StokerLog Version 5 Preview

[Amir]

Your wish is my command

.

Please test this one before I make it official.

http://la.gg/upl/StokerlogPreview5A.msi

 

[Bob Hunter]

Installing now

 

[AndyS]

Amir,

It is readable now with the fixed font. The horizontal spacing is still off, but I am nitpicking or you have not gotten to that yet.

Here is a Screenshot

 

[Bob Hunter]

Amir, I have a couple of comments?
I am playing with your latest version, I like the look and the layout, I am not seeing any horizontal alignment issues on my end.

Where I am having trouble, is in regards to the email system. The email works great even without using gmail. I did notice that the AT&T precanned address is not correct - it has changed with the Cingular change - It should be @txt.att.net

The other thing is that I am not getting graphs with the email. I was testing sending an email every 2 minutes and the graph was supposed to be attached every 2. I tested this to my regular email not the sms on my cell phone.

Another nice feature would be email alarms, I know you had that setup working with the WebAlarm ( I never have understood what ot do with this), but since smtp has been implemented it would be great to get the alarms email that way as well. One thing to consider with this would be to have a settable dead band on the alarm emails, so that if the thing went in and out of alarm several times in x number of minutes you wouldn't flood the user with email.

Thanks for all your work on this.

 

[Amir]

Andy, looks like you are able to override the font for the label still. Not sure how this is possible as I am explicity setting it. Can you double-check that you installed the latest and greatest?

Bob, there is a bug in the graph control I am using which causes it to return a fatal error. It happens once in a while on my machine and so,I am catching and ignoring it. I can enable the reporting of this in the status line so that we can tell if it is the source of the graph not being sent. You may want to try running it on another machine to see if it works better there as it seems to be related to GDI issues.

On AT&T email address, I stole the list from someone who post it here

(Ken?). So I will go ahead and fix that one.

For Alarms, I had forgotten about that

. But it is pretty easy to wire up. How about having it send the same report it is now, plus the alarm indication string? Currently I have protection against alarms repeating so hopefully the logic will work just the same here. Otherwise, I am not looking forward to debugging it

.

Will post a new version soon.

 

[Bob Hunter]

How about having it send the same report it is now, plus the alarm indication string?

This sounds good.

Just setup the newest version on 5 different machines, not a single graph came through on any of the emails.

 

[Amir]

Oops. There was a bug in graph code

. It would only work if you had the Mail window open. I have fixed it plus wiring up the email alarms. I will pretty up the text later but for now, if you have enabled email logging, you will also get the alarms as they occur. Subject line is the name of the probe causing the alarms plus connection being lost.

http://la.gg/upl/StokerLogPreview5b.msi

 

[AndyS]

Amir, here it is with 5b

screenshot

 

[AndyS]

Amir,

I have tested the emailing section now. I can duplicate the not mailing the graph bug. It will occur if you set the email frequency to the same number as the graph frequency. If you set the two setting to different times then you will get the graph emails and the general emails.

 

[Amir]

Andy, if you are running the latest download, then I don't think the count makes a difference. Please see if you see the "GDI" red error messages in the status bar. If so, then that is the real issue.

I am trying to work around it but it is rather complex.

As to fonts, I have to take a deeper look at that. Thanks for the screen shot.

 

[Amir]

Not sure how I missed this but if you only have one pit being monitored, all the food alarms are wrong (shifted by one). So please don't rely on them if you are doing a cook. I am away from home so may not post a fix until Sunday night (have my laptop with me but the version here is a bit messy right now).

 

[Bill Freiberger]

Amir,

I'm doing a test run of version 5 with the Traeger controller. Two things so far.

The font for the current temperature is so big that only three digits show in the window. Right now the temp is 210 point something. I can't read read what's behind the decimal point. When it goes up to 211 point something I can see half of the number behind the decimal point (because the "ones" are narrower and take up less room in the window).

The other is that I had auto lid detection on but it only engaged the first time I opened the lid. The second time it didn't.

Bill

 

[Amir]

Thanks Bill. I can enlarge the box but it shifts everything over. I didn't think folks cared about the decimal point much

. But did fix the pop-up panel so that it shows up there.

On lid open, that is the same way it has worked before. The logic is too simplistic right now so it doesn't always trigger. I need to re-write it to make it better but need lots of help testing it once I do.

 

[Bill Freiberger]

Amir,

It's not so much that I care about the decimal place, it's that I can only half see it. I wonder if it would be better to drop everything past the decimal point for that window rather than have it half obscured. Not a big deal but a window into my particular OCD craziness.

The Open Lid setting worked the third time that I opened the cooker.

For what it's worth, I think that for the Traegers the five minutes setting that the Open Lid keeps the fan off after you open the lid may be too long. Unlike other cookers, the Traeger needs the Stoker to turn on the fan AND supply fuel to the fire. So, if the fan is off too long, you run the risk of burning up all the fuel in the fire pot. Then when the fan/auger kicks back in it will feed more pellets, but they'll never catch fire and burn. I'm using this feature as you have recommended, turning it on right before I open the cooker and then turning it off manually when I'm done. The only reason I bring it up is because I'm Beta testing the Traeger controller and want to document EVERYTHING I notice.

I'm really loving the email feature. I've sent updates to both my phone and my email. I'm sitting at home so I don't need to do either, but it's been fun playing with it.

Bill

 

[Amir]

I will go ahead and make the window bigger so that you we can see the decimal point. Reason is that I can sometimes tell if the fire is dying or not, based on whether the temp is rising ever so slightly.

On Lid Open, I will go ahead and add an Advanced option to set the time delay. I have also been thinking about a way to make this automatic. Not quite sure I know how to do that just yet, and not have it false trigger. If folks have ideas, let me know....

 

[Bill Freiberger]

Another thing I've noticed. When I set Lid Open and then open the lid, the button changes to Auto Lid Det and shows the amount of time in minutes before it resets the target temp. This second line is obscured by the Show Notes button so all I can see is the word Minutes and not the amount of minutes remaining.

Also, my wireless game adapter lost its connection with my LAN and therefore with StokerLog. When I reset it, the Log Time reset to 00:00. Is there a way to keep the old Log Time in those situations?

And finally (for now), when the program adds notes to the graph (Fire Out!, Lid Opened, etc.), they are actually on the graph and therefore difficult to read. When I add my own notes, they're located below the graph (in the timeline) and are easier to read. Is there any way to lower the program generated notes so they're in the white space below the graph? Or, better still, to make the background of the notes solid so they block out the graph information behind them?

Bill

 

[Amir]

Good catch there. Not only was the text coming out the wrong place, but there was a bug that would make the program crash when lid open occured. I am surprised it didn't do that on your system. I fixed both.

I reverted the text boxes to have solid white background again. And moved the box lower, close to the edge of the graph at the bottom.

On not reseting the time, there is a secret solution for that

in version 5. Hold the control key down when you click on start. When you do that, it will not reset the time. There is even a more secret option. If you hold the shift key down, it will also do this, but will skip over putting the stoker in debug mode. This will let the program attach to it faster, and eliminate one of the ways the stoker hangs when it is talking to stokerlog. However, to use this, you must have connected normally with the stoker once and not have power cycled it since. So in case of losing connection with the stoker, you could use the control key to get into it faster and more reliably.

 

[Amir]

So here is an updated version. I am not home so no way to test it properly. I don't think I have broken anything

. But no guarantee either.

Fixes are larger temp box to show decimal point, alarm bug mentioned above, lid open fix and moving the caption box as just mentioned. I also put in a fix to align the boxes better so hopefully this fixes the high-dpi thing.

I also took a shot at fixing the issue with graph not being included in email. Unfortunatley, on my laptop it never failed so I don't know if I have fixed anything. So appreciate someone testing the mail functionality.

http://la.gg/upl/StokerLogPreview5C.msi

 

[Patrick S.]

I'm just finishing up an all night cook with this new version.
Did a butt on top of a brisket.

Things went very well overall. This release is much more stable than the previous one. I never had an issue with having to reboot the stoker because the connection was lost or the stokerlog was not updating like I had before with the older releases.

Attached is a pic of the cook:
Stoker Log

There are 2 major issues on this graph. The dip right before 6:00am was the fire almost dying out because the fuel was running low. I replenished it when the alarm went off & all was well.

But the bigger problem happened around 21:30. For whatever reason, the fan just stopped running on the cooker. I saw the temp dropping & dropping but no fan was blowing at all. I had to go outside & unplug/replug in the fan & then it started up again. I have no idea if this was an issue with the app, or another stoker bug.


As you can see, I also had some problems with the temp control when the lid was picked up. It seemed to not stabilize great. I t worked but just not super great. I think this is acceptable.

The good news is that from 11:00pm - 6:00am (when the coals wen t out), it was steady as a rock & I got a full night's sleep. Just goes to show how smooth it can be if you don't play with the lid at all.

One suggestion: I like the addition of logs via SMTP. Can you make it so alerts go through the SMTP server also? Instead of having to use the alerting URL?

Thanks again for a great app. This meat will probably be ready to pull in 2 hours or so, I'll let you know how it goes.

 

[Amir]

Originally posted by Patrick S.:
I'm just finishing up an all night cook with this new version.
Did a butt on top of a brisket.

Things went very well overall. This release is much more stable than the previous one. I never had an issue with having to reboot the stoker because the connection was lost or the stokerlog was not updating like I had before with the older releases.

I have made one change in the way I read the stoker which could cause the program to hang sometimes although I can't see it causing instability. Otherwise, I think it was luck that it ran better

.

But the bigger problem happened around 21:30. For whatever reason, the fan just stopped running on the cooker. I saw the temp dropping & dropping but no fan was blowing at all. I had to go outside & unplug/replug in the fan & then it started up again. I have no idea if this was an issue with the app, or another stoker bug.

This is a bug that some of us are suspecting is in the stoker. I have only seen it occasionally. The fix is to just change the target temp and it keeps going. I think it has something to do with the fire being stable for a long time and not needing intervention. And then something happens and the stoker thinks it is just an anamoly which will go away.

Here is another secret new feature in version 5

. I log ALL messages from the stoker now in a file. The name of it is: telnet_input.txt. Can you please do a search on your machine and find it? Then copy it to another file name because stoker keeps appending to this file for every cook. Then email it to me if you don't mind and I will take a look to see if I can see anything strange in it. Email is amir.majidimehr at gmail.com. I am hoping this will give enough clues to the developer to try to figure out if this is a bug in the stoker or not. Unfortunately I just realized there are no timestamps in it so it may be tedius work to do this.

As you can see, I also had some problems with the temp control when the lid was picked up. It seemed to not stabilize great. I t worked but just not super great. I think this is acceptable.

This partly happens if you don't disable the stoker. Check the "lid open" option in stokerlog. If it catches this event, then it will shut down the fan so that you don't get the extreme overshoot although it still happens to some extent.

One suggestion: I like the addition of logs via SMTP. Can you make it so alerts go through the SMTP server also? Instead of having to use the alerting URL?

This is now a standard feature. Go to to Settings->Mail and configure it. Then check the "Enable Email Logging" and you are good to go. You will get regular email logs, potentially with the graph attached and also any alarms cause unique emails to be sent with that message.

Thanks again for a great app. This meat will probably be ready to pull in 2 hours or so, I'll let you know how it goes.

My pleasure.