I tried to send this earlier but, the mailing list was down. So I am sending it again in the hopes that it is useful to someone.
---------- Forwarded message ---------- I have been working on a similar problem. After sifting through all kinds of information and suggestions from all kinds of sources (thanks to all of you by the way), I wrote some scripts to try to determine the real problem. I originally thought that my modem, a MultiTech, did not like certain characters, such as \n. It turns out that the problem was in the at2 7 bit encoding as described here (thanks for the fix Riku): http://www.mail-archive.com/[email protected]/msg08263.html This patch, I believe, is in the latest code set. I simply modified the code I have from June as described, and recompiled. I tested long messages using every character from 0x00 to 0x7F, placing them at the end of the second segment as well as within the segment and I did not get a single error. So the 7 bit encoding now seems correct. I also tested the code without the patch just to make sure that I was understanding what the error really was. I sent 64 messages, all long enough to require 2 message segments. The first message had one character in the second segment, the second had 2 characters in the second segment, and so on up to 64. Every 8th message failed because of the miscalculation of the data length, just as was expected from analyzing the patch. You can see that the calculation problem was a little bit cryptic to detect. 7 out of 8 lengths of messages were successful. The patch has solved this problem for me, it seems. So, I think you might find that your fix is not the .\n sequence but the change in length of your message. Implementing the patch should fix this. I hope, anyway. If you find out that this isn't the same problem, sorry for the lengthy discussion. -Tony On Thu, Jan 15, 2009 at 2:03 AM, Jelle Smet <[email protected]>wrote: > Dear all, > > > > I've ran into a couple of error from which I'm not sure if it's a bug or > misuse from my side. > > I'm sending messages larger than 160 characters spanning 2 to 3 messages > concatenated into 1 message. > > To make this work, I use these parameters: > > max-messages = 6 > > concatenation = true > > > > In general my messages went out without any problem but sometimes messages > generated errors and they didn't arrive at the destination. > > > > These were the errors I encountered in the log file using a Siemens M20 (No > Debug information unfortunately) > > 2008-06-13 16:51:37 [2926] [7] ERROR: AT2[/dev/ttyS0]: CMS ERROR: +CMS > ERROR: 304 > > 2008-06-13 16:51:37 [2926] [7] ERROR: AT2[/dev/ttyS0]: CMS ERROR: Invalid > PDU mode parameter (304) > > 2008-06-13 16:51:38 [2926] [7] ERROR: AT2[/dev/ttyS0]: CMS ERROR: +CMS > ERROR: 304 > > 2008-06-13 16:51:38 [2926] [7] ERROR: AT2[/dev/ttyS0]: CMS ERROR: Invalid > PDU mode parameter (304) > > 2008-06-13 16:51:38 [2926] [7] ERROR: AT2[/dev/ttyS0]: CMS ERROR: +CMS > ERROR: 304 > > > > These ware the errors I encountered in the log file using a Siemens TC 63 > > > > 2009-01-06 10:36:50 [19479] [9] DEBUG: boxc_receiver: sms received > > 2009-01-06 10:36:50 [19479] [9] DEBUG: new split_parts created 0x81b3710 > > 2009-01-06 10:36:50 [19479] [9] DEBUG: send_msg: sending msg to box: > <127.0.0.1> > > 2009-01-06 10:36:52 [19479] [7] DEBUG: AT2[/dev/ttyS0]: TP-Validity-Period: > 24.0 hours > > 2009-01-06 10:36:52 [19479] [7] DEBUG: AT2[/dev/ttyS0]: --> AT+CMGS=69^M > > 2009-01-06 10:36:52 [19479] [7] DEBUG: AT2[/dev/ttyS0]: <-- > > > 2009-01-06 10:36:52 [19479] [7] DEBUG: AT2[/dev/ttyS0]: send command > status: 1 > > 2009-01-06 10:36:52 [19479] [7] DEBUG: AT2[/dev/ttyS0]: --> > 0051000B912394552107F00000A73F050003000202D8611D0C16B3C166B6DA0C26A3E5140A6AB2D92A368B41DD382C67B > > F41F6B29CFC66A7ED611D0C3693D16E381A8E369BC5 > > 2009-01-06 10:36:52 [19479] [7] DEBUG: AT2[/dev/ttyS0]: --> ^Z > > 2009-01-06 10:36:52 [19479] [7] DEBUG: AT2[/dev/ttyS0]: <-- > > > 2009-01-06 10:36:52 [19479] [7] DEBUG: AT2[/dev/ttyS0]: <-- ERROR > > 2009-01-06 10:36:52 [19479] [7] ERROR: AT2[/dev/ttyS0]: Generic error: > ERROR > > 2009-01-06 10:36:52 [19479] [7] DEBUG: AT2[/dev/ttyS0]: send command > status: -1 > > 2009-01-06 10:36:52 [19479] [7] DEBUG: AT2[/dev/ttyS0]: --> AT+CMGS=69^M > > 2009-01-06 10:36:52 [19479] [7] DEBUG: AT2[/dev/ttyS0]: <-- > > > 2009-01-06 10:36:52 [19479] [7] DEBUG: AT2[/dev/ttyS0]: send command > status: 1 > > 2009-01-06 10:36:52 [19479] [7] DEBUG: AT2[/dev/ttyS0]: --> > 0051000B912394552107F00000A73F050003000202D8611D0C16B3C166B6DA0C26A3E5140A6AB2D92A368B41DD382C67B > > F41F6B29CFC66A7ED611D0C3693D16E381A8E369BC5 > > 2009-01-06 10:36:52 [19479] [7] DEBUG: AT2[/dev/ttyS0]: --> ^Z > > 2009-01-06 10:36:52 [19479] [7] DEBUG: AT2[/dev/ttyS0]: <-- > > > 2009-01-06 10:36:52 [19479] [7] DEBUG: AT2[/dev/ttyS0]: <-- ERROR > > 2009-01-06 10:36:52 [19479] [7] ERROR: AT2[/dev/ttyS0]: Generic error: > ERROR > > 2009-01-06 10:36:52 [19479] [7] DEBUG: AT2[/dev/ttyS0]: send command > status: -1 > > 2009-01-06 10:36:52 [19479] [7] DEBUG: AT2[/dev/ttyS0]: --> AT+CMGS=69^M > > 2009-01-06 10:36:53 [19479] [7] DEBUG: AT2[/dev/ttyS0]: <-- > > > 2009-01-06 10:36:53 [19479] [7] DEBUG: AT2[/dev/ttyS0]: send command > status: 1 > > 2009-01-06 10:36:53 [19479] [7] DEBUG: AT2[/dev/ttyS0]: --> > 0051000B912394552107F00000A73F050003000202D8611D0C16B3C166B6DA0C26A3E5140A6AB2D92A368B41DD382C67B > > F41F6B29CFC66A7ED611D0C3693D16E381A8E369BC5 > > 2009-01-06 10:36:53 [19479] [7] DEBUG: AT2[/dev/ttyS0]: --> ^Z > > 2009-01-06 10:36:53 [19479] [7] DEBUG: AT2[/dev/ttyS0]: <-- > > > 2009-01-06 10:36:53 [19479] [7] DEBUG: AT2[/dev/ttyS0]: <-- ERROR > > 2009-01-06 10:36:53 [19479] [7] ERROR: AT2[/dev/ttyS0]: Generic error: > ERROR > > 2009-01-06 10:36:53 [19479] [7] DEBUG: AT2[/dev/ttyS0]: send command > status: -1 > > 2009-01-06 10:36:53 [19479] [7] DEBUG: Set split msg status to 3 > > 2009-01-06 10:36:53 [19479] [7] DEBUG: AT2[/dev/ttyS0]: TP-Validity-Period: > 24.0 hours > > 2009-01-06 10:36:53 [19479] [7] DEBUG: AT2[/dev/ttyS0]: --> AT+CMGS=154^M > > 2009-01-06 10:36:53 [19479] [7] DEBUG: AT2[/dev/ttyS0]: <-- > > > 2009-01-06 10:36:53 [19479] [7] DEBUG: AT2[/dev/ttyS0]: send command > status: 1 > > 2009-01-06 10:36:53 [19479] [7] DEBUG: AT2[/dev/ttyS0]: --> > 0051000B912394552107F00000A7A00500030002018ACD6210F4768FC36C3648FEA687E9E9B75BA750149BC5608B387CB > > > 6AED6A10E9E7ED3E5A0BBFC2D2EB37530180D67CBCD60B21B0C8653288ACD62B0451C3E5BC4EA54377ECBDD65763A0C9A8FD1F2F7785DD6C160B45CEC36B3C96E37182EA650149BC560CBF99CB69C > > D4A9AEEA2697CD2185829A6CB6AA531D5C1E2697CB7090F8BD4EBBC3 > > 2009-01-06 10:36:53 [19479] [7] DEBUG: AT2[/dev/ttyS0]: --> ^Z > > 2009-01-06 10:36:58 [19479] [7] DEBUG: AT2[/dev/ttyS0]: <-- > > > 2009-01-06 10:36:58 [19479] [7] DEBUG: AT2[/dev/ttyS0]: <-- +CMGS: 122 > > 2009-01-06 10:36:58 [19479] [7] DEBUG: AT2[/dev/ttyS0]: <-- OK > > 2009-01-06 10:36:58 [19479] [7] DEBUG: AT2[/dev/ttyS0]: send command > status: 0 > > 2009-01-06 10:36:58 [19479] [7] DEBUG: Parts of concatenated message > failed. > > > > > > > > The messages generating the above mentioned erros had following structure: > (I replaced the content for privacy reasons) > > X being capitalized alphanumeric values > > x being alphanumeric values > > 0 being numeric values > . is unchanged > > : is unchanged > > - is unchanged > > There is a \n (carriage return) between each line. > > > > Failing message: > > ###Start Message#### > > XXXX xxxxxx xxxxxxxx: > > > > XXXX-XXX-XXX:xxxx xxxxxxxxxx:0000000000000 > > > > XXXX-XXX-XXX:xxx xxxxxxxxxxxx:0000000000000 > > > > XXXX-XXX-XXX:xxx xxxxxxx:0000000000000 > > > > XXX-XX:xxxxx xxxxxx:0000000000000 > > > > XXX-XXXX:xxxxxx xxx:0000000000000 > > ###EndMessage#### > > Working message: > > ###Start Message#### > > XXXX xxxxxx xxxxxxxx: > > . > > XXXX-XXX-XXX:xxxx xxxxxxxxxx:0000000000000 > > . > > XXXX-XXX-XXX:xxx xxxxxxxxxxxx:0000000000000 > > . > > XXXX-XXX-XXX:xxx xxxxxxx:0000000000000 > > . > > XXX-XX:xxxxx xxxxxx:0000000000000 > > . > > XXX-XXXX:xxxxxx xxx:0000000000000 > > ###EndMessage#### > > > > > > When replacing the \n (carriage returns) with a ".\n" (a dot and carriage > return the problem was solved and the messages arrived without a problem or > error) > > I'm using Kannel 1.4.1 compiled on Suse Linux. > > Unfortunately I can't confirm if the problem can also be reproduced on > version 1.4.2 > > I hope this information is helpful to someone. > > > > Cheers, > > > > Jelle S. >
