Page 3 of 15 FirstFirst 1234567813 ... LastLast
Results 41 to 60 of 286
  1. #41  
    By the way, I installed a new version of Chatter this morning, which meant a soft reset.

    I could believe there's a memory leak in the OS somewhere.

    Tom
  2.    #42  
    Quote Originally Posted by shanta
    I just did another soft rest and still no changes with network time enabled and the application enabled.
    I will look into this but you can still verify the problem by simply maunally powering up your phone. There is no magic about the startup process.

    Jeff
  3.    #43  
    Quote Originally Posted by shanta
    Interesting...I installed this app and when I power off the phone's radio and then power it back on it shows 2 changes, the first one being a date in 1988 and the second one being the correct day time.

    When I do a soft reset and the phone's radio get's powered back it it shows NO changes. Since that soft rest I powered the phones radio on/off 3 times and still no values. Not sure why it registered the first time in powered the radio on/ff but after the soft reset it logs no changes.
    I think it may just be a race condition during startup. IE. The phone gets started before my app gets initialized.
  4. #44  
    There is definitely a memory leak in the OS. Resets happen more often, the less memory you have free.

    GD, Palm...when will you get it right on a single threaded OS?
    And we are supposed to trust Cobalt?
  5.    #45  
    Quote Originally Posted by VetPsychWars
    Jeff, et alia:

    I do not have a custom ROM, and I do have the Sprint roaming patch. I can also say that I was reprovisioned this morning by Sprint. Don't know what they downloaded to me. but they did.

    Ever since I stopped using Versamail, everything has been solid, though the occasional site that delivers a virus load or malevolent javascript can cause problems.

    I'll do reasonable tests, but so far the program has reported zero time resets.

    Jeff, if the time is garbage, I would more likely think of an uninitialized variable or memory that was allocated but not cleared prior to a write?

    Tom
    Its hard to tell where they are getting their time values from. You may be right about the variables. The thing that I do not understand is why they are not checking to see what the current time is and only change the time if the palm time and/or timezone does not match the network time. What they are doing is just wrong.

    Jeff
  6. #46  
    to get your app to consistently report the time changes, I had to enable then diasble your app before bouncing (power off/on) the phone radio.
  7. #47  
    Ah, there we go.

    OK, I see a change. According to the last change, it's 2013?

    Something might be going on. Is it nefarious? Don't know.

    You are using the proper API and using strong typing? :-)

    Tom
  8. #48  
    Quote Originally Posted by scottymomo
    No time changes on Cingular Blue as expected as they don't push network time
    I don't think you are right at all scotty... I should know the details, but I don't. All I know is last two times I have traveled with my Nokia phone when I dot off the plane in a different time zone and powered up, the time adjusted right away. I imagine the info is on the BCCH.
    Treo 650 GSM
  9.    #49  
    Quote Originally Posted by VetPsychWars
    Ah, there we go.

    OK, I see a change. According to the last change, it's 2013?

    Something might be going on. Is it nefarious? Don't know.

    You are using the proper API and using strong typing? :-)

    Tom
    ?????
  10.    #50  
    I found the problem with the startup situation after reset. This version will now work properly. I threw this thing together so quickly I forgot to initialize a variable that needed to be initialized just after reset.

    Jeff
    Attached Files Attached Files
  11. #51  
    I just tried triggering this by turning the phone off (radio off) and then back on. I checked the date/time between each attempt.

    Still no problem on my Treo 650. I wonder if it is a software issue that is conflicting with the radio being turned off/on.
  12.    #52  
    Quote Originally Posted by TonyKInTexas
    I just tried triggering this by turning the phone off (radio off) and then back on. I checked the date/time between each attempt.

    Still no problem on my Treo 650. I wonder if it is a software issue that is conflicting with the radio being turned off/on.
    This is interesting. A couple of questions.

    Do you have Sprint Service?
    Do you have "Enable Local Network Time" option enabled in date&time preferences?
    Do you have the Shadowmite ROM loaded for voice quality?
    Do you have the newest Sprint Roaming Patch loaded?

    Jeff
  13. #53  
    Quote Originally Posted by jeffgibson
    This is interesting. A couple of questions.

    Do you have Sprint Service?
    Do you have "Enable Local Network Time" option enabled in date&time preferences?
    Do you have the Shadowmite ROM loaded for voice quality?
    Do you have the newest Sprint Roaming Patch loaded?

    Jeff

    I have each of the items you listed and with the "Enable Local Network Time" option enabled, your program shows 2 time changes.
  14. #54  
    Sorry, Jeff. :-)

    One of the reasons I decided to be a writer instead of a programmer was that I had no desire to maintain someone else's sloppy code. My little inside joke. I'm used to the bad old days of Unix system programming where an int and a char were the same length, so why bother using the right type?

    I can say, though, that I have never had a spontaneous reset for no apparent reason, and that your program reported the last date change as 2/9/13.

    Whatever that might mean.

    Tom
  15.    #55  
    Quote Originally Posted by VetPsychWars
    Sorry, Jeff. :-)

    One of the reasons I decided to be a writer instead of a programmer was that I had no desire to maintain someone else's sloppy code. My little inside joke. I'm used to the bad old days of Unix system programming where an int and a char were the same length, so why bother using the right type?

    I can say, though, that I have never had a spontaneous reset for no apparent reason, and that your program reported the last date change as 2/9/13.

    Whatever that might mean.

    Tom
    Well you did catch me! I didn't apply the right storage mask for the function call I was using (it wanted a UInt16 and I was giving it a UInt8). That has been fixed in the 1.1 release. This was a display problem only. The resutls that the tool is capturing is correct.
  16. #56  
    Quote Originally Posted by cglaguna
    It is on, but I have been with Cingular for years and never had time pushed before. Plus I manually set the time on mine, did other GSMer's not?
    Same here, cg.
    BTW, does anybody really know what time it is?
    You may be right; I may be crazy. But, the Treo may be just the device I've been looking for.
  17.    #57  
    Quote Originally Posted by JohnH59
    I have each of the items you listed and with the "Enable Local Network Time" option enabled, your program shows 2 time changes.
    That is also my configuration. I was hoping that someone that has *not* loaded Shadowmite's ROM or the Roaming patch could gove this a try. I would like to know if it either of those introduced this problem.

    I have done a hard reset on my 650 and then run this program and a barebones 650 with ROMupdate and roaming patch shows the problem.

    BTW: I can tell you that the times that I lost my saved preferences database and all of my license keys was due to this problem. I just happened to me tonight as I was testing another application. Things are getting a little clearer. The *good* news is that by disabling this time option, you can improve the overall reliability and stability of your 650.

    Jeff
  18. #58  
    The time changes are not real logical from Palm's side. If the Initial Time is okay, why would they have it go to1/5/80 17:59? That does not seem logical at all. If it is okay, within a certain degree, then there should be no changes...
  19. #59  
    How are they keeping time, anyways? Seconds since the epoch? Something else?

    While I trust Jeff knows what he's doing, it wouldn't be the first time that a system call was arbitrarily changed. Also I wonder whether this is mapped to a memory location and whether the 512-byte block comes into play.

    Tom
  20.    #60  
    Quote Originally Posted by VetPsychWars
    How are they keeping time, anyways? Seconds since the epoch? Something else?

    While I trust Jeff knows what he's doing, it wouldn't be the first time that a system call was arbitrarily changed. Also I wonder whether this is mapped to a memory location and whether the 512-byte block comes into play.

    Tom
    Yes the implementation is like Unix. The number of seconds since 12:00 A.M. on January 1, 1904. From that value you do the calcualtion to get the current time.

    Well I think the reason for the problem is hard to pin down. The real issue is why are they setting the time 2 times? The time value being used for the time change is obviously a miscoded statement in the code. Perhaps what is happening is that the code was written to check the date after the time was changed and if it sees a date that is obviously wrong it tries to correct itself. Who knows...... Just like I made the mistake of not masking my display of the date correctly, I am sure a similar thing was done by the P1 developer. The problem here is that we have no control or communication with anyone that can code the fix. P1 says that the 650 is Sprint's responsibility and Sprint appears to not be taking ownership. The problems are just going into a black hole from what the outside world can tell. No one from Sprint or P1 is communicating *anything* but P1 saying - not my problem.

    Jeff
Page 3 of 15 FirstFirst 1234567813 ... LastLast

Posting Permissions