Re: gw_push_pap_compiler.c : bug fixes propositions
Hi Bernard, Aarno, I've started testing the push part of Kannel and I found bugs in pap_compiler.c. I'd like to support WAP Push testing since we have several Ericsson R520m available which are WAP 1.2 Push compliant and I have seen using Materna WAP Gateway that they can receive WAP Push messages. I suppose we need to WBXML encode a XML application which represents the supported content-type and send it via Kannel as SMS Gateway to the phone. Aarno, I hope you remember the request I have made on how to test this using Kannel and Ericsson R520m. I made already a couple of tries sending these WBXML format messages, but _not_ one was interpreter by the phone correctly. Can you guys guild me in how I can help you and us in getting WAP Push operatibility done with Kannel? Thanks in advance. See ya, Stipe Stipe Tolj Department Management Technology Center Research Lab [EMAIL PROTECTED] --- Wapme Systems AG Münsterstr. 248 40470 Düsseldorf Tel: +49-211-74845-0 Fax: +49-211-74845-299 E-Mail: [EMAIL PROTECTED] Internet: http://www.wapme-systems.de --- wapme.net - wherever you are
usage of test_ppg
Hi, what is the sence of the push_url within the usage of test_ppg? I see from test_ppg.c that http_start_request(caller, push_url, ...) is called. And then within reveive_push_reply() the HTTP reply is checked. So what happens on the HTTP server side? I was thinking of Kannel (as PPG) transmitting the Push Content (i.e. hardcoded using -H switch) to the handset?! I need some explaination on that, since I would like to start testing and advancing the Push implementation within Kannel. Thanks in advance. Stipe [EMAIL PROTECTED] --- Wapme Systems AG Münsterstr. 248 40470 Düsseldorf Tel: +49-211-74845-0 Fax: +49-211-74845-299 E-Mail: [EMAIL PROTECTED] Internet: http://www.wapme-systems.de --- wapme.net - wherever you are
[RFC] mysql interface to kannel?
Is there any interest that we may start implementing a DB based interface to Kannel for various functions. This was on my TODO list for quite a time and I would like to kick-off development on that. We have incorporated MySQL client functions to Kannel within our MSISDN Provisioning concept, but there are way more use for storing or retrieving data to a database from Kannel: - logging facility: access.log data may be stored to mysql db, especialy for automatic billing support. - configuration facility: multi-groups like sms-user, sms-service, ota-config may be stored in mysql db, so there may be added or manipulated while runtime and the service would not have to be restarted to get notice of an update. This would make thinks particular slower, but may add incredible functionality gains. I think of a SMS gateway hosting service, where customers get an HTTP based Account and log into a HTTP server (using PHP and MySQL) and configure the way Kannel sends their SMS messages via EMI or other SMSC protocols (more or less) individual, including incoming SMS Services. Of course you have to keep certain options under restriction so that you have control of the system and billing. If there is interest in incorporating (at least in the first step) MySQL interface for Kannel I may get my hands on the above issues. Stipe [EMAIL PROTECTED] --- Wapme Systems AG Münsterstr. 248 40470 Düsseldorf Tel: +49-211-74845-0 Fax: +49-211-74845-299 E-Mail: [EMAIL PROTECTED] Internet: http://www.wapme-systems.de --- wapme.net - wherever you are
Re: Starting 1.1.5 : Problem with shared libraries.
I'am trying to use kannel-1.1.5. I get an error at startup: ./bearerbox: error while loading shared libraries: cannot open shared object file: cannot load shared object file: No such file or directory Kannel-1.0.3 is working and I have configured both files with the same .configure options. Please, could someone explain me what I should do to get it working? I would suggest there is something wrong on your OS iteself. Stipe [EMAIL PROTECTED] --- Wapme Systems AG Münsterstr. 248 40470 Düsseldorf Tel: +49-211-74845-0 Fax: +49-211-74845-299 E-Mail: [EMAIL PROTECTED] Internet: http://www.wapme-systems.de --- wapme.net - wherever you are
Re: Siemens M20 and ringtones
I've just sent this PDU with my Nokia 7110 and it works! Have you already tested the new firmware? Or is anybody able to send ringtones with Siemens M20? yes, we did this for quite a while, but using a Windows NT based GSM-Modem driver, _not_ Kannel. Stipe [EMAIL PROTECTED] --- Wapme Systems AG Münsterstr. 248 40470 Düsseldorf Tel: +49-211-74845-0 Fax: +49-211-74845-299 E-Mail: [EMAIL PROTECTED] Internet: http://www.wapme-systems.de --- wapme.net - wherever you are
Re: deferred databases...
As you can see every item has other needs. I think we should have two abstraction layers here. One which is representing a database and another which is representing the interface between the internal services going to either a config file or to a database interface which then in turn can be MySQL or mSQL or Postgres or whatever. The interface layer would map field names to the data where the database layer would map the calls to the actual database engine. I agree to the two level abstraction layer. Stipe [EMAIL PROTECTED] --- Wapme Systems AG Münsterstr. 248 40470 Düsseldorf Tel: +49-211-74845-0 Fax: +49-211-74845-299 E-Mail: [EMAIL PROTECTED] Internet: http://www.wapme-systems.de --- wapme.net - wherever you are
sendota over emi2?!
Hi, is anyone sending WAP OTA config SMS messages using the sendota facility with a development version later than 1.1.5 or CVS snapshot via emi2 or AT? We still have conciderable errrors with that, see also my previous posting BUG: SMS send using smsbox.c:sendota() are corrupted from Tue, 17 Jul 2001 19:31:24 +0200 Regards, Stipe [EMAIL PROTECTED] --- Wapme Systems AG Münsterstr. 248 40470 Düsseldorf Tel: +49-211-74845-0 Fax: +49-211-74845-299 E-Mail: [EMAIL PROTECTED] Internet: http://www.wapme-systems.de --- wapme.net - wherever you are
Re: kannel on cygwin
I'm trying to compile and run kannel 1.1.5 on cygwin. Everything compiles well but when I try to run the bearerbox it says: 2001-09-04 09:13:43 [1] PANIC: gwlib/gwthread-pthread.c:151: getthreadinfo: Assertion `pthread_equal(threadinfo-self, pthread_self())' failed. I've tried to uncomment line 151 in file gwlib/gwthread-pthread.c which looks like this: gw_assert(pthread_equal(threadinfo-self, pthread_self())); If I do that then I can start the bearerbox and the smsbox but I can only send one SMS. If I try to send another one it freezes. Any ideas how this line can be changed to work on cygwin? I had the same problem. To me it looks like the assert is failing even the threadinfo-self and pthread-self are the same. So pthread_equal has a bug on cygwin or something like that. pthreads are still considered not-fully supported on Cygwin. I would suggest you to use a special pthread-win32 package I arranged for Cygwin 1.1.8, but it works with slight modifications to its pthread.h for 1.3.2 too. Usually you can find the tarball on our Cygwin machine http://kannel.dev.wapme.net/ But the machine is down for hardware and OS upgrade. It will be back-up again tomorrow. Check at least this week-end. Mail me if you have any further questions. Stipe [EMAIL PROTECTED] --- Wapme Systems AG Münsterstr. 248 40470 Düsseldorf Tel: +49-211-74845-0 Fax: +49-211-74845-299 E-Mail: [EMAIL PROTECTED] Internet: http://www.wapme-systems.de --- wapme.net - wherever you are
Push SI and SL work for Ericsson R520m
Hi list, this is a short note to let you (especially Aarno and Tobias) know, that we made the Push content-types session indication (SI) and session loading (SL) run on the GPRS-enabled (still prototype?!) WAP 1.2 mobile phone, the Ericsson R520m. BTW, the Nokia 9210 does non of them. While SI works quite straight forward. SL does it even better: the device receives the push messages and starts automaticly (without asking the user) to dial-in for the requesting URL. This may be a great feature to use location based detection or person tracking in my mind?!?! Know we should be doing the work inside Kannel to make the pushing more automated, especially for SMS based delivery. Thanks again to Aarno, Tobias and Jochen. Stipe [EMAIL PROTECTED] --- Wapme Systems AG Münsterstr. 248 40470 Düsseldorf Tel: +49-211-74845-0 Fax: +49-211-74845-299 E-Mail: [EMAIL PROTECTED] Internet: http://www.wapme-systems.de --- wapme.net - wherever you are
Re: DLR support
this is a unique ID your application specifies. It will be sent back to you when the delivery report arrives so it should be something like a message numer or so. How do you ensure unique IDs across applications ? ie. What if application A and application B are both using the same sendsms interface and are specifying identical IDs ? a kludge may be a central ID repository (aka MySQL DB) which holds all DLR specific IDs accross all applications. Stipe Tolj Department Management Technology Center Research Lab [EMAIL PROTECTED] --- Wapme Systems AG Münsterstr. 248 40470 Düsseldorf Tel: +49-211-74845-0 Fax: +49-211-74845-299 E-Mail: [EMAIL PROTECTED] Internet: http://www.wapme-systems.de --- wapme.net - wherever you are
[REQ] bug within emi2 module some days ago?!
Hi guys, sorry for the stupid question, but our admins here messed up my mail account and I did not receive [EMAIL PROTECTED] for about 3 weeks. While trying a CVS snapshot to see how Andreas Fink's DLR works with emi2 I noticed that there was still a bug in the CVS concerning an assertion problem. Has this been reported and fixed yet? Regards, Stipe [EMAIL PROTECTED] --- Wapme Systems AG Münsterstr. 248 40470 Düsseldorf Tel: +49-211-74845-0 Fax: +49-211-74845-299 E-Mail: [EMAIL PROTECTED] Internet: http://www.wapme-systems.de --- wapme.net - wherever you are
Re: HTTP SMSC
Hi Aaeno, Push Proxy Gateway uses http smsc in a special way. A HTTP SMSCnbsp; generatesnbsp; following HTTP get request when it receives a sendmessage event: http://hostname:port/path?user=foopassword=barto=numbertext=%02%06%17%ae %92localhost%3a8080%00%af%80amp;8d%cf%b4%80%02%05j%00e%c6%0c%03wap.iobox.fi %00%11%031%40wiral.com%00%07%0a%c3%07%19%99%06%25%15%23%15%10%c3 %04+%02%060%01%03Want+to+test+the+fetch%3f%00%01%01udh=%06%05%04%0b %84%23%f0 I'm still having problems on this. What I try is to use test/test_ppg.c to send a WML deck to a Ericsson R520m via SMS bearer. I changed test_ppg.c to have: address address-value=WAPPUSH=+491735333216/TYPE=PLMN /address When I call $ ./text_ppg -H http://localhost:8080/cgi-bin/wap-push.cgi I get push failed. Is Kannels PPG detecting if a push is considered to be made over SMS as bearer or not? I dont't see any relevant code within the wap_push_* and wap_ppg_* files? An detailed setup in how you did it to make a SL or SI routed over SMS bearer would be realy helpfull. Stipe [EMAIL PROTECTED] --- Wapme Systems AG Münsterstr. 248 40470 Düsseldorf Tel: +49-211-74845-0 Fax: +49-211-74845-299 E-Mail: [EMAIL PROTECTED] Internet: http://www.wapme-systems.de --- wapme.net - wherever you are
Re: usage of smsc_http.c
An example would be an SMS aggregator (a virtual SMSC) that provides a http interface to their system. There are a number of companies that provide a http interface for sending SMS messages and a company may wish to connect to one of these systems. I can't think of any of these offhand but I am sure a web search would find you some. ok, so someone calls your SMSC HTTP via HTTP and you call SMSC EMI to your provider as example?! How would the setups look like for your client and yourself? Stipe [EMAIL PROTECTED] --- Wapme Systems AG Münsterstr. 248 40470 Düsseldorf Tel: +49-211-74845-0 Fax: +49-211-74845-299 E-Mail: [EMAIL PROTECTED] Internet: http://www.wapme-systems.de --- wapme.net - wherever you are
anyone using EMI2 with DLR?
Hi, as I reported previously we had huge problems with latest CVS snapshot for emi2. There were assertion errors after using /cgi-bin/sendsms and both smsbox and bearerbox died. I located the problem to be in: gw/smsc_emi2.c:696: i = octstr_search_char(ts,':',0); so it seems when ts == NULL we get into truble here?! BTW, DLR reports don't work either for me. calling http://host:13013/cgi-bin/sendsms?...dlrmask=3dlrid=xyz does not activate the specified sms-service within conf file. Is anyone using this or has similar problems? It may depend on the above assertion problem? Stipe [EMAIL PROTECTED] --- Wapme Systems AG Münsterstr. 248 40470 Düsseldorf Tel: +49-211-74845-0 Fax: +49-211-74845-299 E-Mail: [EMAIL PROTECTED] Internet: http://www.wapme-systems.de --- wapme.net - wherever you are
Re: anyone using EMI2 with DLR?
Ok, if you got this problem then you most probably dont connect to a CMG SMSC but something emulating it or a CMG with an old release. The thing it does in this line is reading the timestamp information out which the SMSC gives back on a request. Apparently yours is empty or such. In my case this is something like timestampinformation:phonenumber and the code separates it to keep track of it. Of course if you get nothing back, its clear that delivery reports cant work for you. and what about the failed assertions that crash smsbox and then bearerbox after one SMS has been processed over emi2? This is definitly some kind of buggy. Stipe [EMAIL PROTECTED] --- Wapme Systems AG Münsterstr. 248 40470 Düsseldorf Tel: +49-211-74845-0 Fax: +49-211-74845-299 E-Mail: [EMAIL PROTECTED] Internet: http://www.wapme-systems.de --- wapme.net - wherever you are
Re: Query regarding Billing
Are there any interfaces that Kannel WAP gateway provides for any third party billing Engine? If so can anyone provide the details of the same(which are these interfaces? What is the format in which traffic data is provided etc?) there are basic constructs for this within Kannel. At least logging facilities that can be expanded. Has somebody already integrated kannel WAP gateway with any billing engine? Please let me know the details. Yes we have. We have a MySQL solution which is used for various reasons with Kannel, i.e. MSISDN provisioning and accounting. It's considered to be very tailored and yet not publicly available. Also are there any plans to implement new WAP billing Framework by Wapforum in kannel wap gateway? Currently billing (what is concerned about WAP anyway) is used in the following manner: a) in case of CSD (GSM): your on-the-air time is beeing charged. b) in case of GPRS: your traffic volume is been charged. I suppose you need some kind of billing facility as third-party provider, an ASP or something like that? Regards, Stipe Tolj Department Management Technology Center Research Lab [EMAIL PROTECTED] --- Wapme Systems AG Münsterstr. 248 40470 Düsseldorf Tel: +49-211-74845-0 Fax: +49-211-74845-299 E-Mail: [EMAIL PROTECTED] Internet: http://www.wapme-systems.de --- wapme.net - wherever you are
Re: kannel language support
Does kannel support messages in different languages? if you consider messages within SMS, yes. if you consider warning messages and everything Kannel internal, no -- currently we develop in English. But there should be an internationalization effort made here, IMO. Stipe [EMAIL PROTECTED] --- Wapme Systems AG Münsterstr. 248 40470 Düsseldorf Tel: +49-211-74845-0 Fax: +49-211-74845-299 E-Mail: [EMAIL PROTECTED] Internet: http://www.wapme-systems.de --- wapme.net - wherever you are
Re: Siemens M20 and Kannel
I'm trying to get Kannel to use a Siemens M20. The M20 is plugged into COM1 and if I connect using minicom I can send AT commands to the M20, and can send and read SMS messages. In the kannel.conf file I have: group = smsc smsc = at modemtype = siemens device = /dev/ttyS0 pin = 2004 And I when I start up the bearerbox I get this: 2001-10-18 14:43:37 [0] INFO: Debug_lvl = -1, log_file = none, log_lvl = 0 2001-10-18 14:43:37 [0] INFO: Reading configuration from /etc/kannel/smskannel.conf 2001-10-18 14:43:37 [0] INFO: 2001-10-18 14:43:37 [0] INFO: Kannel bearerbox II version 1.0.3 starting 2001-10-18 14:43:37 [0] DEBUG: Started thread 1 (gw/bb_http.c:httpadmin_run) Command: ATE0 2001-10-18 14:43:37 [1] DEBUG: HTTP: Creating a new server socket 13000. 2001-10-18 14:43:47 [0] PANIC: Cannot start with SMSC connection failing Can anyone give me any pointers as to what I might be doing wrong? yes, this effect is known for GSM modems connected to serial com lines. I can't tell you exactly why Kannel does not get the hook to the device at once, but we also have to fire up bearerbox at least 2 times before it recognizes the device at a serial com line. Simply retry a second or third time in the shell and you should get a hook. Stipe [EMAIL PROTECTED] --- Wapme Systems AG Münsterstr. 248 40470 Düsseldorf Tel: +49-211-74845-0 Fax: +49-211-74845-299 E-Mail: [EMAIL PROTECTED] Internet: http://www.wapme-systems.de --- wapme.net - wherever you are
Re: AW: Kannel developer meeting
Does the resounding silence mean that there is no interest? from our side we have a _HUGE_ interest for several reasons. Stipe [EMAIL PROTECTED] --- Wapme Systems AG Münsterstr. 248 40470 Düsseldorf Tel: +49-211-74845-0 Fax: +49-211-74845-299 E-Mail: [EMAIL PROTECTED] Internet: http://www.wapme-systems.de --- wapme.net - wherever you are
Re: Still WAP Push
Hi Aarno, T39m works, too. I will test Nokia 8310 soonish. has it been released yet? Stipe [EMAIL PROTECTED] --- Wapme Systems AG Münsterstr. 248 40470 Düsseldorf Tel: +49-211-74845-0 Fax: +49-211-74845-299 E-Mail: [EMAIL PROTECTED] Internet: http://www.wapme-systems.de --- wapme.net - wherever you are
Re: Delivery Report Failures...
I'll check that. Actually there's a way to automatically reconnect if the connection fails. That's probably better. Mixing memory based and MySQL based is not a good idea. Kannel would always have to search the two. mysql_ping() is used to be checking the connection and re-establish it if it has been droped. We use it in our own MSISDN Provisioning extension for Kannel. Stipe [EMAIL PROTECTED] --- Wapme Systems AG Münsterstr. 248 40470 Düsseldorf Tel: +49-211-74845-0 Fax: +49-211-74845-299 E-Mail: [EMAIL PROTECTED] Internet: http://www.wapme-systems.de --- wapme.net - wherever you are
Re: Delivery Report Failures...
Also I've modified delivery reports so they can optionally work with MySQL instead of being memory based. To make this work, you got to modify the Makefile to include the mysqlclient library and you got to edit dlr.c to define MYSQL_DLR and to set your host's username passwords. shouldn't we have a configure switch on that feature? Stipe [EMAIL PROTECTED] --- Wapme Systems AG Münsterstr. 248 40470 Düsseldorf Tel: +49-211-74845-0 Fax: +49-211-74845-299 E-Mail: [EMAIL PROTECTED] Internet: http://www.wapme-systems.de --- wapme.net - wherever you are
Re: Siemens M20 and Kannel
I've tried restarting it about 10 times and unfortunately it hasn't made any difference. Does it matter if the smsbox is or isn't running? no it doesn't matter. BTW, you won't be able to start smsbox without previously running bearerbox :) Stipe [EMAIL PROTECTED] --- Wapme Systems AG Münsterstr. 248 40470 Düsseldorf Tel: +49-211-74845-0 Fax: +49-211-74845-299 E-Mail: [EMAIL PROTECTED] Internet: http://www.wapme-systems.de --- wapme.net - wherever you are
Re: Delivery Report Failures...
Also I've modified delivery reports so they can optionally work with MySQL instead of being memory based. To make this work, you got to modify the Makefile to include the mysqlclient library and you got to edit dlr.c to define MYSQL_DLR and to set your host's username passwords. shouldn't we have a configure switch on that feature? of course :-). the stuff belongs into kannel.conf and a configure --with-mysql-dr should be added. maybe we should add this using a type sensitive switch, like --with-dlr=type type: mysql sdbm ... we make this for the MSISDN provisioning too. This way the user has to take care only once which kind of DLR or provisioning type he requires. Stipe [EMAIL PROTECTED] --- Wapme Systems AG Münsterstr. 248 40470 Düsseldorf Tel: +49-211-74845-0 Fax: +49-211-74845-299 E-Mail: [EMAIL PROTECTED] Internet: http://www.wapme-systems.de --- wapme.net - wherever you are
[RFC] ppg-deny-ip and ppg-allow-ip for PPGs HTTP server
Hi Aarno, are there ppg-deny-ip and ppg-allow-ip ranges possible for the PPG itself? I suggest this would be quite usefull when not explicitely using user credentials. Comments from the list?! Stipe [EMAIL PROTECTED] --- Wapme Systems AG Münsterstr. 248 40470 Düsseldorf Tel: +49-211-74845-0 Fax: +49-211-74845-299 E-Mail: [EMAIL PROTECTED] Internet: http://www.wapme-systems.de --- wapme.net - wherever you are
BUG in PPG when sending malformed PAP document
Hi, I found this bug recently by patching the hard coded PAP document to use a SI instead of the WML deck in test/test_ppg.c. To support SI/SL over SMS I changed the following: [...] quality-of-service [...] network-required=true network=GSM and by mistake added the following: bearer-required=true // BUG! bearer=SMS /quality-of-service [...] which is obviously not a valid XML structured document! Running $ ./test_ppg -H http://localhost/cgi-bin/wap-push.cgi caused wapbox to panic and crash with the following wapbox.log entry: [...] 2001-10-25 17:02:49 [11] DEBUG: HTTP: Creating HTTPClient for `127.0.0.1'. 2001-10-25 17:02:49 [13] INFO: PPG: Accept request /cgi-bin/wap-push.cgi from 127.0.0.1 2001-10-25 17:02:49 [13] DEBUG: PPG: http_read_thread: pap multipart accepted 2001-10-25 17:02:49 [13] PANIC: gwlib/octstr.c:2032: seems_valid_real: Assertion `ostr != NULL' failed. (Called from gwlib/octstr.c:728:octstr_compare.) [...] so obviously we do not parse the PAP document very good here?! Stipe [EMAIL PROTECTED] --- Wapme Systems AG Münsterstr. 248 40470 Düsseldorf Tel: +49-211-74845-0 Fax: +49-211-74845-299 E-Mail: [EMAIL PROTECTED] Internet: http://www.wapme-systems.de --- wapme.net - wherever you are
Re: Production Kannel
CVS is good choice right now. I run it fully productive and have no issues with it. Some newer features might not work as expected or so but the old stuff is very solid. I agree here to Andreas. We have CVS systems running very stable currently. Your'e asking for a lot of fun :-) so let's have it! :) Stipe [EMAIL PROTECTED] --- Wapme Systems AG Münsterstr. 248 40470 Düsseldorf Tel: +49-211-74845-0 Fax: +49-211-74845-299 E-Mail: [EMAIL PROTECTED] Internet: http://www.wapme-systems.de --- wapme.net - wherever you are
Re: For your kind attention!
I am presently setting up Kannel to run as an SMS gateway using compiled version of 1.0.3. I am using a siemens GSM modem as the SMSC. With my present configurations, I can send messages through my smsc from the admin html console via a browser. But messages sent to me are not received and no indications of these are made in any form, even in the logs. If you are running an older firmware version on your Siemens M20 then this may be the cause. Please search the mailing list archive for an update location someone posted recently, where you can get the update for the firmware software. Anyway, 1.0.3 is quite old, considered in terms of smsc_at.c (the GSM modem SMSC implementation). You should be using a CVS snapshot instead. Unfortunatly we currently don't have any latest stable release, so CVS is appropriate for now. I am not receiving messages via the smsc. Is there a specific receive-thread that must be started before messages can be recieved? No, bearerbox is responsible for serial com communication and smsbox for the processing. Does the bearerbox listen continuosly to the smsc and is there a way I can dictate the frequency at which the bearerbox retrieves messages from the smsc(phone modem). No, bearerbox does (or should :)) listen continously for incomding messages. Can you send as much messages as you want via the SMSC, but don't receive any? Or may it be that you send a couple of messages, the SMSC connection drops and you try to receive and wonder why it does not come? Anyway, a bearerbox and smsbox log (not to heavy please :)) would be best for the list to review what you are trying. Stipe [EMAIL PROTECTED] --- Wapme Systems AG Münsterstr. 248 40470 Düsseldorf Tel: +49-211-74845-0 Fax: +49-211-74845-299 E-Mail: [EMAIL PROTECTED] Internet: http://www.wapme-systems.de --- wapme.net - wherever you are
Re: nokia 6210 win2k server?
does the nokia 6210 work with windows 2k server via a Data cable? yes, it should. Stipe [EMAIL PROTECTED] --- Wapme Systems AG Münsterstr. 248 40470 Düsseldorf Tel: +49-211-74845-0 Fax: +49-211-74845-299 E-Mail: [EMAIL PROTECTED] Internet: http://www.wapme-systems.de --- wapme.net - wherever you are
Re: Failover in Kannel.
Is there any possibility to make failover in Kannel with two seperate computers with Kannel on each ? as Paul pointed out, Kannel has no internal constructs to do so. But you may use constructs that are based on a lower level, i.e. you may use a Linux Cluster using the Linux Virtual Server Project at http://www.linuxvirtualserver.org/. We have such a clustered version running. But beware, you will have to do some overhead to garantee that the Kannel's know about eachother and act accordingly. Stipe [EMAIL PROTECTED] --- Wapme Systems AG Münsterstr. 248 40470 Düsseldorf Tel: +49-211-74845-0 Fax: +49-211-74845-299 E-Mail: [EMAIL PROTECTED] Internet: http://www.wapme-systems.de --- wapme.net - wherever you are
Re: SEGV on program termination
am I the onliest one who sees regular crashes on program termination (all the boxes)? They are hard to track down (gdb is useless, no core dumps) so I'd like to hear other ideas. I had such effect on single boxes, but it was related to faulty configuration file. Stipe [EMAIL PROTECTED] --- Wapme Systems AG Münsterstr. 248 40470 Düsseldorf Tel: +49-211-74845-0 Fax: +49-211-74845-299 E-Mail: [EMAIL PROTECTED] Internet: http://www.wapme-systems.de --- wapme.net - wherever you are
Re: REPOST : Problem with emi2 keepalive ?
If you look in the online documentation it doesn't say anything about a configuration option called keepalive. Maybe you're not able to set it. please have always a look in the CVS generated documentation, not the old 1.1.5 version. Stipe [EMAIL PROTECTED] --- Wapme Systems AG Münsterstr. 248 40470 Düsseldorf Tel: +49-211-74845-0 Fax: +49-211-74845-299 E-Mail: [EMAIL PROTECTED] Internet: http://www.wapme-systems.de --- wapme.net - wherever you are
Re: [RFC] changing openssl detection in configure.in
Ok would you mind adding the MacOS X issues I've posted as well? I'm not a frequent configure.in hacker that I can be sure that it works correctly. Can you please state which post you mean? Hmm. be aware that the binaries names will change due to that (example: /usr/local/sbin/bearerbox-cvs-20011120). I would suggest that the latest one is symbolic linked to /usr/local/sbin/bearerbox-cvs to remain compatible with startup scripts. Otherwhise we have to change those every time we update cvs. Are you sure? configure won't change the contents of the file VERSION. It simply attaches the date to the #def VERSION inside config.h. Stipe [EMAIL PROTECTED] --- Wapme Systems AG Münsterstr. 248 40470 Düsseldorf Tel: +49-211-74845-0 Fax: +49-211-74845-299 E-Mail: [EMAIL PROTECTED] Internet: http://www.wapme-systems.de --- wapme.net - wherever you are
Re: [RFC] changing openssl detection in configure.in
The Makefile created should not run ar rc but libtool -static -o instead so you have to garantee libtool is available, which is not tested while configure :( What's the problem wit ar rc? Furthermore the following had to be replaced (safe for all platforms I presume): replace wmlscriptobjs = $(wmlscriptsrcs:.c=.o) $(wmlscriptsrcs:.y=.o) with wmlscriptobjs = $(wildcard wmlscript/*.o) ??? Stipe [EMAIL PROTECTED] --- Wapme Systems AG Münsterstr. 248 40470 Düsseldorf Tel: +49-211-74845-0 Fax: +49-211-74845-299 E-Mail: [EMAIL PROTECTED] Internet: http://www.wapme-systems.de --- wapme.net - wherever you are
Re: [RFC] changing openssl detection in configure.in
Andreas Fink schrieb: The Makefile created should not run ar rc but libtool -static -o instead so you have to garantee libtool is available, which is not tested while configure :( MacOS X has it available. Thats for sure. ar didnt work because some objects are multiple defined in libraries and linking fails because of that. libgw.a for example is created out of all *.c in gw directory. This means there are multiple mains defined. Any objections from the others to change this? Votings please. I would suggest to include or at least test against the presense of libtool. What's the problem wit ar rc? Furthermore the following had to be replaced (safe for all platforms I presume): replace wmlscriptobjs = $(wmlscriptsrcs:.c=.o) $(wmlscriptsrcs:.y=.o) with wmlscriptobjs = $(wildcard wmlscript/*.o) ??? Thats in the created Makefile. make on MacOS X apparently doesnt understand the .c = .o syntax. So why does it understand for the other src dirs?! Is this only specific to wmlscriptobjs? Any objections here from the developers? Stipe [EMAIL PROTECTED] --- Wapme Systems AG Münsterstr. 248 40470 Düsseldorf Tel: +49-211-74845-0 Fax: +49-211-74845-299 E-Mail: [EMAIL PROTECTED] Internet: http://www.wapme-systems.de --- wapme.net - wherever you are
Re: multi-cast implementation
Hi Valter, someone can tell me if multi-cast feature is already implemented in smsbox??? anyone is working on that??? by multi-cast i mean send multiple sms messages using only one http post where the X-Kannel-To header contains multiple receivers! I don't know if the POST method version has been implemented yet, but you may use the GET method using a phone number list for the to argument, see http://www.kannel.3glab.org/download/kannel-userguide-snapshot/userguide.html#AEN2464 Stipe [EMAIL PROTECTED] --- Wapme Systems AG Münsterstr. 248 40470 Düsseldorf Tel: +49-211-74845-0 Fax: +49-211-74845-299 E-Mail: [EMAIL PROTECTED] Internet: http://www.wapme-systems.de --- wapme.net - wherever you are
Re: smsbox:charset_processing called twice
Index: gw/smsbox.c
===
RCS file: /home/cvs/gateway/gw/smsbox.c,v
retrieving revision 1.156
diff -r1.156 smsbox.c
1392,1395d1391
if (charset_processing(charset, body, coding) == -1) {
*status = 415;
ret = octstr_create(Charset or body misformed, rejected);
}
votings from the smsbox hackers for the proposed change?! Andreas?
Nick?
Stipe
[EMAIL PROTECTED]
---
Wapme Systems AG
Münsterstr. 248
40470 Düsseldorf
Tel: +49-211-74845-0
Fax: +49-211-74845-299
E-Mail: [EMAIL PROTECTED]
Internet: http://www.wapme-systems.de
---
wapme.net - wherever you are
Re: memory allocation problem with Kannel
it is givin' this problem when the number of records are reaching the maximum value given by MAX_ALLOCATIONS in the gw_mem_check.c file. so what is the load (i.e. number of incoming requests) when this happens? Stipe [EMAIL PROTECTED] --- Wapme Systems AG Münsterstr. 248 40470 Düsseldorf Tel: +49-211-74845-0 Fax: +49-211-74845-299 E-Mail: [EMAIL PROTECTED] Internet: http://www.wapme-systems.de --- wapme.net - wherever you are
Re: [RFC] changing openssl detection in configure.in
The Makefile created should not run ar rc but libtool -static -o instead Furthermore the following had to be replaced (safe for all platforms I presume): replace wmlscriptobjs = $(wmlscriptsrcs:.c=.o) $(wmlscriptsrcs:.y=.o) with wmlscriptobjs = $(wildcard wmlscript/*.o) I'm noticing similar problems for current CVS on Solaris 2.6. Anyone on Sun having the same problems?! Stipe [EMAIL PROTECTED] --- Wapme Systems AG Münsterstr. 248 40470 Düsseldorf Tel: +49-211-74845-0 Fax: +49-211-74845-299 E-Mail: [EMAIL PROTECTED] Internet: http://www.wapme-systems.de --- wapme.net - wherever you are
Re: [RFC] changing openssl detection in configure.in
I'm noticing similar problems for current CVS on Solaris 2.6. one step back, this was a too early shot. The problem was solaris' make, gmake works fine. Stipe [EMAIL PROTECTED] --- Wapme Systems AG Münsterstr. 248 40470 Düsseldorf Tel: +49-211-74845-0 Fax: +49-211-74845-299 E-Mail: [EMAIL PROTECTED] Internet: http://www.wapme-systems.de --- wapme.net - wherever you are
Re: First version of PPG committed
I committed first version of WAP Push Proxy Gateway. This works with SMS bearer, though recently only with HTTP smsc. This means that it translates PAP protocol requests to HTTP requests. All sendsms interfaces should understand these. It is tested with Ericsson T39m mobile phone. Of course test results and comments are very wellcome. we'll have a shot on that ASAP and report. Thanks for the great work. Stipe [EMAIL PROTECTED] --- Wapme Systems AG Münsterstr. 248 40470 Düsseldorf Tel: +49-211-74845-0 Fax: +49-211-74845-299 E-Mail: [EMAIL PROTECTED] Internet: http://www.wapme-systems.de --- wapme.net - wherever you are
Re: SMSGateway
(I) Is it possible to run SMS Box independently ( not with the bearer box thread, because we don't want WAP box). Kannel is an SMS and WAP gateway. This fact should be considered as non-exklusive AND, which means you can run Kannel as SMS and WAP gateway (both at once), or one of them, i.e. SMS or WAP. Bearerbox is required, due to the fact that it receives all incoming transactions. (ii)Whether this SMS gateway will support POP3 for message transmission, and SMTP for message reception ? If yes then what are the changes we need to do. Yes, but not natively. This means you can use Web Applications to do POP3 and SMTP work seperatly. Use seperate sms-service setting to trigger an URL for your aims. As we made Kannel mostly independent to specific application (or middleware) protocolls, which except to HTTP, I don't consider a native POP3 and/or SMTP support will be incorporated very soon. Use instead the ability of HTTP to do your POP3 and SMTP effort within a Web Application. Regards, Stipe [EMAIL PROTECTED] --- Wapme Systems AG Münsterstr. 248 40470 Düsseldorf Tel: +49-211-74845-0 Fax: +49-211-74845-299 E-Mail: [EMAIL PROTECTED] Internet: http://www.wapme-systems.de --- wapme.net - wherever you are
Re: virtual SMSC configuration!
I am setting up Kannel as an SMS gateway. I've tried using a NOKIA 6210 as a virtual SMSC, since this is the kit available to me presently for this purpose. Setting the SMSC configuration as follows: group =smsc smsc = at smsc-id = GSMMODEM host = localhost port = 1 The settings host and port are completely useless, you should remove them. modemtype = premicell device = /dev/ttyS1 and starting the bearerbox service gives the following error:Cannot start with SMSC connection failing. I will appreciate if you tell me what configuration is required to make this device work for this purpose. On any SMSC AT device you have latency time on the serial COM port. If you start the bearerbox it sometimes doesn't get the hook to the device cleanly. Simply retry a couple of times and you should get a connection. BTW, since Andreas Fink changed a couple of things within smsc_at.c the opening sequence for an AT device is a) taking much londer and b) is more unrealiable for getting the starting hook -- these are at least our experiences. Can someone confirm or disagree with these? Stipe [EMAIL PROTECTED] --- Wapme Systems AG Münsterstr. 248 40470 Düsseldorf Tel: +49-211-74845-0 Fax: +49-211-74845-299 E-Mail: [EMAIL PROTECTED] Internet: http://www.wapme-systems.de --- wapme.net - wherever you are
Re: RAM and page faults
As requested I xcan told you that with the latest cvs, smsc=at2 on windows2000 there are no RAM problems, by the way about 4 page faults every seconds. Everyone know why? Maybe due to cygwin dll, 1.1.8? don't know. You may try Cygwin 1.3.3-2 and see what happens. Stipe [EMAIL PROTECTED] --- Wapme Systems AG Münsterstr. 248 40470 Düsseldorf Tel: +49-211-74845-0 Fax: +49-211-74845-299 E-Mail: [EMAIL PROTECTED] Internet: http://www.wapme-systems.de --- wapme.net - wherever you are
Re: anyone using EMI2 with DLR?
Bi Bruno, do you get the panic even if you don't specify the dlr_* in http push ? yes, even for normal cgi-bin/sendsms this happens. Stipe [EMAIL PROTECTED] --- Wapme Systems AG Münsterstr. 248 40470 Düsseldorf Tel: +49-211-74845-0 Fax: +49-211-74845-299 E-Mail: [EMAIL PROTECTED] Internet: http://www.wapme-systems.de --- wapme.net - wherever you are
BUG in emi2 (was Re: anyone using EMI2 with DLR?)
Hi all, Did you check if the problem still occurs when you comment fakesmsc group entries ? I noticed that bearerbox crashes with cgi-bin/status on Redhat 6.2 with the last CVS version abd only when you add fakesmsc group. There is no fakesmsc group in the config file. There is definitly nothing unusual within the config file. We use a simple emi2 smsc group. Both boxes (smsbox and bearerbox) crash after they write these log etries: bearerbox.log: [...] 2001-10-08 10:51:58 [4] DEBUG: HTTP: Destroying HTTPClient area 0x10035958. 2001-10-08 10:51:58 [4] DEBUG: HTTP: Destroying HTTPClient for `213.70.39.36'. 2001-10-08 10:52:40 [9] DEBUG: boxc_receiver: sms received 2001-10-08 10:52:40 [7] DEBUG: emi2 sending packet: 00/00094/O/51/00491735333216/10F730BC5D76B9CB74/3//746573745039/72 2001-10-08 10:52:40 [7] DEBUG: Got packet from the main socket 2001-10-08 10:52:40 [7] DEBUG: emi2 parsing packet: 00/00020/R/51/A///94 2001-10-08 10:52:40 [7] PANIC: gwlib/octstr.c:2032: seems_valid_real: Assertion `ostr != NULL' failed. (Called from gwlib/octstr.c:839:octstr_search_char.) smsbox.log: [...] 2001-10-08 10:52:40 [3] INFO: smsbox: Got HTTP request /cgi-bin/sendsms from 213.70.39.36 2001-10-08 10:52:40 [3] INFO: sendsms used by foo 2001-10-08 10:52:40 [3] INFO: sendsms sender:foo:wapme.net (213.70.39.36) to:491735333216 msg:test 2001-10-08 10:52:40 [3] DEBUG: Status: 202 Answer: Sent. 2001-10-08 10:52:40 [3] DEBUG: HTTP: Resetting HTTPClient for `213.70.39.36'. 2001-10-08 10:52:40 [0] INFO: Connection closed by the bearerbox 2001-10-08 10:52:40 [0] INFO: Received (and handled?) 0 requests in 45 seconds (0.00 per second) 2001-10-08 10:52:40 [0] INFO: Kannel smsbox terminating. 2001-10-08 10:52:40 [6] DEBUG: Thread 6 (gw/heartbeat.c:heartbeat_thread) terminates. 2001-10-08 10:52:40 [0] DEBUG: Waiting for 2 (gwlib/http.c:server_thread) to terminate 2001-10-08 10:52:40 [3] DEBUG: HTTP: No clients with requests, quitting. 2001-10-08 10:52:40 [3] DEBUG: Thread 3 (gw/smsbox.c:sendsms_thread) terminates. 2001-10-08 10:52:40 [2] DEBUG: Thread 2 (gwlib/http.c:server_thread) terminates. 2001-10-08 10:52:40 [1] WARNING: Destroying fdset with 1 active entries. 2001-10-08 10:52:40 [1] DEBUG: Thread 1 (gwlib/fdset.c:poller) terminates. 2001-10-08 10:52:40 [0] DEBUG: Waiting for 4 (gw/smsbox.c:obey_request_thread) to terminate 2001-10-08 10:52:40 [4] DEBUG: Thread 4 (gw/smsbox.c:obey_request_thread) terminates. 2001-10-08 10:52:40 [5] DEBUG: Thread 5 (gw/smsbox.c:url_result_thread) terminates. 2001-10-08 10:52:40 [0] DEBUG: Immutable octet strings: 176. 2001-10-08 10:52:40 [0] DEBUG: Current allocations: 12 areas, 732 bytes 2001-10-08 10:52:40 [0] DEBUG: Highest number of allocations: 1257 areas 2001-10-08 10:52:40 [0] DEBUG: Highest memory usage: 60404 bytes 2001-10-08 10:52:40 [0] DEBUG: Area 0x10032110, size 12, max_size 12 2001-10-08 10:52:40 [0] DEBUG: Allocated by mutex_create_real at gwlib/thread.c:26 2001-10-08 10:52:40 [0] DEBUG: Claimed by conn_wrap_fd at gwlib/conn.c:399 2001-10-08 10:52:40 [0] DEBUG: Contents of area (first 16 bytes): 2001-10-08 10:52:40 [0] DEBUG: f0 81 3f 00 ff ff ff ff 01 00 00 00 2001-10-08 10:52:40 [0] DEBUG: Area 0x10014670, size 113, max_size 113 2001-10-08 10:52:40 [0] DEBUG: Allocated by octstr_grow at gwlib/octstr.c:107 2001-10-08 10:52:40 [0] DEBUG: Contents of area (first 16 bytes): 2001-10-08 10:52:40 [0] DEBUG: 00 54 54 50 2f 31 2e 31 20 32 30 32 20 46 6f 6f 2001-10-08 10:52:40 [0] DEBUG: Area 0x100321b0, size 16, max_size 16 2001-10-08 10:52:40 [0] DEBUG: Allocated by octstr_create_from_data_real at gwlib/octstr.c:170 2001-10-08 10:52:40 [0] DEBUG: Claimed by octstr_format_valist at gwlib/octstr.c:1969 2001-10-08 10:52:40 [0] DEBUG: Contents of area (first 16 bytes): 2001-10-08 10:52:40 [0] DEBUG: d0 22 03 10 0c 00 00 00 0d 00 00 00 00 00 00 00 2001-10-08 10:52:40 [0] DEBUG: Area 0x1002e0f8, size 16, max_size 16 2001-10-08 10:52:40 [0] DEBUG: Allocated by octstr_create_from_data_real at gwlib/octstr.c:170 2001-10-08 10:52:40 [0] DEBUG: Claimed by init_smsbox at gw/smsbox.c:1872 2001-10-08 10:52:40 [0] DEBUG: Contents of area (first 16 bytes): 2001-10-08 10:52:40 [0] DEBUG: 30 e1 02 10 0d 00 00 00 0e 00 00 00 00 00 00 00 2001-10-08 10:52:40 [0] DEBUG: Area 0x10032558, size 32, max_size 32 2001-10-08 10:52:40 [0] DEBUG: Allocated by client_create at gwlib/http.c:1370 2001-10-08 10:52:40 [0] DEBUG: Contents of area (first 16 bytes): 2001-10-08 10:52:40 [0] DEBUG: d5 32 00 00 78 20 03 10 b0 21 03 10 00 00 00 00 2001-10-08 10:52:40 [0] DEBUG: Area 0x1002e130, size 14, max_size 14 2001-10-08 10:52:40 [0] DEBUG: Allocated by octstr_create_from_data_real at gwlib/octstr.c:178 2001-10-08 10:52:40 [0] DEBUG: Contents of area (first 16 bytes): 2001-10-08 10:52:40 [0] DEBUG: 30 31 32 33 34 35 36 37 38 39 20 2b 2d 00 2001-10-08 10:52:40 [0] DEBUG: Area 0x10032078, size 68, max_size 68 2001-10-08 10:52:40 [0] DEBUG: Allocated by conn_wrap_fd
Re: [Patch] Emi2 not sending timestamp on ACK
It's now on CVS a patch to support ACK without timestamp. I'll try that one and report. Stipe [EMAIL PROTECTED] --- Wapme Systems AG Münsterstr. 248 40470 Düsseldorf Tel: +49-211-74845-0 Fax: +49-211-74845-299 E-Mail: [EMAIL PROTECTED] Internet: http://www.wapme-systems.de --- wapme.net - wherever you are
[PATCH to come] SSL server support for all HTTP server boxes
Hi, I finished my recent work on enabling Kannel's HTTP servers to support SSL via openssl. The patch will follow in a couple of hours. Stipe [EMAIL PROTECTED] --- Wapme Systems AG Münsterstr. 248 40470 Düsseldorf Tel: +49-211-74845-0 Fax: +49-211-74845-299 E-Mail: [EMAIL PROTECTED] Internet: http://www.wapme-systems.de --- wapme.net - wherever you are
[PATCH] SSLv3 support for HTTP servers (admin and smsbox) using openssl
= ota_dispatch;
dispatch_to_appl = appl_dispatch;
-http_open_port(HTTP_PORT);
+http_open_port(HTTP_PORT, ssl);
http_clients = dict_create(NUMBER_OF_PUSHES, NULL);
urls = dict_create(NUMBER_OF_PUSHES, octstr_destroy_item);
diff -urN gateway/gwlib/cfg.def gateway-2001-10-11-ssl/gwlib/cfg.def
--- gateway/gwlib/cfg.def Thu Oct 11 17:18:34 2001
+++ gateway-2001-10-11-ssl/gwlib/cfg.defThu Oct 11 21:25:22 2001
@@ -20,6 +20,7 @@
SINGLE_GROUP(core,
OCTSTR(admin-port)
+OCTSTR(admin-port-ssl)
OCTSTR(admin-password)
OCTSTR(status-password)
OCTSTR(admin-deny-ip)
@@ -43,7 +44,9 @@
OCTSTR(http-proxy-exceptions)
OCTSTR(http-proxy-username)
OCTSTR(http-proxy-password)
-OCTSTR(ssl-certkey-file)
+OCTSTR(ssl-client-certkey-file)
+OCTSTR(ssl-server-cert-file)
+OCTSTR(ssl-server-key-file)
)
@@ -75,6 +78,7 @@
SINGLE_GROUP(smsbox,
OCTSTR(bearerbox-host)
OCTSTR(sendsms-port)
+OCTSTR(sendsms-port-ssl)
OCTSTR(sendsms-chars)
OCTSTR(global-sender)
OCTSTR(log-file)
diff -urN gateway/gwlib/conn.c gateway-2001-10-11-ssl/gwlib/conn.c
--- gateway/gwlib/conn.cMon Oct 8 19:43:04 2001
+++ gateway-2001-10-11-ssl/gwlib/conn.c Thu Oct 11 21:32:58 2001
@@ -3,6 +3,12 @@
* This file implements the interface defined in conn.h.
*
* Richard Braakman
+ *
+ * SSL client implementation contributed by
+ * Jarkko Kovala [EMAIL PROTECTED]
+ *
+ * SSL server implementation contributed by
+ * Stipe Tolj [EMAIL PROTECTED] for Wapme Systems AG
*/
/* TODO: unlocked_close() on error */
@@ -22,6 +28,9 @@
#include openssl/ssl.h
SSL_CTX *global_ssl_context;
+SSL_CTX *global_server_ssl_context;
+X509 *ssl_public_cert;
+RSA *ssl_private_key;
#endif /* HAVE_LIBSSL */
/*
@@ -386,13 +395,14 @@
NULL : octstr_get_cstr(our_host));
if (sockfd 0)
return NULL;
-return conn_wrap_fd(sockfd);
+return conn_wrap_fd(sockfd, 0);
}
-Connection *conn_wrap_fd(int fd)
+Connection *conn_wrap_fd(int fd, int ssl)
{
Connection *conn;
+ unsigned long err;
if (socket_set_blocking(fd, 0) 0)
return NULL;
@@ -418,12 +428,51 @@
conn-listening_pollin = 0;
conn-listening_pollout = 0;
#ifdef HAVE_LIBSSL
-conn-ssl = NULL;
-conn-peer_certificate = NULL;
-conn-ssl_mutex = NULL;
+/*
+ * do all the SSL magic for this connection
+ */
+if (ssl) {
+conn-ssl = SSL_new(global_server_ssl_context);
+conn-peer_certificate = NULL;
+
+SSL_set_fd(conn-ssl, conn-fd);
+/*
+if (!SSL_use_RSAPrivateKey(conn-ssl, ssl_private_key)) {
+error(0, SSL: could not use private key for server connection!);
+ SSL_free(conn-ssl);
+goto error;
+}
+if (!SSL_use_certificate(conn-ssl, ssl_public_cert)) {
+error(0, SSL: could not use public certificate for server connection!);
+ SSL_free(conn-ssl);
+goto error;
+}
+*/
+SSL_set_verify(conn-ssl, 0, NULL);
+BIO_set_nbio(SSL_get_rbio(conn-ssl), 0);
+BIO_set_nbio(SSL_get_wbio(conn-ssl), 0);
+
+conn-ssl_mutex = mutex_create();
+if (!SSL_accept(conn-ssl)) {
+ if ((err = ERR_get_error())) {
+error(0, SSL: Access failed: %.256s, ERR_error_string(err, NULL));
+ }
+error(0, SSL: disconnected.);
+SSL_free(conn-ssl);
+goto error;
+}
+} else {
+conn-ssl = NULL;
+conn-peer_certificate = NULL;
+conn-ssl_mutex = NULL;
+}
#endif /* HAVE_LIBSSL */
return conn;
+
+error:
+conn_destroy(conn);
+return NULL;
}
void conn_destroy(Connection *conn)
@@ -989,7 +1038,20 @@
return(conn-peer_certificate);
}
+RSA *tmp_rsa_callback(SSL *ssl, int export, int key_len)
+{
+static RSA *rsa = NULL;
+debug(gwlib.http, 0, SSL: Generating new RSA key (export=%d, keylen=%d),
+export, key_len);
+if (export) {
+ rsa = RSA_generate_key(key_len, RSA_F4, NULL, NULL);
+} else {
+ debug(gwlib.http, 0, SSL: Export not set);
+}
+return rsa;
+}
+
Mutex **ssl_static_locks = NULL;
+Mutex **ssl_server_static_locks = NULL;
void openssl_locking_function(int mode, int n, const char *file, int line)
{
@@ -1014,6 +1076,28 @@
SSL_library_init();
SSL_load_error_strings();
global_ssl_context = SSL_CTX_new(SSLv23_method());
+debug(gwlib.http, 0, HTTP: SSL library for client side initialized);
+}
+
+void server_ssl_init(void)
+{
+int c, maxlocks = CRYPTO_num_locks();
+
+gw_assert(ssl_server_static_locks == NULL);
+ssl_server_static_locks = gw_malloc(sizeof(Mutex *) * maxlocks);
+for (c=0;cmaxlocks;c++)
+ ssl_server_static_locks[c] = mutex_create();
+
+CRYPTO_set_locking_callback(openssl_locking_function);
+CRYPTO_set_id_callback(gwthread_self
[RFC] security between boxes (remote boxing :))
I wonder if there is any interest (or demand) to support SSL connections for the internal Kannel TCP connections? Is there anyone using bearerbox on one machine and smsbox/wapbox on an other machine which may be on a completely different network and would like to use SSL connections between them? Due to the SSL patch I provided I thing we have both implementations, SSL client and SSL server in there, so we may use this also to let boxes comunicate in a secure way. We may include at least the feature and let the user decide (via config switches) if they want secure internal communications between the boxes. Any comments or suggestions? Stipe [EMAIL PROTECTED] --- Wapme Systems AG Münsterstr. 248 40470 Düsseldorf Tel: +49-211-74845-0 Fax: +49-211-74845-299 E-Mail: [EMAIL PROTECTED] Internet: http://www.wapme-systems.de --- wapme.net - wherever you are
Re: [PATCH] SSLv3 support for HTTP servers (admin and smsbox) using openssl
configure.in:
- LIBS=$LIBS -lssl,
- echo no;echo Either get a multithread-enabled SSL or configure with
--disable-ssl;
- exit 1,
+ LIBS=$LIBS -lssl -lcrypto,
+ echo no;AC_DEFINE(HAVE_LIBSSL)
+ LIBS=$LIBS -lssl -lcrypto;echo WARNING: OpenSSL is not
multithread-enabled!,
It really makes no sense to use non-threaded OpenSSL. It leads to random crashes.
And users are prone to ignoring these kinds of warnings, then coming to the list
whining about those random crashes. I think this was discussed earlier when I
made my client side implementation?
I know, but we would like to keep the support for the Cygwin platform
and there we currently run openssl un-threaded.
I don't like to force users to hack in the configure.in (or even
configure in worst case) to get a runable system out of it.
conn.c:
+void server_ssl_init(void)
conn.h:
+void server_init_ssl(void);
This must be a typo.
Is there a need for different certificates and SSL contexts for clients and servers?
This depends on the user. Some may need different instances for server
and client side. This way the user can decide on it's own if he/she
wants them to match in the config file.
Otherwise you would be to restrictive in my opinion if you would use
only one for both sides. Comments from the others?
smsbox.c:
+
+#ifdef HAVE_LIBSSL
+cfg_get_bool(ssl, grp, octstr_imm(admin-port-ssl));
+#ifdef HAVE_LIBSSL
+cfg_get_bool(ssl, grp, octstr_imm(sendsms-port-ssl));
+#endif /* HAVE_LIBSSL */
+
I think we should actually read this config variable even though we haven't
compiled SSL in and warn the user if they're trying to use SSL.
right! -- I'll do change that.
+} else if (ssl) {
+ panic(0, You MUST specify cert and key files within core group for
SSL!);
+}
Is this really cause for panic?
If you want SSL support for the HTTP server, but there is no cert and
key file specified?? Hmm, I think so, yes.
Stipe
[EMAIL PROTECTED]
---
Wapme Systems AG
Münsterstr. 248
40470 Düsseldorf
Tel: +49-211-74845-0
Fax: +49-211-74845-299
E-Mail: [EMAIL PROTECTED]
Internet: http://www.wapme-systems.de
---
wapme.net - wherever you are
Re: [RFC] security between boxes (remote boxing :))
Thats what I mean. Most users run bearerbox and smsbox on the same machine. If they are separate the its usually on two boxes sitting next to each other and performance is a key issue in those situations (as otherwhise you wouldnt use two boxes). So adding SSL to a performance sensitive environment isnt helping. But in those cases its relatively easy to make the network secure by using a separate network between bearerbox and smsboxes. The only case where I think it would be of course you can use lower-level network security like IPSEC or something for remote communications. But in case of (relatively) easy and open internet based connections you will be communicating in plaintext. helpful is when the bearerbox is far remote from the SMSbox and reached over the internet. in that case however you might run bearerbox smsbox in the remote location and run a local bearerbox smsbox and interface with HTTP between the two locations and there you can use HTTPS with your patch already. but what about this: bearerbox --- (the world) --- smsbox/wapbox (LAN 1)(LAN 2) so if you don't run beaererbox on the machine of LAN 2 you will have plaintext TCP connections between bearerbox from LAN1 and smsbox/wapbox from LAN2? Stipe [EMAIL PROTECTED] --- Wapme Systems AG Münsterstr. 248 40470 Düsseldorf Tel: +49-211-74845-0 Fax: +49-211-74845-299 E-Mail: [EMAIL PROTECTED] Internet: http://www.wapme-systems.de --- wapme.net - wherever you are
[Fwd: SMSGateway]
-- Für weitere Fragen stehe ich Ihnen gerne telefonisch oder via E-Mail zur Verfügung. Mit freundlichen Grüßen Stipe Tolj Department Management Technology Center Research Lab [EMAIL PROTECTED] --- Wapme Systems AG Münsterstr. 248 40470 Düsseldorf Tel: +49-211-74845-0 Fax: +49-211-74845-299 E-Mail: [EMAIL PROTECTED] Internet: http://www.wapme-systems.de --- wapme.net - wherever you are ---BeginMessage--- Hi stipe, I have one more question on fakesmsc. Currently we are using smsgateway, which we will open only transmitter connection to the SMSC, means from the browser we will send message to the cellphone user, we are not planning to do the vice-versa. In this case, we want to test the connection from smsgateway to the fakesmsc.My question is how we will be sure that connection is getting successful .Because as far as my understanding fakesmsc sent message to the client and vice versa. But in a real test ,when we make a bind transmitter connection to the SMSC we get an acknowledgement from SMSC, so how we will test this part with the fakesmsc, I mean the handshaking part. I really appreciate your help, thanks for all your previous response for my questions. Thanks Ravi Ravishankar Tiwari Opsol Integrators Inc (408) 364 9915 ext-87 (408) 364 9916 - Original Message - From: Stipe Tolj [EMAIL PROTECTED] To: Ravi [EMAIL PROTECTED]; [EMAIL PROTECTED] Sent: Friday, October 05, 2001 1:59 AM Subject: Re: SMSGateway Hi Ravi, (i) At what port bearer box listen to the request from SMSC, and how SMSC knows about that port ? (ii) In smskannel.conf file group = smsc smsc = smpp host = 123.123.123.123 port = 600 this specifies the port used at the smsc side (you have to get this from your smsc provider, usually it differs) receive-port = 700 this specifies the port used at your (client) side (either you tell your smsc provider about that, or he enforces you to use a specific port). system-id = STT smsc-password = foo system-type = VMA address-range = we have two different port, onw is port and the other is receive port. Are they belong to the SMSC (machine on which SMSC running)? (see my above comment on that.) If yes then why two different port to connect to the SMSC.I have also seen the code : nop, as I said, one port is at smsc side and one is on your side. Regards, Stipe Tolj Department Management Technology Center Research Lab [EMAIL PROTECTED] --- Wapme Systems AG Münsterstr. 248 40470 Düsseldorf Tel: +49-211-74845-0 Fax: +49-211-74845-299 E-Mail: [EMAIL PROTECTED] Internet: http://www.wapme-systems.de --- wapme.net - wherever you are ---End Message---
Re: minor bug(?) (keepalive) found in smsc_emi2_create
if there is no username, the keepalive is set to unitialized garbage. Is that OK ? of course not. Stipe [EMAIL PROTECTED] --- Wapme Systems AG Münsterstr. 248 40470 Düsseldorf Tel: +49-211-74845-0 Fax: +49-211-74845-299 E-Mail: [EMAIL PROTECTED] Internet: http://www.wapme-systems.de --- wapme.net - wherever you are
Re: CVS patches
On Mon, Oct 15, 2001 at 07:06:18PM +1000, Matt Flax wrote: IF YOU DO NOT WANT TO GIVE ME ACCESS THEN FINE, HOWEVER THIS PATCH IS NECESSARY. I suggest you patch the CVS tree and give me access WHEN YOU DEEM MY KNOWLEDGE, EXPERIENCE AND EFFORT WITH KANNEL SIGNIFICANT. OK ? Matt, Please don't post in all-caps. It's hard to read. yes and my mood is getting tangented too :) [EMAIL PROTECTED] --- Wapme Systems AG Münsterstr. 248 40470 Düsseldorf Tel: +49-211-74845-0 Fax: +49-211-74845-299 E-Mail: [EMAIL PROTECTED] Internet: http://www.wapme-systems.de --- wapme.net - wherever you are
gwmem-check: panic: too many allocations
Hi list, has anyone reviewed why the problem is raising for checking-malloc mode and the specific kanne.store file limit? Stipe [EMAIL PROTECTED] --- Wapme Systems AG Münsterstr. 248 40470 Düsseldorf Tel: +49-211-74845-0 Fax: +49-211-74845-299 E-Mail: [EMAIL PROTECTED] Internet: http://www.wapme-systems.de --- wapme.net - wherever you are
Re: SV: smsbox eating CPU ?
Frederik Ammitzbøll schrieb: See below (sent in July): --- On a different note I believe to have located the source of CPU-hungry boxes. The problem lies in Kannel's handling of HTTP 1.1 requests. Try using the send-sms interface or the Kannel-admin module from a HTTP 1.1 compliant browser, leave the browser window open and watch the bearerbox or smsbox go powerhungry ~1 minute after the request. It happens every time!! My temporary workaround has been to treat all HTTP-requests as HTTP 1.0 (in http.c), but this is only a short-term solution as there are important benefits from using HTTP 1.1. --- This problem has existed for a long time and has never been fixed. Unfortunately, while my workaround has severely helped, two simultaneous HTTP-request to Kannel can sometimes have the same effect. I tried to reconstruct the HTTP/1.1 based effect you describe using the admin interface, but could not see any bearerbox's cpu demand which would lead me to such an assumption?! Stipe [EMAIL PROTECTED] --- Wapme Systems AG Münsterstr. 248 40470 Düsseldorf Tel: +49-211-74845-0 Fax: +49-211-74845-299 E-Mail: [EMAIL PROTECTED] Internet: http://www.wapme-systems.de --- wapme.net - wherever you are
Re: Found and fixed SSL locking bug
I just found and fixed a long standing bug in SSL locking. The problem was that both, the client and the server context used the same SSL locking callback function. Unfortunately this function referenced only one set of mutexes. The fresh and shiny server mutexes never got used. On shutdown the SSL server code then tried to access the already released client SSL mutexes which caused the crash on shutdown. The fix is obvious: use a separate callback for the server context. This is implemented and in CVS. thanks Jörg, this has been obviously my fault :( Stipe [EMAIL PROTECTED] --- Wapme Systems AG Münsterstr. 248 40470 Düsseldorf Tel: +49-211-74845-0 Fax: +49-211-74845-299 E-Mail: [EMAIL PROTECTED] Internet: http://www.wapme-systems.de --- wapme.net - wherever you are
Re: Heads up: SSL client code to verify SSL server certificates
I'm going to commit a patch that adds certificate verification to the Kannel SSL client code. Previously all SSL client connections were doen with verification disabled. This way an encrypted connection was established but the identity of the server was not proven. Please review this code and give me some feedback. I would suggest to post the patch to the devel@ list, so those who are interested and feel responsible have a change to view at the code before commiting to cvs. The one who send the patch (if having cvs write permissions) should at least wait for some time to get the heads up counted. I would suggest the voting structure: -1 : no way, not acceptable 0 : don't know, I'm not the showstopper +1 : yep, that looks ok, go commit Stipe [EMAIL PROTECTED] --- Wapme Systems AG Münsterstr. 248 40470 Düsseldorf Tel: +49-211-74845-0 Fax: +49-211-74845-299 E-Mail: [EMAIL PROTECTED] Internet: http://www.wapme-systems.de --- wapme.net - wherever you are
Re: Heads up: SSL client code to verify SSL server certificates
Sorry if I stepped on some toes. I did not consider the changes to be THAT intrusive. They do not change the structure of the code, they just add a new feature. I took care not to change the behaviour of the system when the new configuration option is not set. I'm willing to back the changes out if you oppose them. Do you? Nono, Jörg, your fine with me. I just wanted to note that it should be done on the voting scheme in general. It hasn't to be on every single change, especially when routined contributers like you know what they are doing and know that they don't create any inconsistency or trouble to others :)) Stipe [EMAIL PROTECTED] --- Wapme Systems AG Münsterstr. 248 40470 Düsseldorf Tel: +49-211-74845-0 Fax: +49-211-74845-299 E-Mail: [EMAIL PROTECTED] Internet: http://www.wapme-systems.de --- wapme.net - wherever you are
Re: Heads up: SSL client code to verify SSL server certificates
Sorry if I stepped on some toes. I did not consider the changes to be THAT intrusive. They do not change the structure of the code, they just add a new feature. I took care not to change the behaviour of the system when the new configuration option is not set. I'm willing to back the changes out if you oppose them. Do you? I agree on stipe's voting idea but only for drastic changes. bugfixes and small modifications should be rolled out. After all, this is a development CVS. yep, of course, same opinion -- it seems I didn't make it as clear as you Andreas. Stipe [EMAIL PROTECTED] --- Wapme Systems AG Münsterstr. 248 40470 Düsseldorf Tel: +49-211-74845-0 Fax: +49-211-74845-299 E-Mail: [EMAIL PROTECTED] Internet: http://www.wapme-systems.de --- wapme.net - wherever you are
Re: Heads up: SSL client code to verify SSL server certificates
BTW, the patch looks good! Stipe [EMAIL PROTECTED] --- Wapme Systems AG Münsterstr. 248 40470 Düsseldorf Tel: +49-211-74845-0 Fax: +49-211-74845-299 E-Mail: [EMAIL PROTECTED] Internet: http://www.wapme-systems.de --- wapme.net - wherever you are
Re: Typo in at2
there's a typo in the modem definitions for at2, diff (smsc_at2.c) follows:
diff -U1 -r1.14 smsc_at2.c
--- smsc_at2.c 2001/11/14 17:17:45 1.14
+++ smsc_at2.c 2001/12/06 13:51:28
@@ -74,3 +74,3 @@
{ falcom,AT+IFC=2,2, 9600 ,
AT+CNMI=1,2,0,0,0, Falcom, NULL, 0, 0, 1, 0, 0 },
-{ ericcson, AT+IFC=2,2, 9600 , AT+CNMI=3,2,0,0,
R520m,NULL, 0, 0, 1, 1, 1 }
+{ ericsson, AT+IFC=2,2, 9600 , AT+CNMI=3,2,0,0,
R520m,NULL, 0, 0, 1, 1, 1 }
};
typo has been fixed in cvs, thanks a lot!
BTW, we would appritiate to attach diffs to the mail, so the long
lines don't get wrapped. Thanks.
Stipe
[EMAIL PROTECTED]
---
Wapme Systems AG
Münsterstr. 248
40470 Düsseldorf
Tel: +49-211-74845-0
Fax: +49-211-74845-299
E-Mail: [EMAIL PROTECTED]
Internet: http://www.wapme-systems.de
---
wapme.net - wherever you are
Re: Ericsson phones supporting WAP bookmarks
Started building support for client provisioning. Do you know what Ericsson phones support WAP bookmarks ? R320, R380 and R520 as far as I recall. Do you mean WAP 2.0 client provisioning or Ericsson proprietary ? Mainly WAP 2.0, but Ericsson proprietary would bo of some interest, too. the Nokia and Ericsson proprietary has been included at least for WAP OTA (whole) configuration settings. It would be easy to add a function that sends only bookmarks. Stipe [EMAIL PROTECTED] --- Wapme Systems AG Münsterstr. 248 40470 Düsseldorf Tel: +49-211-74845-0 Fax: +49-211-74845-299 E-Mail: [EMAIL PROTECTED] Internet: http://www.wapme-systems.de --- wapme.net - wherever you are
Re: what centers i can use
i developed fake sms technology using kannel gate way. but i want develop real sms technology using kannel gateway . so what smsc centers i can use to develop sms technology. and what phones it's supported . please reply me . all information you need is in the user's guide, please visit Kannel's web site and see the section documentation. Regards, Stipe Tolj Department Management Technology Center Research Lab [EMAIL PROTECTED] --- Wapme Systems AG Münsterstr. 248 40470 Düsseldorf Tel: +49-211-74845-0 Fax: +49-211-74845-299 E-Mail: [EMAIL PROTECTED] Internet: http://www.wapme-systems.de --- wapme.net - wherever you are
Re: Ref-Counted octstr
Aarno Syvänen schrieb: Jörg Pommnitz wrote: Currently we often do a octstr_duplicate instead of the octstr_get_ref in my made up example. This is simplified but the pattern occurs quite often... Yeah, octstr_get_ref would be nice. Forget my email: I think about another type of unneeded octstr_duplicate. I'm not into deep here, but would a major change have any global impacts? Stipe [EMAIL PROTECTED] --- Wapme Systems AG Münsterstr. 248 40470 Düsseldorf Tel: +49-211-74845-0 Fax: +49-211-74845-299 E-Mail: [EMAIL PROTECTED] Internet: http://www.wapme-systems.de --- wapme.net - wherever you are
Re: SMSBox PANIC
Oh no. There's still a bug in smsbox.c with msgdata == NULL ? g.. I'll take a look at it. Great, thanks Bruno! Stipe [EMAIL PROTECTED] --- Wapme Systems AG Münsterstr. 248 40470 Düsseldorf Tel: +49-211-74845-0 Fax: +49-211-74845-299 E-Mail: [EMAIL PROTECTED] Internet: http://www.wapme-systems.de --- wapme.net - wherever you are
[FYI] SSL support for inter-box communication
Hi list, I have implemented the changes necessary to make the single boxes communicate using SSL. As I stated in a previous post, this should be considered _optional_ due to the overhead the whole SSL stuff generates for TCP communication if the boxes are running on the same machine. But it is very usefull to secure network communication, i.e. if smsboxes are running on different network machines and bearerbox is addressed for SMS sending. The configuration directives: smsbox-port-ssl = bool wapbox-port-ssl = bool have been introduced and default to no. So the SSL-enabling of inter-box communication is very easy. I have revised the SSL server side code too and will post the diff later to review from Jörg, Jarkko and/or Andreas, if they don't mind, before commiting to CVS. Regards, Stipe [EMAIL PROTECTED] --- Wapme Systems AG Münsterstr. 248 40470 Düsseldorf Tel: +49-211-74845-0 Fax: +49-211-74845-299 E-Mail: [EMAIL PROTECTED] Internet: http://www.wapme-systems.de --- wapme.net - wherever you are
Re: mutex failure?
The things you are using are thread safe, so they lock their
inner state against concurrent access. You might want to add
a wrapper around mutex_unlock in the following way:
#define mutex_unlock(mtx)
mutex_unlock_inline(mtx,__FILE__,__LINE__,__PRETTY_FUNCTION__)
static inline mutex_unlock_inline(Mutex mtx, const char *file, int line,
const char *func) {
debug(mutex, 0, mutex_unlock called from %s: %d in function %s, file,
line, func);
mutex_unlock_real(mtx);
}
I get the same error:
2001-12-13 15:13:38 [16] PANIC: mutex_unlock: Mutex failure!
2001-12-13 15:13:38 [16] PANIC: System error 22: Invalid argument
for smsbox running a Siemens M20. Any ideas?
Stipe
[EMAIL PROTECTED]
---
Wapme Systems AG
Münsterstr. 248
40470 Düsseldorf
Tel: +49-211-74845-0
Fax: +49-211-74845-299
E-Mail: [EMAIL PROTECTED]
Internet: http://www.wapme-systems.de
---
wapme.net - wherever you are
Re: A problem with the HTTP SMSC
I have now a kannel installation that uses the HTTP SMSC with quite high traffic, like 6 to 10 messages per minute all the time. And now there seems to be problems with the HTTP server; the bearerbox log show this: 2001-12-19 12:28:17 [3] ERROR: HTTP: Error accepting a client. 2001-12-19 12:28:17 [3] ERROR: System error 24: Too many open files I took a look at the HTTP code and found out that this error happens when there is no more free server sockets to be opened. The number of the servers is limited to 32. My guess is that the connections are open for too long and therefore the number of connections runs out when requests just keep coming in. So couple of questions: Shouldn't Kannel close the sockets when the HTTP transaction is complete, I mean that a reply to an request has been sent? Or is there a timeout and how much is that timeout? Kannel does use Keep-Alive for HTTP connection on transport layer, hence if the HTTP client send a request with keep-alive = yes, then Kannel should hold that connection until a timeout occures, or am I wrong here? So if you fire up 10 simultanious HTTP client, 10 HTTPClient instances on the server side are fired up and live for lets say 2 Min.. If you multiply that effect you will reach your 32 simultanious connections very quickly. Does anyone know if Kannel checks for Keep-Alive Header in HTTP/1.1 requests? Stipe [EMAIL PROTECTED] --- Wapme Systems AG Münsterstr. 248 40470 Düsseldorf Tel: +49-211-74845-0 Fax: +49-211-74845-299 E-Mail: [EMAIL PROTECTED] Internet: http://www.wapme-systems.de --- wapme.net - wherever you are
Re: label in logs to differenciate sendsms requests
I would like to differenciate sendsms requests for a unique sendsms user defined (there will be several users of a unique sendmsms account). I can't use from variable or text or smsc which are used for another aim. Is there any other variable to do that ? Is someone interesting to add it ? so you are trying to aggregate several logical userid's to one kannel's sendsms userid, is that right? Any that information should be logged to access.log? Hmmm, from the architecture side I would claim you to set up as much same sendsms users as you will have logical user ids. Why isn't that practible in your case? Stipe [EMAIL PROTECTED] --- Wapme Systems AG Münsterstr. 248 40470 Düsseldorf Tel: +49-211-74845-0 Fax: +49-211-74845-299 E-Mail: [EMAIL PROTECTED] Internet: http://www.wapme-systems.de --- wapme.net - wherever you are
Re: label in logs to differenciate sendsms requests
internet service provider A offers an e-mail to SMS service. he forwards the message from his application to kannel and passes on username to his local kannel as customername. The request would be routed to another wholesale provider using kannel as well. There the username would appear as ispname:customername. So if at any time there's a query on a message, this mechanism can be easily used to pull out the logs. Also the wholesaler can make accounting specific to the isp's different customers or services. If I understand correctly, the message will go sendsms request - Kannel - HTTP SMSC 1 (another Kannel) - - HTTP SMSC n (another Kannel) - SMSC I still did not get it -- would someone be so kind to enlight me here :) Stipe [EMAIL PROTECTED] --- Wapme Systems AG Münsterstr. 248 40470 Düsseldorf Tel: +49-211-74845-0 Fax: +49-211-74845-299 E-Mail: [EMAIL PROTECTED] Internet: http://www.wapme-systems.de --- wapme.net - wherever you are
Re: Bug in EMI SMSC
Jacob,
There was a bug in the EMI interface implementation.
When you overwrite the sender phonenumber with another phonenumber the
OTOA-field (Originator Type Of Address) is suposed to be set to 1139 but
it was set to an empty string.
I made a change in gw/smsc_emi2.c on development version 1.1.5 and it
worked.
This is the ouput of a diff between the old file and the new file:
257a258,260
else {
emimsg-fields[E50_OTOA] = octstr_create(1139);
}
Have you checked if this is fixed in the current CVS tree?
If not, would you please provide a patch for it in diff -u format
and our EMI developers will review it before commiting to cvs. Thanks.
Stipe
[EMAIL PROTECTED]
---
Wapme Systems AG
Münsterstr. 248
40470 Düsseldorf
Tel: +49-211-74845-0
Fax: +49-211-74845-299
E-Mail: [EMAIL PROTECTED]
Internet: http://www.wapme-systems.de
---
wapme.net - wherever you are
[PATCH] ssl-enabled inter-box connections
Hi list, attached is the announced patch to support ssl-enabled inter-box connections and various ssl specific (potential) bug fixes. General changes: * supporing ssl-enabled connections between bearerbox and [smsbox|wapbox] * various additional ssl configuration checking * changed the status page to show also ssl-enabled connections (for details have a look in the ChangeLog section, as far as you have time :)) Could the SSL coders around have a review on this, please. If there are no objections, I will commit the patch to CVS in a coupld of hours. Stipe [EMAIL PROTECTED] --- Wapme Systems AG Münsterstr. 248 40470 Düsseldorf Tel: +49-211-74845-0 Fax: +49-211-74845-299 E-Mail: [EMAIL PROTECTED] Internet: http://www.wapme-systems.de --- wapme.net - wherever you are diff -urN gateway/ChangeLog gateway-new/ChangeLog --- gateway/ChangeLog Thu Dec 20 15:46:29 2001 +++ gateway-new/ChangeLog Thu Dec 20 19:21:38 2001 @@ -1,3 +1,45 @@ +2001-12-20 Stipe Tolj [EMAIL PROTECTED] +* doc/userguide/userguide.xml: added documentation for the introduced + configuration directives smsbox-port-ssl and wapbox-port-ssl in + the core group. +* gw/bb_boxc.c: added parameters for calling ssl-enabled box connection + using boolean configuration directives smsbox-port-ssl for bearerbox + to smsbox communication and wapbox-port-ssl for bearerbox to wapbox + communication. Changed the way boxc_status() displays the current + status of kannel's boxes, including ssl information for internal conn. +* gw/bb_http.c: Commented out call to use_global_server_certkey_file() + within httpadmin_start(), this is now done globaly by conn_config_ssl(). +* gw/bearerbox.c: added checking in starter() if corresponding cert and + key files are specified in the conf file for ssl-enabled inter-box + connections. Changed the way the status is displayed. +* gw/shared.c: added OpenSSL version string within widely used + version_report_string(). Added information that established inter-box + connections are ssl-enabled if so. +* gw/shared.h: changed declaration of connect_to_bearerbox() to include + ssl flag. +* gw/smsbox.c: added handling for smsbox-port-ssl conf directive. +* gw/wapbox.c: added handling for wapbox-port-ssl conf directive. +* gwlib/cfg.def: added smsbox-port-ssl and wapbox-port-ssl to allow + boolean configuration of ssl-enabled internal box connections. +* gwlib/conn.c: changed the way ssl connections are established in + conn_wrap_fd() and checked. Fixes also a not previosly detected bug, + that causes non-ssl enabled connections on the same port as wanted + for ssl, hence SSL_accept() = -1 was not handled correctly. Fixed + a potential (non-reported) bug in conn_destroy() that causes errors + if we call unlocked_write() in case of ssl-enabled connection. + Added function conn_get_ssl() to return the corresponding pointer. +* gwlib/conn.h: included declaration of conn_config_ssl() and + conn_get_ssl(). +* gwlib/gwlib.h: added header file inlusion for ssl.h. +* gwlib/http.c: added information to display if HTTP clients are using + ssl to connect to an ssl-enabled HTTP server. + Fixed potential bug if client_create() is called, but the previous + ssl-enabled conn_wrap_fd() has failed, now we check. +* gwlib/http.h: added response code for HTTP_BAD_REQUEST. +* gwlib/ssl.c: added file to hold all ssl specific code. +* gwlib/ssl.h: added file for declaration of all code implemented + in ssl.c. + 2001-12-20: Andreas Fink [EMAIL PROTECTED] * gw/smsc_emi2.c: fixed a problem resulting in timing out at high load. *gw/bb_smscconn.c, gw/msg-decl.h, gw/smsbox.c: diff -urN gateway/doc/userguide/userguide.xml gateway-new/doc/userguide/userguide.xml --- gateway/doc/userguide/userguide.xml Thu Dec 20 15:46:30 2001 +++ gateway-new/doc/userguide/userguide.xml Thu Dec 20 16:47:28 2001 @@ -968,12 +968,34 @@ As with admin-port, this can be anything you want. Must be set if you want to handle any SMS traffic. /entry/row + +rowentryliteralsmsbox-port-ssl (o)/literal/entry + entrybool/entry + entry valign=bottom +If set to true, the smsbox connection module will be SSL-enabled. + Your smsboxes will have to connect using SSL to the bearerbox + then. This is used to secure communication between bearerbox + and smsboxes in case they are in seperate networks operated and + the TCP communication is not secured on a lower network layer. + Defaults to no. + /entry/row rowentryliteralwapbox-port (c)/literal/entry entryport-number/entry entry valign
[RFC] sources stylish enough?!
Is anyone borred the way the sources have been formated in some files? I am. We have a guideline in doc/CodingStyle and hence we should try to follow this, IMO. A way to get back again on stylish clean ground 0 would be to run the specified astyle command against all .c and .h files. Whichs works quite easy anyway. The only problem is that if we clean the source tree that way, we would have a major commit to CVS, since any file would be affected. Any ideas floating around for getting the sources more stylish again? If we decide to run a batched astyle over all source files I would suggest doing this when we tag and roll a new release. Stipe [EMAIL PROTECTED] --- Wapme Systems AG Münsterstr. 248 40470 Düsseldorf Tel: +49-211-74845-0 Fax: +49-211-74845-299 E-Mail: [EMAIL PROTECTED] Internet: http://www.wapme-systems.de --- wapme.net - wherever you are
Re: SSL?
I have openssl-0.9.5a installed on my redhat 7.0 system. When compiling kannel I get this message: Attempting to compile with SSL support. checking for SSL_library_init in -lssl... no ok, that problem is known. -- we have a not very smart way detecting the openssl stuff in configure.in. I was about to change this, but had not yet time to commit to cvs. Do I need to do something else? yes, as a first work-around locate where your libssl.a and libcrypto.a libs are and then do the following: * config.h: #define HAVE_LIBSSL 1 -- will activate the compilation directive * Makefile: add -lssl and -lcryto to the LIBs for linking, add -I/usr/whereever to find the openssl/foobar.h header files. This will compile and build the SSL-enabled binaries for Kannel. Stipe [EMAIL PROTECTED] --- Wapme Systems AG Münsterstr. 248 40470 Düsseldorf Tel: +49-211-74845-0 Fax: +49-211-74845-299 E-Mail: [EMAIL PROTECTED] Internet: http://www.wapme-systems.de --- wapme.net - wherever you are
Re: [RFC] sources stylish enough?!
A way to get back again on stylish clean ground 0 would be to run the specified astyle command against all .c and .h files. Whichs works quite easy anyway. The only problem is that if we clean the source tree that way, we would have a major commit to CVS, since any file would be affected. This would also destroy some formatting. astyle doesn't do everything right, and in fact it does some things badly wrong. (For example, any variable name like return_something gets changed to return _something). Thanks for the objection here Richard. I was not aware of such effects. There's no substitute to doing it by hand. Astyle can help, if you then check the diffs to make sure it didn't mess anything up. Certainly I advise against running it blindly. of course, as you pointed out it would create a logical mess. Seems we may beautify single source files and then check the diffs to see nothing is messed up? Stipe [EMAIL PROTECTED] --- Wapme Systems AG Münsterstr. 248 40470 Düsseldorf Tel: +49-211-74845-0 Fax: +49-211-74845-299 E-Mail: [EMAIL PROTECTED] Internet: http://www.wapme-systems.de --- wapme.net - wherever you are
[ANN] ssl-support for inter-box connections commited
I have just commited the patch to support the ssl-enabled inter-box communication. Please update your cvs checkouts and try the new functions, thanks. If you want to use SSL sockets between bearerbox and smsbox use in core group the directive: ... smsbox-port-ssl = yes ... If you want to use SSL sockets between bearerbox and wapbox use in core group the directive: ... wapbox-port-ssl = yes ... Please keep in mind that you will have to specify the cert and key files using the core group directives: ... ssl-server-cert-file = cert.pem ssl-server-key-file = key.pem ... with your own certificate and private key file. Stipe [EMAIL PROTECTED] --- Wapme Systems AG Münsterstr. 248 40470 Düsseldorf Tel: +49-211-74845-0 Fax: +49-211-74845-299 E-Mail: [EMAIL PROTECTED] Internet: http://www.wapme-systems.de --- wapme.net - wherever you are
Re: Proposed Improvement for Kannel SMS Gateway - Part1
Hi Choong, first thanks a lot for your work. The load balancer will first select SMSC connection (FAKE1) to send out SMS. When the above URL is call again, the load balancer will use SMSC connection (FAKE2) to send out SMS. The SMSC connection will use (FAKE3) and (FAKE4) when the URL is called twice. It will use back SMSC connection (FAKE1) when the URL is called the fifth time. Using this load balancing feature, the sending out SMS will improve as we can have more modems to send out SMS compare using one modem. the round robin strategy would be nice to have, even for non-AT SMSC types. So I guess as long as the patch includes only additive features I'm for a +1 on commit, but I will review the code first before voting for final commit. ii) weighted round robin (not yet implemented) Will submit the patches within next few weeks. 2. Writing to a directory when failed to send SMS Configuration file group = core . . failed-dir = /tmp/failed the failed-dir parameter must be in the core group. The following format is used to write the file :- PNOCR - String connector infoCR - String which userCR - String flagCR - all Integer udhCR - Integer : String messageCR - String The following file naming convention is used :- (random number)_(phone number).err These file(s) can used to resend the failed mesage(s) by an external program. we have already a store file where all not ACK'ed messages are going in. Isn't it possible to use that, or add the structure to make it usable? Finally Merry Chritsmas and Happy New Year to you guys and keep up the good work. Thanks, I'm joining the to all the best wishes for x-mas and new year to all subscribers of the list ... ho, ho, ho :)) Stipe [EMAIL PROTECTED] --- Wapme Systems AG Münsterstr. 248 40470 Düsseldorf Tel: +49-211-74845-0 Fax: +49-211-74845-299 E-Mail: [EMAIL PROTECTED] Internet: http://www.wapme-systems.de --- wapme.net - wherever you are
[ANN] added --with-mysql and --enable-mysql-dir options for configure
I have commited a change to configure.in and the resulting configure script generated by autoconf 2.13 to support two new options: $ ./configure --help [...] --with-mysql[=DLR] where to look for MySQL libs and header files DIR points to the installation [/usr/local] --enable-mysql-dir enable MySQL support for DLR (default: disabled) [...] This change substitues Andreas hard coded libmysqlclient.so searching within configure and configure.in. Stipe [EMAIL PROTECTED] --- Wapme Systems AG Münsterstr. 248 40470 Düsseldorf Tel: +49-211-74845-0 Fax: +49-211-74845-299 E-Mail: [EMAIL PROTECTED] Internet: http://www.wapme-systems.de --- wapme.net - wherever you are
--enable-mysql-dlr of course
Ups, the typo devel striked! it's --enable-mysql-dlr of course. Stipe [EMAIL PROTECTED] --- Wapme Systems AG Münsterstr. 248 40470 Düsseldorf Tel: +49-211-74845-0 Fax: +49-211-74845-299 E-Mail: [EMAIL PROTECTED] Internet: http://www.wapme-systems.de --- wapme.net - wherever you are
Re: new compilation warnings.
gcc -D_REENTRANT=1 -I. -g -O2 -I/usr/include/libxml -I/usr/include -DBROKEN_PTHREADS=1 -o gwlib/conn.o -c gwlib/conn.c gwlib/conn.c: In function `conn_init_ssl': gwlib/conn.c:1175: warning: passing arg 1 of `CRYPTO_set_id_callback' from incompatible pointer type gwlib/conn.c: In function `server_ssl_init': gwlib/conn.c:1192: warning: passing arg 1 of `CRYPTO_set_id_callback' from incompatible pointer type gcc -D_REENTRANT=1 -I. -g -O2 -I/usr/include/libxml -I/usr/include -DBROKEN_PTHREADS=1 -o gwlib/counter.o -c gwlib/counter.c gcc -D_REENTRANT=1 -I. -g -O2 -I/usr/include/libxml -I/usr/include -DBROKEN_PTHREADS=1 -o gwlib/date.o -c gwlib/date.c these are known and should not have any major impact. Jörg, what do you think about fixing this? I was looking at it, but did not get the pointer type conversion for it, do you have any suggestion? Stipe [EMAIL PROTECTED] --- Wapme Systems AG Münsterstr. 248 40470 Düsseldorf Tel: +49-211-74845-0 Fax: +49-211-74845-299 E-Mail: [EMAIL PROTECTED] Internet: http://www.wapme-systems.de --- wapme.net - wherever you are
Re: Stripping Kannel
We have been using Kannel 1.1.5 and cvs for SMS applications. It has been working fine with AT and CIMD2 driver. Is it possible to strip Kannel to bare minimum like bearerbox will have only required SMSC driver and no WDP layer. If yes, how to do it. I thought about some options like 1. Changes in Makefile 2. Commenting whole driver files before compiling 3. And of course going thru whole source code. My ideas must be vague, but there has to be some way to do it. Please comment and guide me. yes there is a simple way, at least to strip down bearerbox to use only wap or sms features. bearerbox.c has a couple of conditional #define checkings for that. To use them specifiy manually within config.h: #define KANNEL_NO_WAP 1 to exclude compilation of WAP specific code within bearerbox.c and #define KANNE_NO_SMS 1 to exclude the SMS specific code. This has not been intended for what you actualy want, it was for debugging and testing, but should do at least a strip down. We may extend this idea for bearerbox to strip down to only needed function, even if I don't think that it is necessary for some realy mattering concerns. We may add a --with-bearerbox-mode=[sms|wap] configure switch which sets a #define in config.h accordingly and those may be tested for the conditional compile for bearerbox.c. Any comments on that from the list? Has anyone real application demands on this? Stipe [EMAIL PROTECTED] --- Wapme Systems AG Münsterstr. 248 40470 Düsseldorf Tel: +49-211-74845-0 Fax: +49-211-74845-299 E-Mail: [EMAIL PROTECTED] Internet: http://www.wapme-systems.de --- wapme.net - wherever you are
Re: contrib/php-admin/
Hi Bernino, first of all thanks for your encouragement towards the Kannel project. We like to see the fruits of our work :) sendol.php :Send an Operator logo. sendgl.php :Send a Group Logo. sendrt.php :Send a Ringing tone. It's not that a major thing implementing these, but it's tricky do decide how you transfer a bitmap to the php script or the ringtone tunes. You may use standardized formats and then PUT them to the HTTP server php script, but usually all of us who are using Kannel for ringtone and logo SMS use a relational database and serverlets, php or other to generate the text= string for them. Any contributions in the style of send*.php would be highly appritiated anyway. My question is: Does the missing files reside somewhere ? Maybe liw has them ? Have you asked him? Stipe [EMAIL PROTECTED] --- Wapme Systems AG Münsterstr. 248 40470 Düsseldorf Tel: +49-211-74845-0 Fax: +49-211-74845-299 E-Mail: [EMAIL PROTECTED] Internet: http://www.wapme-systems.de --- wapme.net - wherever you are
Re: a howto compile cvs on FreeBSD
4. The configure.in does not setup -I/usr/local/include, so if thats not part of your gmake configuration per default, you should edit the generated makefile and add this include flag, in the CFLAG variable. -I/usr/local/include should not be necessary, due that system headers should reside in /usr/include and for all optional (xml2, ssl, mysql) libs we are testing in configure.in and modify CFLAGS accordingly. Have you any scenario where pre-defined inclusion of /usr/local/include would be necessary? We may then include it. Also take care that configure.in have some static defined paths to various dependencies such as ssl libraries. That's to be changed and has been announced already. The SSL detection will be smarter. 5. Use 'gmake' instead of make to do all compilations That's also valid for Sun's make on Solaris 2.x. [EMAIL PROTECTED] --- Wapme Systems AG Münsterstr. 248 40470 Düsseldorf Tel: +49-211-74845-0 Fax: +49-211-74845-299 E-Mail: [EMAIL PROTECTED] Internet: http://www.wapme-systems.de --- wapme.net - wherever you are
Re: english - specified...
During my test of kannel, I run into a little error in english: Spesified is spelled Specified I run a quick test on http://babel.altavista.com/ and it seems that Specified is the right one, or am I wrong here, others? Stipe [EMAIL PROTECTED] --- Wapme Systems AG Münsterstr. 248 40470 Düsseldorf Tel: +49-211-74845-0 Fax: +49-211-74845-299 E-Mail: [EMAIL PROTECTED] Internet: http://www.wapme-systems.de --- wapme.net - wherever you are
Re: wap-push.cgi
can someone send me an example of this file : wap-push.cgi, to see what it does, please? As Aarno pointed out, wap-push.cgi is _no_ file itself as you may know it from perl or bash CGI scripts. It's used as URI trigger in the wapbox to know that a certain services is requested. Please have a look into the user's guide for information of WAP Push. Stipe [EMAIL PROTECTED] --- Wapme Systems AG Münsterstr. 248 40470 Düsseldorf Tel: +49-211-74845-0 Fax: +49-211-74845-299 E-Mail: [EMAIL PROTECTED] Internet: http://www.wapme-systems.de --- wapme.net - wherever you are
Re: problems with CPU time with kannel 1.1.6
Anders, I too had problems like this with latest cvs. I tracked the problem down to my smsc_cgw.c implementation, and the gwthread_pollfd() function, which always seemed to return immediately without respecting the given timeout value. I'm not sure what caused this problem, but I changed to conn_wait() instead, which worked a whole lot better. could you please provide a patch that solved your problem so the list can review it and we get it commited to cvs if necessary. Thanks. Stipe [EMAIL PROTECTED] --- Wapme Systems AG Münsterstr. 248 40470 Düsseldorf Tel: +49-211-74845-0 Fax: +49-211-74845-299 E-Mail: [EMAIL PROTECTED] Internet: http://www.wapme-systems.de --- wapme.net - wherever you are
[Fwd: kannel gateway]
forwareded: [EMAIL PROTECTED] --- Wapme Systems AG Münsterstr. 248 40470 Düsseldorf Tel: +49-211-74845-0 Fax: +49-211-74845-299 E-Mail: [EMAIL PROTECTED] Internet: http://www.wapme-systems.de --- wapme.net - wherever you are ---BeginMessage--- Hello! Sorry for this message but: i have installed current snapshot under linux env. and my smsc configuretion looks like: group = smsc smsc = emi2 smsc-id = MTS port = 5006 host = xxx.xxx.xxx.xxx connect-allow-ip = xxx.xxx.xxx.xxx receive-port = 5006 denied-prefix = 206 #smsc-username = #smsc-password = flow-control = 1 and while messages sending i see the following messages in kannel.log (see below). is it problem that sequence number (TRN) does not go back to zero? Thanks. 6865746F6D2067726166696B61207261626F7479204D547349206E61736820706C616E207679676C79616469742074616B3A202E205A76 6F6E6974652C2065736C69206275647574200D75746F63686E656E6979612E0D0D532E2054736869706B6F762E2020506F736C65646E69 6A2076617269616E742067726166696B6120434E3D536572676579204D205368636869706B6F762F4F3D494D424D/5C 2001-12-29 11:17:36 [9] DEBUG: boxc_receiver: sms received 2001-12-29 11:17:36 [7] DEBUG: Got packet from the main socket 2001-12-29 11:17:36 [7] DEBUG: emi2 parsing packet: 61/00044/R/51/A//89021733070:29120727/75 2001-12-29 11:17:36 [7] DEBUG: emi2 sending packet: 62/00068/O/51/89021733070/206/3//534B /C7 2001-12-29 11:17:36 [1] DEBUG: Dumping 1 messages and 0 acks to store 2001-12-29 11:17:36 [7] DEBUG: Got packet from the main socket 2001-12-29 11:17:36 [7] DEBUG: emi2 parsing packet: 62/00044/R/51/A//89021733070:29120728/77 2001-12-29 11:18:21 [9] DEBUG: boxc_receiver: sms received 2001-12-29 11:18:21 [7] DEBUG: emi2 sending packet: 63/00188/O/51/89021733070/206/3//52415320 4C6F7475732028207564616C796F6E6E796A20646F73747570202920434E3D536572676579204D205368636869706B6F762F4F3D494D42 4D534B/88 2001-12-29 11:18:21 [7] DEBUG: Got packet from the main socket 2001-12-29 11:18:21 [7] DEBUG: emi2 parsing packet: 63/00044/R/51/A//89021733070:29120812/72 2001-12-29 11:18:21 [1] DEBUG: Dumping 0 messages and 0 acks to store 2001-12-29 11:25:33 [9] DEBUG: boxc_receiver: sms received 2001-12-29 11:25:33 [7] DEBUG: emi2 sending packet: 64/00088/O/51/89021733068/206/3//74657374 206D657373616765/F9 with best wishes German Aksenov phone: (095)258-7258 ext. 7836 ---End Message---
[ANN] sendsms-url and smsota-url config directives added
Hi list, I just commited a minor change to cvs which allows to have user specified sendsms and sendota request URLs for smsbox. This allows to set the URLs respectively in smsbox group as follows: smskannel.conf: ... group = smsbox ... sendsms-url = /cgi-bin/sendsms sendota-url = /cgi-bin/sendota ... These are also hard coded default values if no sendsms-url or/and sendota-url are specified. The documentation changes in userguide.xml will follow. Happy new your to all Kannel developers and users! Stipe [EMAIL PROTECTED] --- Wapme Systems AG Münsterstr. 248 40470 Düsseldorf Tel: +49-211-74845-0 Fax: +49-211-74845-299 E-Mail: [EMAIL PROTECTED] Internet: http://www.wapme-systems.de --- wapme.net - wherever you are
Re: DLR and SMPP..
When using DLR with my SMSC that uses SMPP kannel crashes (cvs version): 2001-12-31 21:42:27 [9] DEBUG: ** SMPP handle_pdu Got DELIVER REPORT 2001-12-31 21:42:27 [9] PANIC: gwlib/octstr.c:2032: seems_valid_real: Assertion `ostr != NULL' failed. (Called from gwlib/octstr.c:728:octstr_compare.) Andreas, have you any idea here? Stipe [EMAIL PROTECTED] --- Wapme Systems AG Münsterstr. 248 40470 Düsseldorf Tel: +49-211-74845-0 Fax: +49-211-74845-299 E-Mail: [EMAIL PROTECTED] Internet: http://www.wapme-systems.de --- wapme.net - wherever you are
Re: Errors in the docs..
After reading some source I see that (at least) the smpp implementation
does not handle multiple receivers (seperarated with space). (this however
works with cimd2). Therefor i suggest that the following is removed from
the documentation:
Phone number of the receiver. To send to multiple receivers, separate each
entry with space (' ', '+' url-encoded) - but note that this can be
deactivated via sendsms-chars in the 'smsbox' group.
I may suggest that those developers using and supporting SMPP should
provide a patch to integrate the multiple-receiver support?!
Who is in charge of the smsc_smpp.c module?
Stipe
[EMAIL PROTECTED]
---
Wapme Systems AG
Münsterstr. 248
40470 Düsseldorf
Tel: +49-211-74845-0
Fax: +49-211-74845-299
E-Mail: [EMAIL PROTECTED]
Internet: http://www.wapme-systems.de
---
wapme.net - wherever you are
Re: DLR and SMPP..
When using DLR with my SMSC that uses SMPP kannel crashes (cvs version): 2001-12-31 21:42:27 [9] DEBUG: ** SMPP handle_pdu Got DELIVER REPORT 2001-12-31 21:42:27 [9] PANIC: gwlib/octstr.c:2032: seems_valid_real: Assertion `ostr != NULL' failed. (Called from gwlib/octstr.c:728:octstr_compare.) it seems that there is a octstr_compare() call within smsc_smpp.c at line 449-450 that cause this assertion. The following atteched patch should fix this NULL pointer assertion for octstr_compare?! Stefan, please try this patch and report and if there are no objections we will commit the patch to cvs. Stipe [EMAIL PROTECTED] --- Wapme Systems AG Münsterstr. 248 40470 Düsseldorf Tel: +49-211-74845-0 Fax: +49-211-74845-299 E-Mail: [EMAIL PROTECTED] Internet: http://www.wapme-systems.de --- wapme.net - wherever you are --- gateway/gw/smsc_smpp.c.orig Tue Jan 1 14:26:20 2002 +++ gateway/gw/smsc_smpp.c Tue Jan 1 14:27:26 2002 @@ -446,8 +446,8 @@ /* we get the following status: DELIVRD, ACCEPTD, EXPIRED, DELETED, UNDELIV, UNKNOWN, REJECTD */ - if ((octstr_compare(stat,octstr_imm(DELIVRD))==0) - || (octstr_compare(stat,octstr_imm(ACCEPTD))==0)) + if ((stat != NULL) ((octstr_compare(stat,octstr_imm(DELIVRD))==0) + || (octstr_compare(stat,octstr_imm(ACCEPTD))==0))) dlrstat = DLR_SUCCESS; else dlrstat = DLR_FAIL;
Re: DLR and SMPP..
Andreas Fink wrote: Why dont we modify octstr_compare that a NULL string can be compared too? NULL would then be equal to . Does that make sense? I'm not sure how deep the impact would be. Anyone from the core architects still reading the list and have their heads up, please?! Stipe [EMAIL PROTECTED] --- Wapme Systems AG Münsterstr. 248 40470 Düsseldorf Tel: +49-211-74845-0 Fax: +49-211-74845-299 E-Mail: [EMAIL PROTECTED] Internet: http://www.wapme-systems.de --- wapme.net - wherever you are
anyone using mutli-cast GET requests
for the sendsms HTTP interface? Any if yes, which smsc? I'm currently figuring out in the code where this is supported for the GET method, but I can't find it. Any I remember that the user's guide that's that multiple receivers are allowed seperated with a blank in the to= argument. Stipe [EMAIL PROTECTED] --- Wapme Systems AG Münsterstr. 248 40470 Düsseldorf Tel: +49-211-74845-0 Fax: +49-211-74845-299 E-Mail: [EMAIL PROTECTED] Internet: http://www.wapme-systems.de --- wapme.net - wherever you are
Re: anyone using mutli-cast GET requests
Stefan Cars wrote: What do you meanwith multi-cast GET ? the user guide states at http://www.kannel.3glab.org/download/kannel-userguide-snapshot/userguide.html#AEN2521 that the to= argument may contain multiple receivers, but I can't see any implementation of this is the code?! I'm just curious if someone is using this, otherwise we have a too optimistic documentation here :)) Stipe [EMAIL PROTECTED] --- Wapme Systems AG Münsterstr. 248 40470 Düsseldorf Tel: +49-211-74845-0 Fax: +49-211-74845-299 E-Mail: [EMAIL PROTECTED] Internet: http://www.wapme-systems.de --- wapme.net - wherever you are
Re: anyone using mutli-cast GET requests
Yes, i'm using this. IT works with CIMD2 at least. so where is the splitting and looping of the single receiver number of the to GET variable implemented? In smsc_cimd2.c? Is the code within the current cvs tree or have you an own version? Stipe [EMAIL PROTECTED] --- Wapme Systems AG Münsterstr. 248 40470 Düsseldorf Tel: +49-211-74845-0 Fax: +49-211-74845-299 E-Mail: [EMAIL PROTECTED] Internet: http://www.wapme-systems.de --- wapme.net - wherever you are
Re: nyone using the load balancer ?
Choong Hong Cheng wrote: Anyone testing the load balancer ? not yet, but it's on my TODO list. Stipe [EMAIL PROTECTED] --- Wapme Systems AG Münsterstr. 248 40470 Düsseldorf Tel: +49-211-74845-0 Fax: +49-211-74845-299 E-Mail: [EMAIL PROTECTED] Internet: http://www.wapme-systems.de --- wapme.net - wherever you are
Re: nyone using the load balancer ?
Is it in the cvs tree ? I cant seem to find it ? the proposed patch has been send to the devel@ list, AFAIK. Choong, can you please state which subject it was, or resend it. Thanks. Stipe [EMAIL PROTECTED] --- Wapme Systems AG Münsterstr. 248 40470 Düsseldorf Tel: +49-211-74845-0 Fax: +49-211-74845-299 E-Mail: [EMAIL PROTECTED] Internet: http://www.wapme-systems.de --- wapme.net - wherever you are
Re: DLR and SMPP..
[EMAIL PROTECTED] wrote: As I see it this approach with octstr_compare is a part of the Kannel's protective coding style where the NULL pointers are not meant to be used as any other similar pointer. So the assert prevents the NULL pointer to go on loosely and stops it before it may cause a segmentation fault on some other part of the program that is not ready to handle NULL pointers. Therefore the line between a NULL pointer and an empty octstr; first one is an accident that shouldn't happen, the latter is a way to say that there is no content. I agree in the view of protective coding style, hence we may distringuish between a NULL pointer and an empty content, IMO. Stipe [EMAIL PROTECTED] --- Wapme Systems AG Münsterstr. 248 40470 Düsseldorf Tel: +49-211-74845-0 Fax: +49-211-74845-299 E-Mail: [EMAIL PROTECTED] Internet: http://www.wapme-systems.de --- wapme.net - wherever you are
