Print

Please login or register.
1 Hour 1 Day 1 Week 1 Month Forever
Login with username, password and session length

[roseway]

Thanks again to all.

@oldjim: The location of the config files was changed some time ago in response to user requests. Because there wasn't universal agreement I included the option to choose one of two locations, both of which should be accessible to the user. I think I'm going to leave it that way.

@NewtronStar: I think your second experience may be a red herring. If an MDWS upload is delayed for any reason, such as a temporary network issue, the GUI freezes and doesn't redraw until the temporary blockage clears. I think it's most likely that this is what happened to you on that occasion.

@tbailey2: The occasional divide by zero error has been bugging me for a long time, as I've never been able to reproduce it and haven't been able to locate a cause. The best suggestion so far seems to be to set the error averages to the 24 hour setting, not the modem uptime setting, and this seems to have fixed (or at least alleviated) the issue.

[Oldjim]

Your decision except that as I posted the default isn't readily available to the user who is at Windows default settings as it is hidden

[simmotech]

I have just had the Division by Zero and again immediately after a restart - didn't get that at all in the previous version.

Also the Errors/Averages panel is not quite tall enough - shows a scrollbar.

I've followed your suggestion and changed to 24 hour averages which it managed to save even in a corrupt state and restarted - seems OK for now.

[tbailey2]

@tbailey2: The occasional divide by zero error has been bugging me for a long time, as I've never been able to reproduce it and haven't been able to locate a cause. The best suggestion so far seems to be to set the error averages to the 24 hour setting, not the modem uptime setting, and this seems to have fixed (or at least alleviated) the issue.

I'm not monitoring them....

[ktz392837]

@tbailey2: The occasional divide by zero error has been bugging me for a long time, as I've never been able to reproduce it and haven't been able to locate a cause. The best suggestion so far seems to be to set the error averages to the 24 hour setting, not the modem uptime setting, and this seems to have fixed (or at least alleviated) the issue.

Perhaps a pain but could you wrap all divide operations in a function called nodivideerror and check for -/+ve near zero and change to 1 to stop error.  Then expand with logging/exceptions to track down the line causing the issue?

[roseway]

Thanks for the suggestion, but I've already been through my own code line by line, and checked every instance of a division. In every case it's either not possible for there to be a zero divisor, or where there's doubt I've wrapped it in an error trap. So although something in my own code may be the cause of the problem, the error is actually happening in a library function. Chasing that by hand would take forever.

I do have some ideas for more investigation, but I'm rather hampered by a personal issue which is limiting my time at the PC. I'll do my best.

[NewtronStar]

V5.7.1 looks like it has settled down on the RPi and still uploading to MDWS and I'm very grateful for your ongoing support into DSLstats. was starting to think the only version that would worked on the RPi was V5.6 and previous versions 

and yep the RPi modem/router uptime averages was selected now changed it back to 24 hour averages.

[roseway]

Thanks for that, NS.

[Starman]

Upgraded too myself on my RPI - I'll let you know if there are any problems

[NewtronStar]

Upgraded too myself on my RPI - I'll let you know if there are any problems

Just make sure you have selected 24 hour averages in the Errors/Averages !

[tbailey2]

Just had a look at the server and there's another divide by zero alert up. Could have been there all day as haven't checked today. Still uploading though, I'll leave it there for now. As I said, I don't have the error box checked, only SNR and Connection speed in that section.

[MikeZ]

My DSLstats fell over at 11pm last night with a Divide By Zero error. I've been unable to restart it because the error comes up within a few seconds.

I've gone back to the previous version and restored my AppData/Local/dslstats directory from yesterday's backup.

[adrianw]

I recently upgraded to dslstats32W-5.7.1
Initially all was well.
Yesterday evening I started getting division by zero errors.
I have automatic start recording set so was unable to make any configuration checks or changes before the dialog box came up.

I have reverted to dslstats32W-5.6, which seems to be working.
I do have 24 hour averages set in Errors / Averages.

A possible clue.

My HG612 has been up for 51 days, with a lot of traffic going through it.

The dslstats32W-5.6 event log contains
27 Feb 2016 06:07:02   Configuration files stored in C:\Users\aw1\AppData\Local\dslstats\
27 Feb 2016 06:07:02   The configured modem/router is Huawei HG612
27 Feb 2016 06:07:02   Snapshot folder is C:\Users\aw1\AppData\Local\dslstats\
27 Feb 2016 06:07:05   Recording started
27 Feb 2016 06:07:09   Implausible up traffic value reported by modem/router
 
The Telnet Data display contains (recently)
ptm1      Link encap:Ethernet  HWaddr F4:55:9C:77:96:6C 
          UP BROADCAST RUNNING MULTICAST  MTU:1500  Metric:1
          RX packets:668639566 errors:0 dropped:0 overruns:0 frame:0
          TX packets:1212938853 errors:0 dropped:258 overruns:0 carrier:0
          collisions:0 txqueuelen:1000
          RX bytes:4294967295 (3.9 GiB)  TX bytes:2132956719 (1.9 GiB)

Perhaps a big number / wrap around related problem?

[kitzuser87430]

R Pi new version up and running OK from 8 PM last night.

[licquorice]

I've had a couple of Invalid Floating Point errors on my Rpi since upgrading to v5.7.1