Traceability on SMS

[Roman Novgorodov]

Hello

Unfortunately  we ca not reproduce trouble on our hardware.
There are not timeouts or empty Ref IDs. :-(
We will try again with other devices.

Roman Novgorodov
DeepSoftware.Com

[henry64]

The timeout issue is gone after I stopped trying to delete the message that wasn't stored, although I dont understand why that should give a timeout instead of an "error".
Maybe that timeout actually messed up the return of correct refId?

Regarding the missing refid:

I have started a new project from scratch and now it seems quite "stable" and I get a RefId back.  8)
I have noticed that the SmsSendWaitRefID doesn't return before the message has been sent whilest the "sms sent" and even "sms received" events are fired before return of SmsSendWaitRefID.
So it seems that I won't need that "sent event" since SmsSendWaitRefID indicated the same if it returns a RefId.

So I'll get back to you within a few days if the problem re-emerges. Okay?

[Roman Novgorodov]

Hello

You can try to change UseMainThread property or baud rate of connection.

We will try to reproduce problem on our side.

Roman Novgorodov
DeepSoftware

[henry64]

Hi Roman

I'm expirencing some troubles getting the RefId back to my application, although in the log it seems that the RefId exists on a lower level but just doesn't get consistently
returned in RefId.

I have attached a nrcommlib.log where three messages (98,99 &100) are being sent and reports are being received, however at application level RefId's are missing, see the adsmsdispatch. log

Look for the "Got RefId: 98" after "sending..."

RefID := nrGsm.SmsSendWaitRefID(aPDU);
LogEvent(lcDispatchEvents, format('Got RefId: %s',[RefId]));


It's not consistenly missing nor in what order RefId's are missing

[henry64]

Here's another example where I send 3 messages and the nrcommlig.log show no error and obviously retrieves the RefId allright. In this case RefId 108 is not returned in RefId.

It seems (but is only a guess) that when setting Report := True and if a report sms is received while waiting for SmsSendWaitRefID then it will mess up something .



2011-05-29 16:20:41 [Dispatch event    ] Sending SMS PhoneNo:+4520102200 Message:Repeated aeoa AEOA DTU-System-AlarmDirecor-Server/Alarmpunkt (TEST)
2011-05-29 16:20:45 [Dispatch event    ] SMS Sent 4520102200 Repeated aeoa AEOA DTU-System-AlarmDirecor-Server/Alarmpunkt (TEST) id:  107
2011-05-29 16:20:51 [Dispatch event    ] Got RefId: 107
2011-05-29 16:20:51 [Dispatch event    ] Sending SMS PhoneNo:+4529804991 Message:Repeated aeoa AEOA DTU-System-AlarmDirecor-Server/Alarmpunkt (TEST)
2011-05-29 16:20:56 [Dispatch event    ] SMS Sent 4529804991 Repeated aeoa AEOA DTU-System-AlarmDirecor-Server/Alarmpunkt (TEST) id:  108
2011-05-29 16:20:56 [Dispatch event    ] SMS Report 4520102200 REPORT: Sms is Delivered to 4520102200 at 2011-05-29 16:20:49 id: 107,
2011-05-29 16:21:01 [Dispatch event    ] Got RefId:
2011-05-29 16:21:01 [Dispatch event    ] Sending SMS PhoneNo:+4529802359 Message:Repeated aeoa AEOA DTU-System-AlarmDirecor-Server/Alarmpunkt (TEST)
2011-05-29 16:21:05 [Dispatch event    ] SMS Sent 4529802359 Repeated aeoa AEOA DTU-System-AlarmDirecor-Server/Alarmpunkt (TEST) id:  109
2011-05-29 16:21:08 [Thread og Mutex   ] THeartbeatThr.Execute hTimer received
2011-05-29 16:21:08 [Thread og Mutex   ] Heartbeat started
2011-05-29 16:21:08 [Thread og Mutex   ] Heartbeat done
2011-05-29 16:21:11 [Dispatch event    ] Got RefId: 109

[henry64]

I'm using nsmSkipSave in the above approach, I have now tried another approach using nsmNotifyOnly - the idea is to let the device store the incomming messages and then when completed the "sending" messages, then read from device memory.

Now, it appears that there is a problem in TnrPduSms.DecodePDUDateTime when reading smslist, occurs in the gsmdemo!!

'000000FF0006850A9154020122000180911244508001809112'


I'm stuck here  :o

[Roman Novgorodov]

Hello

Thanks for LOG files.

It seems like problem is following:
After SMS sending TnrGsm waits RefID, but new SMS (delivery report in your case) is coming first.

We will try to resolve this problem.

Roman Novgorodov
DeepSoftware LLC

[henry64]

Excellent  8)

I have moved the "gsmdemo" issue into a new thread so we can keep this thread strictly to the subject.

[Roman Novgorodov]

Hello

I have sent you email message with download link for test.
Please post  here the result of testing.

Roman Novgorodov
DeepSoftware LLC

[henry64]

Please send me the source code package, I need to change a few things in order to prevent BSOD (setcommask) must be replaced as I have showen ealier   ::)

[henry64]

I've testet a bit and the refid is still missing, here's log from the gsmdemo - from what I can see in the log, refid 176 is actually returned somewhere internally, but is not returned by the SmsSendWaitRefID method.

[Roman Novgorodov]

Hello

Thank you for your information.

It's strange. We reproduced problem on our Huawei and fixed it.
We have no wavecom device for deep test.
I have sent new compiled GSMDemo on your email some minutes ago.

Roma Novgorodov
DeepSoftware LLC

[Roman Novgorodov]

Hello

Please check your email.
I have sent new build with new changes.

Roman Novgorodov
DeepSoftware LLC