-sync. Is it possible that you have one of those? Does the bitloading graph show changes from sample to sample?
-
There are some routers which don't refresh the bitloading except when there's a re-sync. Is it possible that you have one of those? Does the bitloading graph show changes from sample to sample?
FWIW, I have noticed that Bitloading graphs using the Huawei HG612 with the new BLOB don't change from sample to sample via both your program & mine, whereas it does when using the original unlocked BLOB.
I have also noticed the Huawei HG622's GUI doesn't auto-refresh every few seconds either, but the data from the command 'xdslcmd info --stats' & Bitloading data does update whenever it is sampled.
-
FWIW, I have noticed that Bitloading graphs using the Huawei HG612 with the new BLOB don't change from sample to sample via both your program & mine, whereas it does when using the original unlocked BLOB.
There is a posting by Asbokid, somewhere in this forum, where he mentions that fact. ::)
-
Yes, it came up in a discussion starting here: http://forum.kitz.co.uk/index.php/topic,12450.msg235213.html#msg235213
-
As an aside just noticed that my line did a re-sync today after 90 days of uptime. Just wonder if TT force a re-sync after 90 days? Could not see a reason for it otherwise as the SNR was good.
To come back to this question, I've been having a bit of instability recently, and since it started, my connection re-syncs in the small hours each morning for no apparent reason. Each time it re-syncs the interleave values seem to change slightly. So I'm guessing that TT's DLM is taking note of error rates and so on, and trying to optimise the connection. Perhaps your re-sync was also the result of action by DLM?
-
As an aside just noticed that my line did a re-sync today after 90 days of uptime. Just wonder if TT force a re-sync after 90 days? Could not see a reason for it otherwise as the SNR was good.
To come back to this question, I've been having a bit of instability recently, and since it started, my connection re-syncs in the small hours each morning for no apparent reason. Each time it re-syncs the interleave values seem to change slightly. So I'm guessing that TT's DLM is taking note of error rates and so on, and trying to optimise the connection. Perhaps your re-sync was also the result of action by DLM?
I have asked on their forum but the reply was 'it must have been a power problem or noise spike' which I know from running your program it was not. There recent error rates have not been any different to the past 90 days.
I dont know if they will reply to my next question which is can they look at their logs and see a reason. I have given them the time it happened.
Stuart
Had another re-sync today and since then it has had zero error for both FEC and CRC, speed is good and SNRM seems very stable. Wonder if they've been messing with the exchange or my cabinet recently?
-
Eric I spoke too soon about having no errors. Unfortunately I was relying on the DSLStats error graphs and sadly this is not a good idea. The problem seems to be that if a re-sync happens the error graphs do not start again after it reconnects, I had to stop and start DSLStats in order for the errors to be plotted once again.
Stuart
-
I'll check that out, Stuart.
-
May I ask one thing, so far I've not seen an option to get to the new version screen while the program is running, if I'm right and there is not an option can you add an option for manually checking for a new version, and perhaps an option for checking every n hours or days? Would be useful for anyone running the program 24x7.
I've added an option for the user to specify how often the program checks for updates (between 1 and 14 days), and there's a button to do a check at any time.
soon as I run this new version (3.1) i had lots of strange behaviour on my computer, this is what happened.
the DSL Stats app opened straight to tray, i could not open it from the tray, i opened task manager and i could not kill the dslstats.exe *32 process, then all of a sudden it looked like it was logging me out as my screen blended to black (like does on shutdown) and it made the "disconnection" noise..the screen came back on and said a USB device had failed? also internet connected was lost but came back on, Task Manager was showing only half the stuff in task manager too..
I can't explain this behaviour, unless it was related to the check for an updated version - if the website was unavailable at the time, there was a long timeout before the program would appear, so the system tray icon would have been unresponsive. I've now reduced the timeout to 5 seconds, and also changed the way in which update checks are carried out.
but there are some problems: (im using Netgear DG834Nv1 with DGTeam 1018)
1) AS Uptime isnt reported in the bottom window
2) Nothing is logging in BitSwaps per tone (it is checked for monitoring)
3) The BitSwaps Per minute graph is showing zero (which I dont believe is correct)
1. Now corrected.
2. As discussed earlier, this graph is cumulative, so it may take some time for data to appear after starting the program.
3. I don't believe there was anything wrong here, but I'm prepared to be persuaded otherwise.
EDIT 2: It seems to have issues running alongside DMT Tool (which RS-W didnt) - ive been getting "no data in last response" and "no response from router" when DMT Tool was running only for few minutes.
As discussed earlier, I don't believe there were any changes in this version which would have affected this, so I'm assuming that this was just a random conflict at the time. If it continues to be the case, please tell me.
it would be great if it saved settings as they were made...or had SAVE/APPLY button somewhere? because if one shuts down windows without first exiting the program it wont remember their settings or save images etc... I dont recall having any problems with RS-W in this respect? it seemed to me that the settings were saved as amended? as i never shut down RS-W I just shut down windows.
The program now saves its settings when you click the green button.
also, I think it should start all graphs up in compressed view... expanded view should only be chosen by the user for a more refined inspection, the save to folder works great but any "expanded view" images (QLN, HLOG) dont save as a viewable image, it says INVALID IMAGE, when i change to compressed view and exit the program and re-check its ok..
I can't reproduce this problem, so I haven't made any changes in this area at present. If you prefer the graphs to start up in compressed form, then set them that way and the setting will be remembered.
EDIT: bug report - when changing colors for download...everything but the "bit swaps per minute" change color, they remain red...even if i change the 'other data' color to blue (like I did with download just messing about)
EDIT2: bug report - SNR Graph sometimes doesnt save correctly, well hasnt for last 3 saves, I have since disabled it in hope that it will be included together in the Bit-loading graph rather than seperate graphs...all other graphs are fine
1. Now fixed.
2. I can't reproduce this problem. The relevant code is the same as for the other graphs, so I'm at a bit of a loss.
My bitswaps per tone graph hasn't worked for around a month now, but I have around 50 Bitswaps per minute so I'm assuming this is something to do at my end?
Unless you're using one of those routers which don't refresh the bitloading data between re-syncs, I can't explain this. If it continues to be the case, I'd appreciate more information to help me get to the bottom of it.
Eric I spoke too soon about having no errors. Unfortunately I was relying on the DSLStats error graphs and sadly this is not a good idea. The problem seems to be that if a re-sync happens the error graphs do not start again after it reconnects, I had to stop and start DSLStats in order for the errors to be plotted once again.
I found the source of this, and I believe it's now fixed.
[Edit] Now confirmed - I had a re-sync earlier, and the error graphs continued to plot afterwards.
-
Eric any ideas when you might release this new version?
Stuart
-
Tomorrow hopefully, unless events conspire against that.
-
Hi Eric, yeah after some time the tone per graph worked, as for crashing etc iam thinking it was something to do with avast is not liking it or something because it works now... Although since then and while it's been running my system has crashed twice then failed to boot, early days but looks like it may be something to do with us joypad as since removing that it hasn't happened, hard to say what the exact cause was but it sounds like it was dslstats & avast caused a crash which has caused something else too happen.. But what I dunno... Watch this space lol..
At moment the only thing that doesn't work (that you haven't addressed) is the snr & QLN graphs won't save in expanded mode, I don't use expanded mode so not so bothered....
Thanks again for your effort, really appreciate it :-)
Sent from my Sony Xperia Miro on Tapatalk
-
Thanks for your comments, snadge.
At moment the only thing that doesn't work (that you haven't addressed) is the snr & QLN graphs won't save in expanded mode, I don't use expanded mode so not so bothered....
I will come back to this, but at present I can't reproduce it - those graphs save normally in either mode when I try them.
-
Thanks for your comments, snadge.
At moment the only thing that doesn't work (that you haven't addressed) is the snr & QLN graphs won't save in expanded mode, I don't use expanded mode so not so bothered....
I will come back to this, but at present I can't reproduce it - those graphs save normally in either mode when I try them.
it saves files with the names but they are 0kb in size and cant be opened - if you have yours in 'list view' then check them and see if the open or of a proper size.
edit: ok ive re-checked and when expanded it either saves just the bit that you can see (not the whole thing) or it fails with 0kb files, I just done some now and the HLOG one has nothing in it - also ive noticed that sometimes the HLOG and QLN graphs save as 0kb files even when not 'expanded' and ive also had 1 SNR one save like that too...
see attachments, first one is the HLOG expanded ive just got by closing the program (save on exit) as you can see there is nothing in the graph, and the rest are the 0kb files (not expanded)
[attachment deleted by admin]
-
...just to further add to my above post....
...does DSLstats use or modify anything to do with COM Surrogate or rpcrt4.dll ...?
-
it saves files with the names but they are 0kb in size and cant be opened - if you have yours in 'list view' then check them and see if the open or of a proper size.
edit: ok ive re-checked and when expanded it either saves just the bit that you can see (not the whole thing) or it fails with 0kb files, I just done some now and the HLOG one has nothing in it - also ive noticed that sometimes the HLOG and QLN graphs save as 0kb files even when not 'expanded' and ive also had 1 SNR one save like that too...
see attachments, first one is the HLOG expanded ive just got by closing the program (save on exit) as you can see there is nothing in the graph, and the rest are the 0kb files (not expanded)
I've just checked again, and the graphs are saved properly when I click the "Take snapshot" button, or when I set them to autosave, whether they are expanded or not. What I haven't checked (yet) is whether they always save correctly when set to autosave on exit. It's possible that your problem only affects the Windows version (I develop on Linux) and that's something else I have to check.
When the graphs are in expanded form, the snapshots only include the visible portion. I could in principle make them include the whole thing, but it would actually be rather complicated, and in the case of VDSL2 it would result in rather large images. I could put this on the todo list, but I can't promise to get around to it very soon.
Concerning the HLog graph which you saved on exit, did you enable the HLog graph just for testing purposes and then exit immediately? If so, this would explain its lack of content - nothing appears on the graph until the next sample after it's been enabled. The same will apply to the QLN and SNR graphs.
..just to further add to my above post....
...does DSLstats use or modify anything to do with COM Surrogate or rpcrt4.dll ...?
I'm afraid I don't even understand the question. I use a cross-platform high-level development system (Lazarus), do all the development on a Linux 64-bit machine, then build it for the various platforms. I don't get involved in the low-level implementation details. The information is certainly available in the Lazarus source, but wouldn't begin to know where to look.
-
I only take occasional snapshots on the windows version and haven't seen any problems.
The only comment I would make is that ideally the snapshot names could do with including the seconds in the name, otherwise you've got to wait a minute between consecutive snapshots.
-
The only comment I would make is that ideally the snapshot names could do with including the seconds in the name, otherwise you've got to wait a minute between consecutive snapshots.
Yes, I think you're right. The filenames used to include seconds several versions ago, but I left off the seconds after a discussion on the subject at the time. On reflection that probably wasn't the best decision, and I'll put the seconds back in. I'm just about to release v3.2, so this will have to wait until v3.3.
-
Eric I just found the new version and it's up and running ;)
Stuart
-
That was quick, Stuart. :)
-
It's definitely got stuff in the graph when saving, for me the snapshot button doesn't save images into the folder I have specified? When I click take snapshot nothing happens?? Can anyone else replicate this in windows 7 64bit?
The question about com surrogate was cos I've been having more problems and this com surrogate stopped working when I tried to browse a folder on my phone After I had to reinstall the driver again... I only started having these problems the very second I ran this new version of dslstats , it's lookin like I'm gunna have to format n reinstall windows because I think multiple problems have occurred as a consequence.
So you use lizards to generate a windows version but don't test it because you use Linux? Is this right?
If I reinstall then I will create a back up image before using dslstats again alongside avast is... Again just to see if I can replicate the problem
Sent from my Sony Xperia Miro on Tapatalk
-
So you use lizards to generate a windows version but don't test it because you use Linux? Is this right?
I use Lazarus on a Linux machine to develop the program. Then I transfer the source to a WinXP machine which also has Lazarus installed, to build the Windows version. I do some testing on the WinXP machine to make sure that the program runs and the main features work, but it doesn't get as much detailed testing as I do on the Linux development machine.