Upon receipt of the SMS from the wireless cloud, this iSMS server will try 3 times in a row to POST the message to the server. If the POST attempts are unsuccessful, the message will be placed into the Receive API Failure log and is not be re-sent / re-queued. The message is also placed in the Inbox and Received SMS trace log. The originator will have to resend once the net server is back up or the net server administrator will have to manually review the iSMS server to learn what messages came in while the net server was down.
The Polling method of Receive API was implemented to address this concern.
Sincerely
Jeff
How does the iSMS deal with a recipient net server failure used for the NON-Polling API solution?
Is the SMS stored and re-queued?
In event of the recipient server being unavailable before the iSMS reaches memory capacity will the iSMS stop receiving untill it can deliver the SMS?
So this was easier than I thought. I didn’t know that when the receive API was enabled it just forwards all messages to the server listed. I was searching for the mechanism that would make the message be sent to the application. I’ll post up a sample in a couple days… It’s going well though.
Thanks,
Ed
Brian,
Are you saying this is a new feature with the SF devices we already have?
I am lookng in the API doc and do not see any reference. This sounds like an interesting feature
Randy
After I updated to 1.47 I had to restart the modem and wait about 10 minutes for the SIM to show as active.
I’m not exactly sure what you’re trying to do, but besides doing some trigger features, to run a command after a SMS you would need a program running on a PC or webserver that would use the iSMS API to receive incoming SMS, interpret the SMS and execute your command on that PC/webserver, and then send a SMS with the results using the API.
For more detailed information we would need to know more about the programming language being used and the setup.
There are some iSMS API programming examples that may be helpful on our FTP site at https://webfiles.multitech.com/engineering/sample-code/sms-finder/
For non-development problems, it’s best to create a case in our tech support portal at support.multitech.com.
Hi,
Did you review the receive API?
http://www.multitech.net/developer/products/isms/isms-receive-api/
Do you have an example of what you tried to do (sample code)? What URI did you post to?
Does anyone have example code using PHP to send a txt message to the SMS server ?
We are getting the following response :
<HTML><HEAD><TITLE>501 Not Implemented</TITLE></HEAD>
<BODY><H1>501 Not Implemented</H1>
POST to non-script is not supported in Boa.
</BODY></HTML>
Hi Kirk,
We’re sorry you didn’t get a response right away through this forum. I see that you had to open a support case through our WEB portal and were then able to close out the issue.
I just wanted to post your findings to this thread. Please Note: We added that <Response> element to our XML data structure in iSMS version 1.44. I’ll mention to the Engineer of the C# example code to update the example.
Thanks
Jeff Blees
Engineering Technician
Kirk Wrote:
Figured it out. The iSms was doing everything as configured. On my default.aspx page I had to set ValidateRequest=”false” otherwise IIS thought it was malicious code being sent and rejected the post.
Also, in the C# code sample MultiTech has this line
XmlNodeList xmlNodeList = xmlDoc.SelectNodes(“/MessageNotification”);
I couldn’t get that to work so I changed it to
XmlNodeList xmlNodeList = xmlDoc.SelectNodes(“/Response/MessageNotification”);
I have the following settings and I can’t get the iSms to send out the receive Sms. Is there another setting I’m missing? I did save and restart the iSms.
Non Polling Receive API Status = checked
Protocol = HTTP
Server = 192.168.1.3
Port = 80
iSms IP = 192.168.1.20
Server Default Page = Default.aspx
Post Interval = 0
I tried both with Server Authentication and without. My test server doesn’t require Authentication on that website
Maybe if was Jeff Kline. I new the Jeff part was right. I’m not having any problems. It’s working great during my testing. I just want to make sure I didn’t have to take action when it sends me the log. Is the apimsgid generation tied to the device or to the port? If I send sms out port 1 and it gets apimsgid = 1, then I send sms out port 2 (both in the same device) will it get apimsgid = 2 ? Also do I have to specify a port number when I perform a querymsg?
Hi Kirk,
I’m sorry but I can’t find a record of us working together before (maybe you were working with Jeff Kline – you have some cases in our Portal with him).
Neither your initial comments or my reply, in today’s thread, say anything about the iSMS not working when a log file becomes full. Are you having a failure?
Emailing the logs is part of a maintenance function. If you do not want to receive the periodic emails – “delete” the “Log File Full” item in the “Send Email Notification for” column in the Administration menu. Then when the log file becomes full, it gets renamed and backed up locally in the /var/log/ folder on the iSMS server. Then after 3 cycles of it being backed up/renamed – it gets deleted.
I’m not sure what all the variables are to the msgid number that is generated by our send API.
I know if you default the box it will start over.
I goes up to at least 99999.
A system reboot/restart does not cause the number to start over.
If your APP uses (submits to more than one) iSMS server, then Yes, you will have to track separately. The msgid generated by each iSMS server is independent of the other server.
There is an alternative to this. You can utilize our “Load Balancing” feature – however it’ really not a balancing feature. It’s more of load sharing when I get too busy feature. The current online User Guide is quite accurate with it’s description of the feature. You only have to submit to one unit, but you have no control over which unit is used to send out the message.
I’ll look into the msgid rollover question more and get back to you.
Sincerely
Jeff Blees
Engineering Technician
Got an 8-port and I did go in and disable all but the first one, just like you originally told me when I bought it. The log comes once a week. I didn’t know it that would prevent the modem from working if the log filled up. Numbers less than ten make sense…I forgot about Short Codes. My app checks length anyway. What about the apimsgid the modem assigns, does this ever start over? What if I have two devices? Will I need to track per device then ID?
Hi,
I search an product to receive 6 SMS/Second .
Did you know if MULTI-TECH product in France could receive 6 SMS / Second ?
Could i use JAVA SMSLIB API on LINUX Server ?
Regards
Philippe
Hello,
I’ve just updated, I have 1.31 now I have 1.47 firmware.
Sometimes the API takes too long to detect the connection, and I do not have the signal LEDs on, and in admin web page I have “Disabled – SIM Inactive” in SIM Status.
But if I go to telnet and type AT+CSQ it replay +CSQ: 22,0.
Whats wrong?
Thanks,
JS
Hello Yemane,
We do not test with Kannel. Regardless of which Software & or platform you are trying to use, all have to communicate the same way with the iSMS SMS Servers (via API based commands).
Here is a link to various pieces of sample code.
https://webfiles.multitech.com/engineering/sample-code/sms-finder/
Our API commands start on page 74 of the User Guide.
http://www.multitech.com/en_US/DOCUMENTS/Collateral/manuals/S000461B.pdf
Sincerely
Jeff Blees
Engineering Technician
Hello Alex,
The “inbox” (per port) is essentially just a log of what messages the particular port (modem with sim card) has received from the cellular network.
The contents of each inbox does not reflect if the SMS/text messages listed (in each inbox) was delivered or not to your host application.
The “Receive API query commands” do Not implement/need/use a modem number specific option (“filter” option).
The “Send API commands” that we support/use Does include a modem specific option so you can pick which sim/account you want to send out SMS messages through. Maybe you are confusing this send function with our receive function?
Even though the WEB menus display an “inbox” per port – all SMS messages that come from the mobile network, regardless of what modem port they came in on, are held in our “receive queue”. Depending on how you set up our Receive API – you either query our pool to get the sms or we automatically post the message to your host application. The response we give/post contains (among other things) which port the SMS was received on.
How do you have the “Non Polling Receive API Configuration” section, of the “SMS API” menu, configured?
Sincerely
Jeff Blees
Engineering Technician
Hello Raza,
The iSMS is compatible with any software/application that can utilize/conform to the specific API commands implemented by the iSMS SMS Server.
Please review our API Appendix starting on page 74.
http://www.multitech.com/en_US/DOCUMENTS/Collateral/manuals/S000461B.pdf Check
SMSlib needs to submit messages to us, via HTTP, following our documented syntax.
Sincerely
Jeff Blees
Engineering Technician
Hi David,
If you have only 1 SIM in the SF400 – are the other 3 ports disabled? The “Status” check box for each port without a SIM card needs to be unchecked (this is found in the “cellular modem” setup page within the Network Setup menu).
If you have not already, please open a case in our support portal. http://www.multitech.com/en_US/SUPPORT/
The iSMS does not have a setting that dis-allows the sending of duplicate messages. I know from testing it can send duplicate messages over and over.
What is the contents of the HTTP Send API Live Log (it’s a log of what you submit to us) and what is the contents of the “SMS Live Log” (a per port log of how our SMS server controls/utilizes the modem)?
Even though both are T-Mobile SIMs, the problem could still be with one SIM? Have you tried swapping the SIM cards around?
What version of firmware is the SF400 running?
Is the failure all the time – or does it follow the contents of the SMS?
Can you send a simple short sms (twice in a row) that does not contain any blank lines like “Hi, this is a test.”
Can you send a simple sms via our Send SMS GUI menu, twice in a row?
Sincerely
Jeff Blees
Engineering Technician
Hi Brian,
There has not been any decision on adding this. Today I asked if Engineering could give any additional details.
My simple understanding of how the iSMS product currently works (current API commands, GUI menu choices and how our cellular modem is currently communicating with Telco) suggests to me that adding this would be at least a somewhat significant change & require a fair amount of testing – however I’m not a software engineer and so I’ve asked Engineering to look into this aspect and determine what it might take to add this functionality.
I think maybe by the end of the week – more information on this perspective should become available.
Sincerely
Jeff Blees
Engineering Technician
Thanks Jeff. I have been looking at the admin guide and appendices but will take another look. i beleive that the machine i was working on is at least part of my issue. i just built a VM to do testing with and can at least send a message via the perl script now…so i am part way there . still getting authentication issues trying to query the iSMS status even thought the user works.
when i said i was able to send messages via the browser, what i did was take the URL that the script was using and just past it in to the browser. not using the UI.
pretty sure that I have the send API properly configured. followed the manual.
i will take a look at those sample scripts that you put in there.
thanks for your help.
matt
Hi Matt,
Have you reviewed our Perl examples and how to found at:
https://webfiles.multitech.com/engineering/sample-code/sms-finder/perl/
? If not, please do.
When you say “I am able to send messages straight from a browser, but…” – do you mean using a browser to access the iSMS server on port 80 and navigating our menus to send with?
How do you have all of the options set in the “SMS API” menu (found in the SMS Services menu)?
Have you reviewed the API appendix in the administrators user guide (starting on page 74)?
http://www.multitech.com/en_US/DOCUMENTS/Families/MultiModemiSMS/manuals.aspx
Lots of customer’s are using Perl scripts to send out text messages through iSMS server.
Sincerely
Jeff Blees
Engineering Technician
Hi Mark,
I take it you are submitting messages via our HTTP Send API function/process. Our Send API provides a MsgID Number in response. You can then perform a MsgID query to track the progress of our transmitting the SMS over to Telco/service provider. This information is depicted in our API Appendix, in the current online Administrator’s User Guide.
We do not track if the Cellular Network has delivered the SMS to the mobile End point.
Sincerely
Jeff Blees
Engineering Technician
Thank you Steve,
but what is the meanings of the “strange” characters I can see in the first 7 bytes ?
For example, each part of a long message is sent in XML as follows (the YYY… sequence is the text body part of the text I sent):
“é@oHt¿£YYYYYYYYY”
but the first 7 characters ? Apart from different values, it seems that the rules for the UDH reported from wiki should fit, at least at a semantical level.
Thanks again
The iSMS doesn’t support PDU mode, so we don’t pass the PDU headers to the receive API. We only pass the message contents. So a three part message will appear as three separate messages.
Hi,
I use a servlet to receive the XML POST of the Receive API.
In case of Concatenated SMS I receive 7 bytes headers which only in part adheres to the rules at:
http://en.wikipedia.org/wiki/Concatenated_SMS
More precisely, the bytes received ar (for a 3 parts message):
1st msg: E9 40 6F 28 31 20 40
2nd msg: E9 40 6F 28 31 A1 40
3rd msg: E9 40 6F 28 31 BF 40
The only rules matched as reported by wiki are:
“Field 4” is the same for all parts.
The “Field 5” should contain the total number of parts, but it seems a bit too much for 3 message parts.
The “Field 6” is the only variable parts, but should evaluate from 1 to N (where N is the number of parts) while in my tests it seems to be random.
Where am I wrong ? Cold you please help me ?
Thanks
Francesco
Hi Anthony,
Sorry for the delayed response (I have been out sick).
Regarding SMS messages sent by the iSMS server (your previous question regarded messages received) – they are kept in one database, regardless of which port it was sent out on. This would be the “Outbox” data base in the SMS Services menu. We do not generate a text file of the numbers we’ve sent to.
As you submit messages for sending (via our API), you get back a MsgID number for each submitted message. You can then query for the status of that MsgID number – to learn if we successfully transmitted the SMS to Telco or not.
Sincerely
Jeff Blees
Engineering Technician
There is no provision for setting the UDH (User Data Header) of an SMS through the iSMS API at this time. Sorry.
Hi,
In order to be able to send notifications to a J2ME application I need to be able to send an sms to a specific port. I can prepare a binary sms with the appropriate UDH to achieve this – the question is, can I send it through the send api ? Would setting enc=2 and sending the raw bytes in hex work ?
Thanks,
Veso
