Hi list,
seems Cygwin (at least 1.3.10) did not have a reentrant
gethostbyname() here, right?!
See attached Kannel automatic compilation test log that runs on our
Cygwin machine.
Stipe
[EMAIL PROTECTED]
---
Wapme Systems AG
Alex Judd schrieb:
Anyone else having problems with CVS smsc_at.c?
I'm having problems with a Wavecom Fastrack M1200. But I was
considering this is a device issue.
What have been the recent changes in that module?
Stipe
[EMAIL PROTECTED]
I'm getting this error often,
2003-10-01 15:51:01 [6] ERROR: error connecting to server
`xx.xx.xx.xx' at port `5016'
2003-10-01 15:51:01 [6] ERROR: SMPP[SMPP_ABCD]: Couldn't connect to
server.
2003-10-01 15:51:01 [6] ERROR: SMPP[SMPP_ABCD]: Couldn't connect to
SMS center (retrying in 10
Pak,
first of all, please keep sending to the devel@ mailing list instead
of addressing persons directly. They tend to feel annoyed ;)
Pak Bei wrote:
To send messages we need to call the smsbox with the sendsms
interface using http post cgi-bin/sendsms.
Beside cgi-bin/sendsms we can use
As I reported earlier, wml_compiler.c has bug.
If you try to process attached test.wml with wml_tester
(see kannel test directory) then you will see PANIC message.
[0] INFO: WBXML: No wbxml version given, assuming 1.1
[0] DEBUG: WBXML: WML with ExternalID -//WAPFORUM//DTD WML 1.1//EN
[0]
Yury Mikhienko wrote:
No, not radius-acct section.
with RADIUS config I'm mean the radius server config.
My problem detail:
1. Kannel get account request from NAS properlly.
2. Kannel succesfully forward that request to radius server
3. Kannel got responce from radius server.
but radius
Hi Marc,
Marc Ricart wrote:
Is there any specification on Openwave OTA provisioning ?
Anyone could share more information ?
there is, but it's not open and under NDA. Unfortunatly I never seen
it on paper ;(
I' ve tested the information found in this list on a SIEMENS SL 55.
I want to
Vjacheslav Chekushin wrote:
I have some questions about new transcoding feature, implemented in
cvs tree. (see http://www.mail-archive.com/[EMAIL PROTECTED]/msg00196.html)
(In Kannel 1.3.1 all content, compiled by libxml2 was sent to phone in UTF-8)
Stipe Tolj wrote:
...
defines
Vjacheslav Chekushin wrote:
BUG #33 can be closed, applying followng patch:
I think it is clear that we must not send NULL value to wap-appl thread.
(NULL value will destroy apllication thread)
--- ../devel_1.2.1_base/wap/wsp_unit.c Thu Jan 2 12:04:16 2003
+++ wap/wsp_unit.c Thu
Vjacheslav Chekushin wrote:
There is transcoding staff in wap-appl.c (static void return_reply()) and
wml_compiler.c (some parsing routines, that do charset_convert).
There are some bugs in this transcoding staff (I affraid - many bugs) but
I want waste my time to clean this code.
I just
Hi list,
I commited these two issues today:
2003-10-08 Stipe Tolj [EMAIL PROTECTED]
* utils/foobar-config.sh: added generic foobar-config shell script
generator script to generate gw-config which may be used by 3rd
party configure processes to determine Kannel's gwlib
Damjan wrote:
I commited these two issues today:
2003-10-08 Stipe Tolj [EMAIL PROTECTED]
* utils/foobar-config.sh: added generic foobar-config shell script
generator script to generate gw-config which may be used by 3rd
party configure processes to determine
David Chkhartishvili schrieb:
Is this patch commited to CVS?
not yet, but will do this in a couple of moments.
Usually we have a *rule*. That applied patches from the mailing list
will have a mail in the thread that states that it has been applied.
(as a backup, you can always check via
Alexander Malysh wrote:
Hi again,
oops, Slava thank you for pointing ;)
David can you confirm that attached patch works for you?
+1, commited to cvs.
Thanks again Vjachslav for pointing to it.
Stipe
[EMAIL PROTECTED]
---
For MO, we have now a '%B' escape code for the sms-service to pass
this binfo field into any HTTP requests and for MT we have a 'binfo'
CGI variable in the sendsms interface.
BTW, both have been documented in the user's guide.
Stipe
[EMAIL PROTECTED]
Hi Paul,
Paul Keogh wrote:
In EMI2 we forward the XSer 0c field and in SMPP we forward the
service_type from the deliver_sm.
The EMI-UCP specification suggests that this field is reserved for
future use.
What syntax are you using for this field and is it operator specific ?
AFAIS,
Hi Alex,
Alex Kinch wrote:
Just rebuilt our test box to use cvs-20031016 (previously was
running 1.3.1-devel). I'm now experiencing problems with incoming
SMPP. The message comes in, and shows up the bearerbox debug log -
but nothing happens. It doesn't go anywhere!
Basically the SMPP
Alexander Malysh wrote:
have you notice a command status in deliver_sm_resp? it's 0x0a == Invalid
source address. And see source address in deliver_sm: '+' in front which is
not allowed.
So you should try to contact telco and tell them that smsc is broken.
And second: please try attached
I think it's nice if we could specify billing info using
X-Kannel-BInfo header from sms-service.
I'm not sure about smsc_http.c ... is that correct?
Diff output from latest cvs attached.
definetly +1. Applied to cvs. Thanks a lot.
Stipe
mailto:[EMAIL PROTECTED]
corret,
Kannel as a WAP gateway receives/sends UDP packets over IP to the
device. Hence the device has to establish an IP session before
transmitting and receiving of WDP (UDP) datagrams.
Stipe
mailto:[EMAIL PROTECTED]
---
Wapme
Juan Enrique Gómez wrote:
El mar, 28-10-2003 a las 16:27, Alex Judd escribió:
Just a stupid question, how will you download the binary data from the
MMSc center?, cause (at least in Spain) the operators has the MMS
content available via GPRS but on a private network, not available
through
Jörg Pommnitz schrieb:
Stipe Tolj wrote:
correct. Usually the MMSC URL is located in a private network. The
only practible way IMO is to use a GPRS modem device to download the
MMS via the device.
But this is expensive and unreliable in terms of throughput.
Why should
Right now I'm using kannel towards Telenor (Norwegian telco running Logica
CMG SMSC) and it works fine (when you remember to specify
interface-version=33, that part took some time to figure out :-P).
Actually the only part of kannel I use is the smpp parsing/sending,
forwarding of the
Rory Campbell-Lange schrieb:
I've hacked the dlr mechanism to do an update rather than a delete.
Pretty simple to do -- even though I don't know C!
I'd be very keen for Kannel to dispense with the http interface
altogether and only supply database connectivity. Working with Postgres
If you mean, postgresql support for kannel then it's pretty simple :) and I
will look in it over weekend...
great, patches are welcome.
Consider doing this the right way by having the PostgreSQL support
implemented in our database pool architecture (see gwlib/dbpool.[ch])
Stipe
Paul Keogh wrote:
I'm +1 for this idea but I'm still not available to look at the code.
Also think image/png - image/gif = convert %s %s to
convert png's
to gifs if content is png and device doesn't accept it (SE-T610 for
example).
I think this is more within the UAProf space than
Im having a problem sending operator logos with the latest CVS kannel.
The same service has worked perfectly with an over 2 years old version of
the gateway.
But I think I can see the problem in the log dump below.
Why the heck is there a ISO-8859-11 right in the middle of the binary data,
Jörg Pommnitz wrote:
I suspect Kannel needs some cleanup for this to work properly. Currently
each protocol entity (e.g. application layer, WSP, WTP,...) contain a lot
of redundant code:
They provide an external interface, to init, cleanup and handle events and
then do a lot of things by
Peter,
check the mailing list archive. Siemens uses Openwave OTA specs for
their devices.
Stipe
mailto:[EMAIL PROTECTED]
---
Wapme Systems AG
Münsterstr. 248
40470 Düsseldorf, NRW, Germany
phone: +49.211.74845.0
fax:
Alexander Malysh wrote:
Hi,
'at2' module was renamed to just 'at' , so please update your config...
these changes will have to be reflected in the NEWS file for the next
development release.
Stipe
Stefan Cars wrote:
This patch allows you to set the source_addr_ton when sending the SM by
using a new cgi var called source-addr-ton. This is necessary when
speaking to some providers. Maybe the same thing should be done for the NPI
aswell, but I have not come across any case where that
Aarno Syvänen wrote:
WSP version is 1.2 (This is not same thing as the wap version).
This is because not all phones tell Kannel their own wsp version,
and we of course want support all phones (no version information
means 1.2).
but Kannel support all WSP encoding even up to 1.5 if the
Okey, I'll try to see if I have any spare time next week to start work on
something like that, could be useful in many aspects!
great, keep going.
@all: seems we need to keep maintaining the STATUS file to reflect
such things. I'll get as the ammount of developers has decreased
drastically,
Alexander Malysh wrote:
Hi Igor,
++1 for this patch...
It's 100% wrong to use octstr_imm for non const c-strings. And this can also
explain why we have seen panics with too many immutable octstr...
so, is there any need in changing code at different other places?
ie.
Aarno Syvänen wrote:
Hi List,
Some new PPG fixes (see attachment). There are mostly about
performance,
removing some useless octstr_append_char. In addition, PPG now do not
search optional headers from the message body. For doing this a new
octstr
function is supplied.
And, yes,
Alexander Malysh wrote:
On Tuesday 11 November 2003 13:44, Stipe Tolj wrote:
Alexander Malysh wrote:
Hi Igor,
++1 for this patch...
It's 100% wrong to use octstr_imm for non const c-strings. And this can
also explain why we have seen panics with too many immutable octstr
Igor Ivoilov wrote:
Certainly!
The same error. octstr_imm() is sutable only for static strings.
Stipe are you going to change it?
How often octstr_case_search is used in such circumstances?
Does it worth to create octstr_case_search(Octstr*,char*) ?
I'll check the sources if I can find
Alexander Malysh wrote:
On Tuesday 11 November 2003 13:56, Stipe Tolj wrote:
Aarno Syvänen wrote:
Hi List,
Some new PPG fixes (see attachment). There are mostly about
performance,
removing some useless octstr_append_char. In addition, PPG now do not
search optional headers
I have a question regarding the SSL issue. I use kannel cvs v20031110 - SSL
enabled. Generally works fine, but check this out:
1) I issue the URL
http://kannel-server:13000/status.html from a client PC through Internet
Explorer
I get Page can not display in the IE window (which is correct)
Thomas Hager schrieb:
On Wed, 2003-11-12 at 01:58, Alex Kinch wrote:
Hi Thomas,
hi alex,
Version is cvs-20031103. Maybe it is a bug, and this probably now
belongs on the devel-list so I've cc'd it in.
i've quickly checked current cvs and it seems as if the bug i described
is fixed.
David Tully wrote:
Sounds like a provider issue..
We have a provider that does just this - delivers DLRs on one bind that
relate to another bind. I get round this by giving the binds the same
SMSC-ID.
correct!
This is currently the only practible solution. Name all SMPP smsc
links with
ok, you mean it would break in terms of routing if you use the
smsc=foobar variable in the sendsms interface, right?
To solve this, do the following: name the links with the same smcs-id,
to ensure you get the DLRs processed. But, and that's the trick,
include unified smsc-ids to allowed-smsc-id
Alex Kinch schrieb:
Sorry to sound like a broken record, but wondered how smppbox was
coming along. Actually I was also wondering how the sqlbox/send
using sql insert trigger was going, but one thing at a time :-)
I still have no convinient way in makeing things modulary configure,
hence
Tarun Dua wrote:
We get the following delivery report format in delivery_sm PDU from the
SMPP smsc
id:1 sub:1 dlvrd:1 submit date:031113120612 done date:031113120616
stat:4 err:0
However according to kannel source gw/smsc/smsc_smpp.c
--
/*
Aarno Syvänen wrote:
This was indeed a fakewap specific thing: leak appears when there are
no headers.
But even when stress testing pure wap, you probably want to add headers
sending
to fakewap, so this leak is not important.
which may bring us to the point again to re-code fakewap with
Jörg Pommnitz wrote:
which may bring us to the point again to re-code fakewap with real
WSP and WTP code from the libwap.a libs.
Which means that the one (or more) threads per stack entity
pattern of doing things in Kannel must go. Writing a client that
has to start, manage and
Developers, CVS commiters,
I'd like to go ahead and change the copyright owner from WapIT Ltd
to Kannel Group, as I mentioned this in a previous announcement.
The license I proposed was derived closely from Apache's software
license (ASL) and which is 100% compatible to our BSD-style license.
It
Kalle,
thanks for nitpicking, here you go with an updates one.
Stipe
mailto:[EMAIL PROTECTED]
---
Wapme Systems AG
Münsterstr. 248
40470 Düsseldorf, NRW, Germany
phone: +49.211.74845.0
fax: +49.211.74845.299
mailto:[EMAIL
Aarno Syvänen wrote:
PAP is mostly used to send SI, SLs and MMS notifications, so content is very
specific here (and not the actual content).
Aarno
On 14.11.2003, at 09:32, ghhou wrote:
Dear sir,
Does kannel support multiple push contents in PAP interface? That is, in the
In a word, now kannel doesn't support Complex Multipart (embedded multipart
message MIME) sepecified in rfc2049.
not for the PPG part. The gwlib/mime.[ch] is used within our MMSC for
these purposes and could be addopted to be used in PPG sources too.
But If you have such a drastic need, you
How do I resolve this error?
[6] ERROR: EMI2[IP:PORT]: Got negative ack. op:60, trn:0, error:4 (Operation
not allowed), message: Session not accepted, number of sessions exceeeded
[6] ERROR: EMI2[IP:PORT]: Server rejected our login, giving up
you could not log in to your SMSC connection. I
What is Xser field?
XSer is a EMI/UCP 4.x specific item. See the official specs for more
indeep information.
Our operator ask me to insert the billing code
and tariff class in the Xser field, how can it be done?
Is it already implemented in the existing version?if not is there any
forgot to say. It's in the current cvs head tree, not in 1.3.1 or
previous.
Stipe
mailto:[EMAIL PROTECTED]
---
Wapme Systems AG
Münsterstr. 248
40470 Düsseldorf, NRW, Germany
phone: +49.211.74845.0
fax: +49.211.74845.299
Igor Ivoilov wrote:
I have changed packet size to be 1400 on my production server
and don't see any problem for now.
I have not see javagames more than 358k till now.
So as a temporary work around it's very good me.
I don't have any other problem. But we provide access to our wap gw only to
Robert Hassing wrote:
Hi All,
I have this problem with the WAP gateway I did setup.
The problem is that a maximum of 254/255 packets can be send concerning 1
file.
This means that if I want to download a file with a size higher as 143KB the
transmission fails :-(
The error I got is
Does Kannel suport HTTP redirection?
yes, you can even specify how many redirection hops Kannel will follow
as HTTP client. Please see user's guide for more details, or explain
in which situation you are trying to use it.
Stipe
mailto:[EMAIL PROTECTED]
Hi all,
can't this be done using a specific 'data_coding' value in SMPPs
PDUs?`
Or what about using alt-charset in the smsc group to use iconv()
functionality to convert the character encoding?
Unfortunatly I'm not aware of too much greek things regarding SMS
character encoding.
Stipe
Daniele Dal Verme wrote:
I'm running ver 1.3.1 on my Linux Box RH 7.3. I have one wavecom modem
attached to my ttyS1 serial port. This is the smsc section:
group = smsc
smsc = at
modemtype = wavecom
device = /dev/ttyS1
speed = 28800
pin = 1234
I need to execute some AT command to
Bruno, nice to hear you again.
BUT: don't crosspost to mailing lists!!! *g*
Bruno Rodrigues wrote:
Igor Ivoilov [EMAIL PROTECTED] wrote:
I have changed packet size to be 1400 on my production server
and don't see any problem for now.
I have not see javagames more than 358k till now.
Vjacheslav Chekushin wrote:
Hi, list.
It is not enough just to change packet size. We must avoid PANIC if
maximal document size exceeded.
don't know yet, still untested from my side.
But(!), the PANIC itself can be considered as bug, since we don't want
to stop the wapbox using a PANIC if
Vjacheslav Chekushin wrote:
It is not enough just to change packet size. We must avoid PANIC if
maximal document size exceeded.
Wap gw must send abort indication instead of PANIC if segment's count
more than 255.
Attached patch do it.
++1, commited to cvs.
Stipe
mailto:[EMAIL
[EMAIL PROTECTED] wrote:
I believe that the best way to do it, is to alter the C code of the
charset.c file. But what about the iconv() functionality and alt-charset you
are referring to? The alt-charset variable in the SMSC group, also uses the
charset.c file?
manually changing .c files is
[EMAIL PROTECTED] wrote:
You are right about this. In the CVS manual I checked out the alt-charset
directive. It is supposed to be a nummeric value, but I don't know which
value do I have to put there. Can you give me an example, or just give me
the range of this numeric field ?
the value
there may be a good change to use libsdb to interact via ODBC to the
MS SQL server. But I guess non of us here has ever tried this.
And to be honest, I'd wouldn't consider it as a production solution.
Stipe
mailto:[EMAIL PROTECTED]
Bruno Rodrigues schrieb:
Igor Ivoilov [EMAIL PROTECTED] wrote:
Nokia 6600 does not need wap gw, the same to SonyEricsson P800.
I think that when we implement ESAR there will very few phones which
use wap-gw ;)
And guess what's CMG's wap 2.0 gateway based ?
tchan tchan tchan
Phuah Yee Keat wrote:
I was previously just a user of Kannel (such a great project!), and now
am dwelving into developing a new smsc module, for our local telco
support in Malaysia. The standard's (or just a spec) name is CPA
(Content Provider Architecture).
Its a http-get type interface
thanks for testing.
commited to cvs.
Stipe
mailto:[EMAIL PROTECTED]
---
Wapme Systems AG
Münsterstr. 248
40470 Düsseldorf, NRW, Germany
phone: +49.211.74845.0
fax: +49.211.74845.299
mailto:[EMAIL PROTECTED]
Angel Fradejas schrieb:
Indeed, kannel 1.3.1 outputs a warning every time it sees param 069.
My advise: switch as soon as you can to CVS-version, as it is far more
stable than 1.3.1
agreed. CVS head is currently very stable.
Stipe
mailto:[EMAIL PROTECTED]
Rory Campbell-Lange wrote:
How can anyone help to get the mmsbox up and running?
hmmm, you can't help ;) the beast is still living in a cage here at
Wapme and waits to be released.
The only way to get this release faster done is to consider modular
configure/make process enhancements so we
Would you pls tell me where to get mmsbox software? I can't found it in
Kannel.
it's yet not released, consider it property of Wapme at the present
time.
Stipe
mailto:[EMAIL PROTECTED]
---
Wapme Systems AG
Münsterstr. 248
40470
ghhou wrote:
If there is comparative URL in the redirection page, then error will occurs.
For the terminal keeps the original Host, it will send the request to the
original web site.
sorry, seems it's too early in the morning ;) -- I didn't get the
point. Can you please re-fraze this?!
A will redirect the request to B. But B can
only return the page sent before.
Best Regards,
Guanghua Hou
- Original Message -
From: Stipe Tolj [EMAIL PROTECTED]
To: ghhou [EMAIL PROTECTED]
Cc: [EMAIL PROTECTED]
Sent: Wednesday, November 26, 2003 4:25 PM
Subject: Re: HTTP redirect
ghhou wrote
Raditha Dissanayake wrote:
Stipe,
this sounds like a beast that everyone want's to get released from it's
cage :-) I know 'me too' posts are discouraged but can't resist this one
yep, seems it's a most wanted animal living in this little Wapme box
;)
So it's not up to me. Main holding
Martin,
please don't cross-post to lists!
I'm on the look out for tools/emulators with which to test the SMSC
connections. What do people use? What do the Kannel developers use? I'm
talking about the SMPP, CIMD, UCP/EMI, OIS, SM/ASI etc.
Now, Kannel has it's own fake SMSC connection module
Vjacheslav Chekushin wrote:
Hi, list
In wsp_server_method_states.def:
/*
* The Siemens S35 adds an extra NUL character to the end
* of the request body which may not work with certain cgi
* scripts. It is removed
Paul Keogh wrote:
There can't be that many S35 floating around still; I recall that the
WAP browser UI was so bad to be almost unusable.
I'd recommend forgetting about the S35 and removing this hack...
+1 from me. Any objections by others?
Stipe
mailto:[EMAIL PROTECTED]
ok, while digging in the code this seems to be the more relevant part:
static struct emimsg *make_emi31(PrivData *privdata, int trn)
{
struct emimsg *emimsg;
if(octstr_len(privdata-username) ||
octstr_len(privdata-my_number)) {
emimsg = emimsg_create_op(31, trn, privdata-name);
Alex Kinch wrote:
Came across this today as a new DLR format on SMPP v3.4. Any
thoughts - apart from telling the carrier in question to actually
follow some standards? :-)
By the way it's not a small carrier, it's quite a large one
beginning with V.. no prizes for guessing.
Alex
Alexander Malysh wrote:
just import as independent cvs module and later (if me or others have time to
do configure stuff) we can put it into gateway module.
yep, was thinking of about exactly that... ;)
Stipe
mailto:[EMAIL PROTECTED]
Alexander Malysh wrote:
correct , the way it's done now ...
ok, the rumble is opened...
In the left corner: Paul, chief developer at ANAM, irish wireless
software solution provider.
In the right corner: Alexander, chief Kannel developer at Centrium,
Bulk-SMS carrier.
:]]
Stipe
Cesar Garcia wrote:
Hi,
How I can submit modifications to the CVS?
in terms of modifications, we have the following guide of rules:
* user provides a 'diff -u' formated patch of their changes against
current cvs head tree
* user submits the patch with a [PATCH] prefix in the subject
seems they are all narrow minded.
Sometimes it appears that those people did have got the point about
agreeing to a protocoll/specification. They don't get it that it will
be for the sake of all and for their personal/business benefit.
It's like in (normal) life. To some you can show and they
Paul Keogh wrote:
but it's simple:
alt_dcs flag is used to define either 0xFX coding (alt_dcs=1)
or 0x0X coding
(alt_dcs=0). alt_dcs may be SMS_PARAM_UNDEF==-1 only if
message was generated
on the extern interface (e.g. sendsms in smsbox) and user has
not supplied
any dcs values.
Kita B. Ndara wrote:
I was having problems sending WAP POST messages
larger than 8kb or so. I realised after some poking
around that Wapbox kept reporting a Timeout-A event.
More digging turned up fact that this is controlled to
an extent by the wapbox conf timer-freq. Once I set
this
Vjacheslav Chekushin wrote:
Have somebody tried OMA bootstrap provisioning?
I've sent configuration to phone using USERPIN method successfully, but
can't do the same with NETWPIN.
I'm using semi-octet encoded IMSI as key:
123456789012345 = (hex encoded string, length 8) 21 43 65 87 09 21 43
[EMAIL PROTECTED] wrote:
Hi List,
attached you can find a patch that implements UUID (Universal Unique Id)
support for kannel. The uuid functions where borrowed from libuuid
(http://e2fsprogs.sourceforge.net) that implements DEC compatible UUIDs (see
definetly +1, applying patch (including gw/smsc_fake.c fix).
Stipe
mailto:[EMAIL PROTECTED]
---
Wapme Systems AG
Münsterstr. 248
40470 Düsseldorf, NRW, Germany
phone: +49.211.74845.0
fax: +49.211.74845.299
mailto:[EMAIL
Hi all,
any results on this issue?!
Paul, I'll try to re-create the behaviour in our smppbox and claim
your rights at Alex if we definetly can't handle the values
transparently allong Kannel's Msg structure.
Stipe
mailto:[EMAIL PROTECTED]
Igor Ivoilov wrote:
I can not find in wapbox access log file records of failed attemps
to fetch the given url.
Let's say in a case when this url is not available from the wap
gateway.
Is it done on purpose? Is there a way to configure wapbox to log it?
actualy it is not a purpose. But it's
Hi all,
any update on this issue?!
Stipe
mailto:[EMAIL PROTECTED]
---
Wapme Systems AG
Münsterstr. 248
40470 Düsseldorf, NRW, Germany
phone: +49.211.74845.0
fax: +49.211.74845.299
mailto:[EMAIL PROTECTED]
Igor Ivoilov wrote:
Any suggestions why this would be semantically necessary to log to
access-log?
For me access log shows users activity. Even if they are not able to get something
I have to see what they have tried to access.
that's the question.
If you consider the Apache httpd,
Hi list,
Alex has introduced the new UUIDs for message IDs and there are some
sections marked with #warning's in gw/smsc/smsc_soap.c, which need to
be resolved.
Who is currently actively using the SOAP SMSC module? It is yet not
bound to the SMSCConn abstraction layer. He/she should resolve the
I have just buy a Wavecom Fastrack M1206B modem,
but kannel doesn't arrive to detect him :
[5] DEBUG: AT2[test]: -- AT^M
2003-12-09 18:28:54 [5] DEBUG: AT2[test]: -- AT^M
2003-12-09 18:28:58 [5] ERROR: AT2[test]: no answer from modem
2003-12-09 18:28:58 [5] ERROR: AT2[test]: Opening failed.
there is no PANIC in the wapbox.log, so the wapbox process does
SEGFAULT?!
Can you please generate a core dump file and analyze it with gdb to
see at which position the segmentation fault happens?!
I can try to reproduce the crash on our side meanwhile...
Stipe
mailto:[EMAIL PROTECTED]
Here is gdb output:
Core was generated by `/usr/local/kannel-rad2/sbin/wapbox -v 0 /etc/
kannel2.conf'.
Program terminated with signal 9, Killed.
#0 0xff033084 in ?? ()
did you include debug symbols into the build?
If possible, recompile using the following additional configure
switch:
Vjacheslav Chekushin schrieb:
Hi, Stipe
No, it doesn't PANIC. It simply doesn't serve request because think
that client doesn't support wmlc.
To be correct WAP gw send Not Acceptable response.
ok, checking the sources...
Stipe
mailto:[EMAIL PROTECTED]
Alexander Malysh schrieb:
have you tried w/ debug options (cflags: -g ) on solaris ? if yes, what say
the gdb to coredump?
just guess: if it works for linux but not for solaris, then is extern library
to blame...
Alex is correct here.
If you have only libxml2 as direct dependency
can you point us directly to the section of the WSP spec where you
see that a WSP header may be empty, hence with 0x00 as value?!
Stipe
mailto:[EMAIL PROTECTED]
---
Wapme Systems AG
Münsterstr. 248
40470 Düsseldorf, NRW, Germany
Bruno Rodrigues wrote:
Stipe Tolj [EMAIL PROTECTED] wrote:
Bruno Rodrigues wrote:
you get a 5xx gateway timeout if you can't reach the other server
so you're voting +1 in getting non-access error codes to access-log??
Yep, +1 for Gateway Timeouts (no reply) and other errors like
Hi list,
here is a patch that will allow to log 502 HTTP response code for
proxied WAP GW requests that did not have any internal HTTP response
code, ie. wapbox could not connect to the port or host was not
reachable.
Please try and vote on commitment.
Stipe
mailto:[EMAIL PROTECTED]
201 - 300 of 1437 matches
Mail list logo
