Sensors times defaulting back to around May 20th on HS3 restart.

[cerjzc]

Starting around version 103 or 104 of BlRadar it seems that after a HS3 is restarted my sensor times revert back to around May 20th which is close to when I updated to version .103 BlRadar. I checked the hspi_BLRadar.ini and the dates and times appear to be stored there. So I turned on debugging for BLRadar and shut down HS3 and it shows in the debug log it is saving the updates to the configuration file, but when I check the file it still has the dates from back in May. So I located an old version of BLRadar which was version .98 and installed it and started HS3. I then updated a sensor that I had been using to test and shutdown HS3 and sure enough the hspi_BLRadar.ini was updated with the current date of June 7th. I then restarted HS3 while on .98 of BLRadar and it now shows the correct time for the sensor that it was last triggered a few minutes ago instead of showing around 17 days.

I was running BLRadar .106 which is the current version and HS3 version .181. I'm going to leave .98 of BLRadar in place until Blade has a chance to look at it. If I recall this is around the time he came out with the version or at least some updates for the linux BLRadar.

Thanks,
Jeff

[Blade]

Log a ticket on my web site for this please

[cerjzc]

Blade,

I'm still seeing this issue on anything above around .98 of BlRadar. I had stuck with .98 since logging a ticket about this, but recently saw you put out a few updates so I tried them, but found after I cycle HS all of the blradar devices reset their time back to it appears when I updated to anything above .98. So after an HS cycle I will have a bunch of devices that either show they have dead batteries or that show they haven't triggered for days or weeks. Is it possible to have it update whatever file that stores this information at a certain update interval so it will be fairly up to-date on an HS cycle or crash? I'm guessing it was getting updated for me on the earlier versions, but somewhere around .99 or so it stopped.

I checked with a couple people I work with that run HS and blradar and they confirmed they are seeing the same issue with their setup. The ticket I opened on your helpdesk is 793 and like I said I'm not sure that .98 was the last working version it just happened to be the one I had a backup copy of when I first noticed the problem around version .106.

Thanks,
Jeff

[Blade]

Can you confirm the issue is still happening in the latest in the updater.
If it is please capture me a new debug log so I have an updated one and attach it to your ticket zipped

[cerjzc]

I had tested with version .109, but not .110. I'm trying to duplicate the problem now with .110 and so far I haven't been able to. Was there a change from .109 to .110 that might have corrected this? I normally don't notice it for sure until a day or two later when I go to cycle HS and the sensor times all get reset to when I updated to above .98, but currently I have cycled multiple times on .110 and the times appear to be correct/current.

I will keep testing and see what happens.

Jeff

[cerjzc]

Well another update. I left things running on .110 and while it appeared that the times were correct right after the HS restart now devices that had been updated just before the most recent shutdown have now reverted to previous times. I had a few devices that showed 16 days before and I caused all of them to trigger and they were up todate, but now these devices have reverted from their current time back to 16 days ago or for other devices they have gone back to when I upgraded from .98 to .110 which is around 1 hour and 10 minutes ago. Of course I had turned off debugging when it looked like it had worked. Let me turn it back on and see if I can capture another log.

Jeff

[Blade]

There is very strange. I have not seen this happening and I really rely on BLRadar the most in my house. See if you can capture a debug showing the issue and I am sure we can nail this down and get it fixed. It is tough to find these type of issues.

[cerjzc]

Bob I did send you the logs I had, but of course it might have missed exactly when the problem happened. I am back on .98 currently, but when I get time in the next couple days I will give this another test and try to track down the exact time it changes and get the logs for it. Can you maybe tell me how the information is stored/saved in your config files for the last echo/no echo? Does it only get saved to the config file on shutdown or is there some sort of update interval?

Thanks,
Jeff

[Blade]

I will have to check for sure but I believe it gets written to the ini file on shutdown of the plugin. I will validate this weekend and see if anything jumps out at me

[zgebones]

Blade,

I reproduced the issue and attached debugging to ticket # 848 after upgrading from .98 to .110 tonight.

[Blade]

OK I will have a look as soon as I can.