Sorry, the previous message should have been sent to the devel list :P
Anyway, I think I've found the problem, and it's not really on
bearerbox/smsbox, but on sqlbox.
The problem seems to arise if I queue a message manually on the sqlbox,
inserting a record on send_sms. If I leave the sms_type field empty, the
message gets delivered, but the sms_type comes as null.
If that message goes to the store and I reboot kannel, the problem arises.
Seems like bearerbox accepts a null sms_type when the message is already
queued, but not if it comes from the store, since it goes through
dispatch_into_queue and it panics.
I can make a patch to not accept NULL on sqlbox's sms_type (using an ENUM
NOT NULL for the field would do the trick IMHO).
I'll make a patch to fix it, any comments?
Regards,
Alejandro
On Thu, Feb 28, 2008 at 8:37 PM, Alejandro Guerrieri <
[EMAIL PROTECTED]> wrote:
> Dear List,
>
> I'm having problems restarting kannel when messages are on the store (I'm
> using the spool, but the problem happens with the file mode as well).
>
> Sometimes (many times, in fact), if I shutdown kannel when messages are
> being left on the store, I cannot restart without clearing it first.
>
> The error I get is:
>
> 2008-02-28 20:07:08 [27772] [0] PANIC: Not handled sms_type within store!
> 2008-02-28 20:07:08 [27772] [0] PANIC:
> /usr/local/sbin/bearerbox(gw_panic+0xe2) [0x80ce972]
> 2008-02-28 20:07:08 [27772] [0] PANIC: /usr/local/sbin/bearerbox
> [0x8055349]
> 2008-02-28 20:07:08 [27772] [0] PANIC: /usr/local/sbin/bearerbox
> [0x805f836]
> 2008-02-28 20:07:08 [27772] [0] PANIC: /usr/local/sbin/bearerbox
> [0x805f204]
> 2008-02-28 20:07:08 [27772] [0] PANIC: /usr/local/sbin/bearerbox
> [0x805f229]
> 2008-02-28 20:07:08 [27772] [0] PANIC: /usr/local/sbin/bearerbox
> [0x805f768]
> 2008-02-28 20:07:08 [27772] [0] PANIC:
> /usr/local/sbin/bearerbox(main+0x7ba) [0x8054dfa]
> 2008-02-28 20:07:08 [27772] [0] PANIC:
> /lib/tls/i686/cmov/libc.so.6(__libc_start_main+0xe0) [0xb791d050]
> 2008-02-28 20:07:08 [27772] [0] PANIC: /usr/local/sbin/bearerbox
> [0x8053b01]
>
> I did a lot of tests, using the msg_dump() function, and in fact the
> messages are in good shape, except for the "sms_type" value, that gets
> loaded with -1. This obviously triggers the "Unhandled sms_type..." message.
>
> I've narrowed the error to bearerbox.c, dispatch_into_queue(Msg). I've
> patched it to dump messages:
>
> static void dispatch_into_queue(Msg *msg)
> {
> *msg_dump(msg, 0);*
> gw_assert(msg != NULL),
> gw_assert(msg_type(msg) == sms);
>
> *debug("", 0, "TEST sms_type: %d", msg->sms.sms_type);*
> switch (msg->sms.sms_type) {
> case mt_push:
> case mt_reply:
> case report_mt:
> gwlist_append(outgoing_sms, msg);
> break;
> case mo:
> case report_mo:
> gwlist_append(incoming_sms, msg);
> break;
> default:
> panic(0, "Not handled sms_type within store!");
> }
> }
>
> And this is what I get:
>
> ....
> 2008-02-28 20:12:55 [27893] [0] INFO: Kannel bearerbox II version
> cvs-20080226 starting
> 2008-02-28 20:12:55 [27893] [4] DEBUG: Thread 4
> (gw/bb_boxc.c:sms_to_smsboxes) maps to pid 27893.
> ...
> 2008-02-28 20:12:55 [27893] [5] DEBUG: Thread 5 (gw/bb_boxc.c:smsboxc_run)
> maps to pid 27893.
> 2008-02-28 20:12:55 [27893] [0] DEBUG: Msg object at 0x819e938:
> 2008-02-28 20:12:55 [27893] [0] DEBUG: type: sms
> 2008-02-28 20:12:55 [27893] [0] DEBUG: sms.sender:
> 2008-02-28 20:12:55 [27893] [0] DEBUG: Octet string at 0x819ea00:
> 2008-02-28 20:12:55 [27893] [0] DEBUG: len: 5
> 2008-02-28 20:12:55 [27893] [0] DEBUG: size: 6
> 2008-02-28 20:12:55 [27893] [0] DEBUG: immutable: 0
> 2008-02-28 20:12:55 [27893] [0] DEBUG: data: xx xx xx xx xx
> xxxxx
> 2008-02-28 20:12:55 [27893] [0] DEBUG: Octet string dump ends.
> 2008-02-28 20:12:55 [27893] [0] DEBUG: sms.receiver:
> 2008-02-28 20:12:55 [27893] [0] DEBUG: Octet string at 0x819ea18:
> 2008-02-28 20:12:55 [27893] [0] DEBUG: len: 10
> 2008-02-28 20:12:55 [27893] [0] DEBUG: size: 11
> 2008-02-28 20:12:55 [27893] [0] DEBUG: immutable: 0
> 2008-02-28 20:12:55 [27893] [0] DEBUG: data: xx xx xx xx xx xx xx xx xx
> xx xxxxxxxxxx
> 2008-02-28 20:12:55 [27893] [0] DEBUG: Octet string dump ends.
> 2008-02-28 20:12:55 [27893] [0] DEBUG: sms.udhdata:
> 2008-02-28 20:12:55 [27893] [0] DEBUG: Octet string at 0x819ea40:
> 2008-02-28 20:12:55 [27893] [0] DEBUG: len: 0
> 2008-02-28 20:12:55 [27893] [0] DEBUG: size: 0
> 2008-02-28 20:12:55 [27893] [0] DEBUG: immutable: 0
> 2008-02-28 20:12:55 [27893] [0] DEBUG: Octet string dump ends.
> 2008-02-28 20:12:55 [27893] [0] DEBUG: sms.msgdata:
> 2008-02-28 20:12:55 [27893] [0] DEBUG: Octet string at 0x8197b18:
> 2008-02-28 20:12:55 [27893] [0] DEBUG: len: 156
> 2008-02-28 20:12:55 [27893] [0] DEBUG: size: 157
> 2008-02-28 20:12:55 [27893] [0] DEBUG: immutable: 0
> 2008-02-28 20:12:55 [27893] [0] DEBUG: data: 55 74 69 6c 69 7a 61 20 65
> 6c 20 6e 75 65 76 6f Utiliza el nuevo
> ...
> 2008-02-28 20:12:55 [27893] [0] DEBUG: data: 69 61 20 61 20 6c 61 20 76
> 65 7a 2e ia a la vez.
> 2008-02-28 20:12:55 [27893] [0] DEBUG: Octet string dump ends.
> 2008-02-28 20:12:55 [27893] [0] DEBUG: sms.time: 1202072929
> 2008-02-28 20:12:55 [27893] [0] DEBUG: sms.smsc_id:
> 2008-02-28 20:12:55 [27893] [0] DEBUG: Octet string at 0x81a4450:
> 2008-02-28 20:12:55 [27893] [0] DEBUG: len: 11
> 2008-02-28 20:12:55 [27893] [0] DEBUG: size: 12
> 2008-02-28 20:12:55 [27893] [0] DEBUG: immutable: 0
> 2008-02-28 20:12:55 [27893] [0] DEBUG: data: xx xx xx xx xx xx xx xx xx
> xx xx xxxxxxxxxxx
> 2008-02-28 20:12:55 [27893] [0] DEBUG: Octet string dump ends.
> 2008-02-28 20:12:55 [27893] [0] DEBUG: sms.smsc_number:
> 2008-02-28 20:12:55 [27893] [0] DEBUG: sms.service:
> 2008-02-28 20:12:55 [27893] [0] DEBUG: Octet string at 0x81a4478:
> 2008-02-28 20:12:55 [27893] [0] DEBUG: len: 7
> 2008-02-28 20:12:55 [27893] [0] DEBUG: size: 8
> 2008-02-28 20:12:55 [27893] [0] DEBUG: immutable: 0
> 2008-02-28 20:12:55 [27893] [0] DEBUG: data: 33 37 30 30 39 39
> 35 3700995
> 2008-02-28 20:12:55 [27893] [0] DEBUG: Octet string dump ends.
> 2008-02-28 20:12:55 [27893] [0] DEBUG: sms.account:
> 2008-02-28 20:12:55 [27893] [0] DEBUG: Octet string at 0x81a44a0:
> 2008-02-28 20:12:55 [27893] [0] DEBUG: len: 0
> 2008-02-28 20:12:55 [27893] [0] DEBUG: size: 0
> 2008-02-28 20:12:55 [27893] [0] DEBUG: immutable: 0
> 2008-02-28 20:12:55 [27893] [0] DEBUG: Octet string dump ends.
> 2008-02-28 20:12:55 [27893] [0] DEBUG: sms.id:
> 84de8b2b-ab42-47ae-8d2b-45ecbce50e11
> 2008-02-28 20:12:55 [27893] [0] DEBUG: sms.sms_type: -1
> 2008-02-28 20:12:55 [27893] [0] DEBUG: sms.mclass: -1
> 2008-02-28 20:12:55 [27893] [0] DEBUG: sms.mwi: -1
> 2008-02-28 20:12:55 [27893] [0] DEBUG: sms.coding: -1
> 2008-02-28 20:12:55 [27893] [0] DEBUG: sms.compress: -1
> 2008-02-28 20:12:55 [27893] [0] DEBUG: sms.validity: -1
> 2008-02-28 20:12:55 [27893] [0] DEBUG: sms.deferred: -1
> 2008-02-28 20:12:55 [27893] [0] DEBUG: sms.dlr_mask: -1
> 2008-02-28 20:12:55 [27893] [0] DEBUG: sms.dlr_url:
> 2008-02-28 20:12:55 [27893] [0] DEBUG: Octet string at 0x81a44b8:
> 2008-02-28 20:12:55 [27893] [0] DEBUG: len: 0
> 2008-02-28 20:12:55 [27893] [0] DEBUG: size: 0
> 2008-02-28 20:12:55 [27893] [0] DEBUG: immutable: 0
> 2008-02-28 20:12:55 [27893] [0] DEBUG: Octet string dump ends.
> 2008-02-28 20:12:55 [27893] [0] DEBUG: sms.pid: -1
> 2008-02-28 20:12:55 [27893] [0] DEBUG: sms.alt_dcs: -1
> 2008-02-28 20:12:55 [27893] [0] DEBUG: sms.rpi: -1
> 2008-02-28 20:12:55 [27893] [0] DEBUG: sms.charset:
> 2008-02-28 20:12:55 [27893] [0] DEBUG: Octet string at 0x81a4500:
> 2008-02-28 20:12:55 [27893] [0] DEBUG: len: 0
> 2008-02-28 20:12:55 [27893] [0] DEBUG: size: 0
> 2008-02-28 20:12:55 [27893] [0] DEBUG: immutable: 0
> 2008-02-28 20:12:55 [27893] [0] DEBUG: Octet string dump ends.
> 2008-02-28 20:12:55 [27893] [0] DEBUG: sms.boxc_id:
> 2008-02-28 20:12:55 [27893] [0] DEBUG: Octet string at 0x81a4518:
> 2008-02-28 20:12:55 [27893] [0] DEBUG: len: 6
> 2008-02-28 20:12:55 [27893] [0] DEBUG: size: 7
> 2008-02-28 20:12:55 [27893] [0] DEBUG: immutable: 0
> 2008-02-28 20:12:55 [27893] [0] DEBUG: data: 73 71 6c 62 6f
> 78 sqlbox
> 2008-02-28 20:12:55 [27893] [0] DEBUG: Octet string dump ends.
> 2008-02-28 20:12:55 [27893] [0] DEBUG: sms.binfo:
> 2008-02-28 20:12:55 [27893] [0] DEBUG: Octet string at 0x81a4540:
> 2008-02-28 20:12:55 [27893] [0] DEBUG: len: 17
> 2008-02-28 20:12:55 [27893] [0] DEBUG: size: 18
> 2008-02-28 20:12:55 [27893] [0] DEBUG: immutable: 0
> 2008-02-28 20:12:55 [27893] [0] DEBUG: data: 70 72 6f 63 65 73 61 64 6f
> 5f 72 65 63 6f 62 72 procesado_recobr
> 2008-02-28 20:12:55 [27893] [0] DEBUG: data:
> 6f o
> 2008-02-28 20:12:55 [27893] [0] DEBUG: Octet string dump ends.
> 2008-02-28 20:12:55 [27893] [0] DEBUG: sms.msg_left: -1
> 2008-02-28 20:12:55 [27893] [0] DEBUG: sms.split_parts: (nil)
> 2008-02-28 20:12:55 [27893] [0] DEBUG: sms.priority: -1
> 2008-02-28 20:12:55 [27893] [0] DEBUG: sms.resend_try: -1
> 2008-02-28 20:12:55 [27893] [0] DEBUG: sms.resend_time: -1
> 2008-02-28 20:12:55 [27893] [0] DEBUG: Msg object ends.
> *2008-02-28 20:12:55 [27893] [0] DEBUG: TEST sms_type: -1
> *2008-02-28 20:12:55 [27893] [0] PANIC: Not handled sms_type within store!
> 2008-02-28 20:12:55 [27893] [0] PANIC:
> /usr/local/sbin/bearerbox(gw_panic+0xe2) [0x80ce972]
> 2008-02-28 20:12:55 [27893] [0] PANIC: /usr/local/sbin/bearerbox
> [0x8055349]
> 2008-02-28 20:12:55 [27893] [0] PANIC: /usr/local/sbin/bearerbox
> [0x805f836]
> 2008-02-28 20:12:55 [27893] [0] PANIC: /usr/local/sbin/bearerbox
> [0x805f204]
> 2008-02-28 20:12:55 [27893] [0] PANIC: /usr/local/sbin/bearerbox
> [0x805f229]
> 2008-02-28 20:12:55 [27893] [0] PANIC: /usr/local/sbin/bearerbox
> [0x805f768]
> 2008-02-28 20:12:55 [27893] [0] PANIC:
> /usr/local/sbin/bearerbox(main+0x7ba) [0x8054dfa]
> 2008-02-28 20:12:55 [27893] [0] PANIC:
> /lib/tls/i686/cmov/libc.so.6(__libc_start_main+0xe0) [0xb796e050]
> 2008-02-28 20:12:55 [27893] [0] PANIC: /usr/local/sbin/bearerbox
> [0x8053b01]
>
> This happens with all messages on store (if I replace the panic() with
> debug() I get a line for each and every message!)
>
> So, the messages seems to be ok, but for some reason the sms_type gets
> garbled on the store, or not properly load during init().
>
> Here's an hexdump of the file on disk (I've garbled the numbers for
> privacy):
>
> 00000000 00 00 00 02 00 00 00 05 78 78 78 78 78 00 00 00
> |........xxxxx...|
> 00000010 0a 78 78 78 78 78 78 78 78 78 78 00 00 00 00 00
> |.xxxxxxxxxx.....|
> 00000020 00 00 9c 55 74 69 6c 69 7a 61 20 65 6c 20 6e 75 |...Utiliza el
> nu|
> 00000030 65 76 6f 20 61 6e 69 6f 20 71 75 65 20 63 6f 6d |evo anio que
> com|
> 00000040 69 65 6e 7a 61 20 70 61 72 61 20 65 73 74 61 62 |ienza para
> estab|
> 00000050 6c 65 63 65 72 20 70 72 6f 79 65 63 74 6f 73 20 |lecer
> proyectos |
> 00000060 72 65 61 6c 69 73 74 61 73 20 79 20 70 6f 73 69 |realistas y
> posi|
> 00000070 62 6c 65 73 2e 20 45 73 20 64 69 66 69 63 69 6c |bles. Es
> dificil|
> 00000080 20 63 61 6d 62 69 61 72 20 64 65 20 75 6e 20 64 | cambiar de
> un d|
> 00000090 69 61 20 70 61 72 61 20 65 6c 20 6f 74 72 6f 2c |ia para el
> otro,|
> 000000a0 20 70 72 6f 62 61 20 68 61 63 65 72 6c 6f 20 75 | proba
> hacerlo u|
> 000000b0 6e 20 64 69 61 20 61 20 6c 61 20 76 65 7a 2e 47 |n dia a la
> vez.G|
> 000000c0 a6 2d 61 00 00 00 0b 78 78 78 78 78 78 78 78 78
> |.-a....xxxxxxxxx|
> 000000d0 78 78 ff ff ff ff 00 00 00 07 33 37 30 30 39 39
> |xx........370099|
> 000000e0 35 00 00 00 00 00 00 00 24 38 34 64 65 38 62 32
> |5.......$84de8b2|
> 000000f0 62 2d 61 62 34 32 2d 34 37 61 65 2d 38 64 32 62
> |b-ab42-47ae-8d2b|
> 00000100 2d 34 35 65 63 62 63 65 35 30 65 31 31 ff ff ff
> |-45ecbce50e11...|
> 00000110 ff ff ff ff ff ff ff ff ff ff ff ff ff ff ff ff
> |................|
> 00000120 ff ff ff ff ff ff ff ff ff ff ff ff ff 00 00 00
> |................|
> 00000130 00 ff ff ff ff ff ff ff ff ff ff ff ff 00 00 00
> |................|
> 00000140 00 00 00 00 06 73 71 6c 62 6f 78 00 00 00 11 70
> |.....sqlbox....p|
> 00000150 72 6f 63 65 73 61 64 6f 5f 72 65 63 6f 62 72 6f
> |rocesado_recobro|
> 00000160 ff ff ff ff ff ff ff ff ff ff ff ff ff ff ff ff
> |................|
> *
> 00000180
>
> Any hints? Anyone else capable of reproducing this?
>
> Regads,
> --
> Alejandro Guerrieri
> Magicom
> http://www.magicom-bcn.net/
> LinkedIn: http://www.linkedin.com/in/aguerrieri
--
Alejandro Guerrieri
Magicom
http://www.magicom-bcn.net/
LinkedIn: http://www.linkedin.com/in/aguerrieri
