summaryrefslogtreecommitdiffstats
AgeCommit message (Collapse)AuthorFilesLines
2013-09-13Added test for "rach access-control-class" VTY command.kluchnikov/wipIvan Kluchnikov1-0/+35
2013-09-13Added vty command for setting Access control classes.Ivan Kluchnikov1-0/+54
2013-09-12Added test for "auth policy" VTY command.Ivan Kluchnikov1-0/+36
2013-09-11Added new auth policy black-list.Ivan Kluchnikov5-5/+16
In this mode by default we set authorized = 1 for all new subscribers. BSC accepts all MS, except subscribers not authorized in DB. All subscribers with authorized = 0 are part of the black list and not accepted.
2013-09-04bsc_msc: Fix possible null pointer dereferenceHolger Hans Peter Freyther1-3/+1
In case the allocation of con is failing, do not attempt to print con->name as this will be a null pointer dereference. Fixes: Coverity CID 1076318
2013-09-03nat: Remember the original dest local reference in the parsed structHolger Hans Peter Freyther3-1/+7
In case of the RLSD coming from the MSC we are patching the address in-situ but for local calls set con = NULL. We then answered the RLSD with the wrong reference and the MSC kept on trying.
2013-09-03nat: Fix the log message of the code in case of an allocation failureHolger Hans Peter Freyther1-1/+1
This wrong log message appears to be the result of copy and paste
2013-09-03bsc_msc: Stop the re-connect timer in case the msc is lostHolger Hans Peter Freyther1-0/+1
It was possible to cause a crash by enabling and disabling the MSC connection. The enabling lead to scheduling a connection and the second call was not stopping the timer.
2013-09-03bsc_msc: Add a name field to the MSC Connection to differentiate linksHolger Hans Peter Freyther3-20/+45
Assign a static name to a MSC Connection and use it. In case there are multiple connections we can now more easily identify them. This is only used for the NAT right now, the BSC could start to name the various MSC connections too.
2013-09-02vty: Attempt to fix the build when SMPP is not enabledHolger Hans Peter Freyther1-8/+19
2013-09-02vty: Rename 'mgcp-through-msc-ipa' command to 'use-msc-ipa-for-mgcp'Jacob Erlbeck2-9/+9
Currently the 'mgcp' command fails in the 'config-nat' node, because it get confused with 'mgcp-through-msc-ipa' which is executed instead because of the prefix based command selection. Thus the latter command is renamed by this patch to avoid the common prefix. The workaround in the test suite is removed.
2013-09-02vty: Use generic 'end' and 'exit' commandsJacob Erlbeck13-63/+35
Add bsc_install_default() and replace all install_default() This patch adds bsc_install_default() which calls install_default() and add 'exit' and 'end'. All other calls to install_default() are replaced by calls to bsc_install_default(). Since 'exit' and 'end' are now added automatically to each node, the explicit registrations of these commands are removed by this patch, too. The related tests succeed now without work-arounds (except for the 'config' node itself which is part of libosmocore).
2013-09-02vty: Generalize ournode_exit() and ournode_end()Jacob Erlbeck1-104/+36
ournode_exit() duplicates most of bsc_vty_go_parent(). This patch fixes the inconsistencies of both functions within bsc_vty_go_parent() and replaces the implementation of ournode_exit() by a call to it. This makes 'exit' behave exactly like ^D in all openbsc nodes. ournode_end() has been changed to walk through the intermediate nodes until one of the top nodes is reached. This allows for cleanups to be done on the way. Note that in config mode if the tree is searched along the nodes toward the config node and a command is not found this way, a rollback is done by just replacing the vty's node and index member variable by the saved old values which might break the whole thing, when there has been a free() on the way.
2013-09-02vty: Add test to check vty node hierarchy and related commandsJacob Erlbeck1-3/+152
These tests check for the availability of 'exit' and 'end' in each configuration node and for the node specific commands to traverse the tree. In addition, using these commands from within inner contexts is checked. This will detect problems, when an outer command word is a prefix of an inner command, like with 'mgcp' and 'mgcp-through-msc-ipa'. Several assertions are disabled due to inconsistencies and missing commands (see above).
2013-09-02SMPP: add missing VTY parameter 'deliver-src-imsi' for ESMEHarald Welte1-0/+24
the parameter was already used in the code and saved from VTY code, but somehow the functions for parsing it didn't make it into master.
2013-09-02nat: Attempt to follow the MODULE_METHOD pattern for methodsHolger Hans Peter Freyther3-7/+7
Rename methods to be like bsc_ussd_ACTION.
2013-09-02nat: Use bsc_nat_msc_is_connected instead of accessing it directlyHolger Hans Peter Freyther1-2/+2
2013-09-02bsc: Fix compiler warning about undeclare functionsHolger Hans Peter Freyther1-0/+1
Addresses: osmo_bsc_sccp.c:280:2: warning: implicit declaration of function ‘gsm0480_send_ussdNotify’ [-Wimplicit-function-declaration] osmo_bsc_sccp.c:281:2: warning: implicit declaration of function ‘gsm0480_send_releaseComplete’ [-Wimplicit-function-declaration]
2013-09-02nanobts: Fix initialization of two concurrent BTS.Holger Hans Peter Freyther1-15/+51
It was possible that the wrong NSEI information was sent to the BTS. This is because patch_nm_tables is not called before sending the data to the BTS. This will break when two BTS connect more or less at the same time. Stop using the arrays directly and instead introduce a method that will patch the table and return the data and length. This makes sure that all users patch the table before we send the data to the BTS. I bootstrapped a sysmoBTS and I can bring up rsl. The device is running with a dummy load so I couldn't do more verification.
2013-08-31vty: Fix BSC_NODE prompt stringJacob Erlbeck1-1/+1
Change '%s(bsc)#' to '%s(config-bsc)# '. The missing trailing blank breaks osmopy's VTYInteract.command() because the blank is contained in the end patterns which are checked to decide whether to leave the select loop. Thus trying to execute the 'bsc' command there blocks the test script forever.
2013-08-28bsc/vty: Add 'no bsc-welcome-text' commandJacob Erlbeck2-1/+25
There was no command to reset a bsc-welcome-text string, so it has been added.
2013-08-28bsc/ussd: Optionally send USSD message on MSC disconnectionJacob Erlbeck4-4/+105
Send an USSD message on each MS connection if the connection to the MSC has been lost. Add a vty config command 'bsc-msc-loss-txt' in 'config-msc' to set the notification string and to enable the feature. Ticket: OW#957
2013-08-27nat: Add example bsc entry for pabloHolger Hans Peter Freyther1-0/+7
2013-08-27libmgcp: add enum mgcp_rolePablo Neira Ayuso4-4/+14
This enum indicates if the mgcp is running on the BSC or the BSC-NAT.
2013-08-27libmgcp: attach mgcp_ prefix to udp_sendPablo Neira Ayuso1-11/+16
This patch is a cleanup.
2013-08-18Fix: Handle CM service request on already secured channel correctlyAndreas Eversberg1-0/+5
Second part of the previous patch. Log an error message in case we receive a location updating request on an already authenticated channel.
2013-08-14libmgcp: add mgcp prefix to functions and constantsPablo Neira Ayuso1-33/+37
Functions and constants that belong to the libmgcp scope are prefixed with mgcp_ and MGCP_. This patch is a cleanup.
2013-08-14nat/ussd: Add 'show ussd-connection' vty commandJacob Erlbeck3-0/+81
This command returns the current state of the connection to the USSD side channel provider. It shows whether a provider has been connected and authorized or not. Fixes: OW#953
2013-08-13nat: Fix potential memory leak when reading a messageHolger Hans Peter Freyther1-0/+1
Spotted while going through the code with Jacob. We could have leaked the msgb in case of error.
2013-08-13nat: Misc changes spotted while going through the file with JacobHolger Hans Peter Freyther1-3/+3
* Spell Configure correctly * Use %s and VTY_NEWLINE instead of \n
2013-08-03Merge branch 'laforge/smpp-fixes'Holger Hans Peter Freyther1-10/+43
These commits have been deployed by sysmocom and it should qualify as some form of smoke testing. Let's see how things go from here. The by IMSI alertion has not been used, the rest should have been executed by the system.
2013-08-03SMPP: use VTY setting for E212/E164 in ALERT NOTIFICATIONHarald Welte1-24/+25
There's a VTY option by which for every ESME the user can specify if the E.212 or E.164 number should be used in DELIVER-SM. The ALERT notifications generate by subscriber LU have so far always contained the E.212 (IMSI) rather than E.164 (MSISDN) which is a bit inconsistent. Rather than copying code, we create a new function that implements ALERTing all ESMEs.
2013-08-03SMPP: convert a SMMA to a SMPP ALERT NOTIFICATIONHarald Welte1-0/+28
2013-08-03SMPP: don't get stuck in case of SMS memory exceededHarald Welte1-0/+4
If the MS memory for SMS is exceeded and we get an RP-layer error, we need to report that back to the (transaction-mode) ESME. Otherwise the ESME will wait forever after sending a SUBMIT-SM without ever receiving a response to it. Thanks to Holger for catching this.
2013-07-31Merge branch 'zecke/features/big-rewrite'Holger Hans Peter Freyther21-39/+843
2013-07-31nat: Use the DCC region for number rewriting and increase debug outputHolger Hans Peter Freyther1-8/+27
2013-07-31nat: Implement a post-routing for the NAT softwareHolger Hans Peter Freyther7-19/+141
* The post-routing is applied after the first re-writing. To do this the new number is copied back into the called data structure. * Add a testcase that goes from 0172 to 0049 and then back to 0049 using the post rule with a table lookup.
2013-07-31nat: Fix some memory leaks in the testcases.Holger Hans Peter Freyther1-0/+5
The test cases did not free some of the resources it allocated. Improve the situation a bit. There is still data allocated that is not freed at exit.
2013-07-31nat: Allow to use the prefix lookup to rewrite numbersHolger Hans Peter Freyther10-9/+203
* Increase the rewritten rule to five digits (this is the easiest for the unit test). This will add another 40kb to the runtime size. * Create a unit test that tests adding and removing the prefix rules. * Use the regexp match to replace from one package
2013-07-31nat: Introduce a prefix lookup tree (trie) for number rewritingHolger Hans Peter Freyther13-3/+450
* It is a trie. The max depth of the trie is the length of the longest prefix. The lookup is O(lookuped_prefix), but as the prefix length is limited, the lookup time is constant. * Each node can hold the entire prefix, has place for the rewrite rule with up to three digits. * A trie with 20k entries will take about 3MB ram. * Filling the trie 100 times takes ~800ms on my i7 laptop * 10.000.000 lookups take 315ms.. (for the same prefix). * 93/99 lines are tested, 6/6 functions are tested, 49 of 54 branches are tested. Only memory allocation failures are not covered * A late addition is to handle the '+' sign and to increase the number of chars in the rewrite prefix. The timing/line coverage has not been updated after this change.
2013-07-31nat: Add a no number-rewrite command and call it through a VTY testHolger Hans Peter Freyther2-4/+21
The test is just testing the invocation but does not verify that the side effect of this call. It is good enought for now.
2013-07-31gprs_gmm: Avoid assertion in the sending code during routing area updateHolger Hans Peter Freyther1-1/+1
Use old and new tlli as specified by the caller. WIP <000f> sgsn_libgtp.c:321 Received DELETE PDP CTX CONF, cause=128(Request accepted) <0013> gprs_sndcp.c:320 SNSM-DEACTIVATE.ind (lle=0x8095d08, TLLI=e3ddd574, SAPI=11, NSAPI=5) <0011> gprs_bssgp.c:376 BSSGP TLLI=0xe3ddd574 Rx UPLINK-UNITDATA <0012> gprs_llc.c:551 LLC SAPI=1 C FCS=0xb85a96CMD=UI DATA <0011> gprs_bssgp.c:376 BSSGP TLLI=0xe3ddd574 Rx UPLINK-UNITDATA <0012> gprs_llc.c:551 LLC SAPI=1 C FCS=0xbe5a96CMD=UI DATA <0002> gprs_gmm.c:214 Starting MM timer 3350 while old timer 3350 pending <0012> gprs_llc.c:417 LLC TX: unknown TLLI 0xe3ddd574, creating LLME on the fly Assert failed msgb_tlli(msg) == mmctx->llme->tlli || msgb_tlli(msg) == mmctx->llme->old_tlli || tlli_foreign2local(msgb_tlli(msg)) == mmctx->llme->tlli || tlli_foreign2local(msgb_tlli(msg)) == mmctx->llme->old_tlli gprs_llc.c:123 backtrace() returned 18 addresses /home/ich/install/openbsc/lib/libosmocore.so.4(osmo_generate_backtrace+0x16) [0xb7bb6a36] /home/ich/source/gsm/openbsc/openbsc/src/gprs/osmo-sgsn() [0x805224a] /home/ich/source/gsm/openbsc/openbsc/src/gprs/osmo-sgsn() [0x804a2f6] /home/ich/source/gsm/openbsc/openbsc/src/gprs/osmo-sgsn() [0x804b687] /home/ich/source/gsm/openbsc/openbsc/src/gprs/osmo-sgsn() [0x804bc35] /home/ich/source/gsm/openbsc/openbsc/src/gprs/osmo-sgsn() [0x804bd4e] /home/ich/source/gsm/openbsc/openbsc/src/gprs/osmo-sgsn() [0x804d20c] /home/ich/source/gsm/openbsc/openbsc/src/gprs/osmo-sgsn() [0x8052be4] /home/ich/source/gsm/openbsc/openbsc/src/gprs/osmo-sgsn(bssgp_prim_cb+0x55) [0x804f5a4] /home/ich/install/openbsc/lib/libosmogb.so.2(bssgp_rcvmsg+0x3b8) [0xb7b626b8] /home/ich/source/gsm/openbsc/openbsc/src/gprs/osmo-sgsn() [0x804f4f1] /home/ich/install/openbsc/lib/libosmogb.so.2(gprs_ns_rcvmsg+0x8c7) [0xb7b5ecf7] /home/ich/install/openbsc/lib/libosmogb.so.2(+0x4311) [0xb7b5f311] /home/ich/install/openbsc/lib/libosmocore.so.4(osmo_select_main+0x192) [0xb7bb2ed2] /home/ich/source/gsm/openbsc/openbsc/src/gprs/osmo-sgsn() [0x804fbcd] /lib/i386-linux-gnu/i686/cmov/libc.so.6(__libc_start_main+0xf5) [0xb796a8f5] /home/ich/source/gsm/openbsc/openbsc/src/gprs/osmo-sgsn() [0x8049db1] Program received signal SIGABRT, Aborted. 0xb7fde424 in __kernel_vsyscall () (gdb) bt #0 0xb7fde424 in __kernel_vsyscall () #1 0xb797f83f in __GI_raise (sig=sig@entry=6) at ../nptl/sysdeps/unix/sysv/linux/raise.c:56 #2 0xb7982cf3 in __GI_abort () at abort.c:90 #3 0x0805224f in _bssgp_tx_dl_ud (mmctx=0x80932d0, msg=0x80963a8) at gprs_llc.c:120 #4 gprs_llc_tx_ui (msg=0x80963a8, sapi=1 '\001', command=0, mmctx=0x80932d0) at gprs_llc.c:496 #5 0x0804a2f6 in gsm48_gmm_sendmsg (msg=0x80963a8, command=0, mm=0x80932d0) at gprs_gmm.c:241 #6 0x0804b687 in gsm48_tx_gmm_ra_upd_ack (mm=0x80932d0) at gprs_gmm.c:851 #7 0x0804bc35 in gsm48_rx_gmm_ra_upd_req (mmctx=0x80932d0, msg=0x8091ce8, llme=0x8095630) at gprs_gmm.c:1004 #8 0x0804bd4e in gsm0408_rcv_gmm (mmctx=0x80932d0, msg=0x8091ce8, llme=0x8095630) at gprs_gmm.c:1036 #9 0x0804d20c in gsm0408_gprs_rcvmsg (msg=msg@entry=0x8091ce8, llme=0x8095630) at gprs_gmm.c:1566 #10 0x08052be4 in gprs_llc_rcvmsg (msg=0x8091ce8, tv=0xbfffdcb0) at gprs_llc.c:882 #11 0x0804f5a4 in bssgp_prim_cb (oph=oph@entry=0xbfffdc8c, ctx=ctx@entry=0x0) at sgsn_main.c:114 #12 0xb7b626b8 in bssgp_rx_ul_ud (tp=0xbfffdcb0, msg=0x8091ce8, ctx=<optimized out>) at gprs_bssgp.c:398 #13 bssgp_rx_ptp (bctx=0x8091a08, tp=0xbfffdcb0, msg=0x8091ce8) at gprs_bssgp.c:820 #14 bssgp_rcvmsg (msg=0x8091ce8) at gprs_bssgp.c:1016 #15 0x0804f4f1 in sgsn_ns_cb (event=GPRS_NS_EVT_UNIT_DATA, nsvc=0x8090740, msg=0x8091ce8, bvci=1801) at sgsn_main.c:92 #16 0xb7b5ecf7 in gprs_ns_rx_unitdata (msg=0x8091ce8, nsvc=0x8090740) at gprs_ns.c:616 #17 gprs_ns_rcvmsg (nsi=nsi@entry=0x807fd38, msg=msg@entry=0x8091ce8, saddr=saddr@entry=0xbfffedc0, ll=ll@entry=GPRS_NS_LL_UDP) at gprs_ns.c:841 #18 0xb7b5f311 in handle_nsip_read (bfd=0x807fd58) at gprs_ns.c:991 #19 nsip_fd_cb (bfd=0x807fd58, what=1) at gprs_ns.c:1024 #20 0xb7bb2ed2 in osmo_select_main (polling=0) at select.c:158 #21 0x0804fbcd in main (argc=3, argv=0xbffff234) at sgsn_main.c:369 (gdb) frame 5 #5 0x0804a2f6 in gsm48_gmm_sendmsg (msg=0x80963a8, command=0, mm=0x80932d0) at gprs_gmm.c:241 241 return gprs_llc_tx_ui(msg, GPRS_SAPI_GMM, command, mm); (gdb) p msgb_tlli(msg) $1 = 3822966132 (gdb) frame 4 #4 gprs_llc_tx_ui (msg=0x80963a8, sapi=1 '\001', command=0, mmctx=0x80932d0) at gprs_llc.c:496 496 return _bssgp_tx_dl_ud(msg, mmctx); (gdb) p mmctx $2 = (void *) 0x80932d0 (gdb) frame 3 #3 0x0805224f in _bssgp_tx_dl_ud (mmctx=0x80932d0, msg=0x80963a8) at gprs_llc.c:120 120 OSMO_ASSERT(msgb_tlli(msg) == mmctx->llme->tlli (gdb) p mmctx $3 = (struct sgsn_mm_ctx *) 0x80932d0 (gdb) p *mmctx $4 = {list = {next = 0x8092e28, prev = 0x805c318 <sgsn_mm_ctxts>}, imsi = "901700000003094\000", mm_state = GMM_REGISTERED_NORMAL, p_tmsi = 296043751, p_tmsi_old = 2075232571, p_tmsi_sig = 0, imei = "353943044782210\000", msisdn = '\000' <repeats 14 times>, ra = {mnc = 70, mcc = 901, lac = 1, rac = 0 '\000'}, cell_id = 0, cell_id_age = 0, sac = 0, sac_age = 0, new_sgsn_addr = 0, ciph_algo = GPRS_ALGO_GEA0, ms_radio_access_capa = {len = 11 '\v', buf = "4\307\003*\240B|\255\341\030\v", '\000' <repeats 38 times>}, ms_network_capa = { len = 2 '\002', buf = "\345\200\000\000\000\000\000"}, drx_parms = 3329, mnrg = 0, ngaf = 0, ppf = 0, recovery = 0, radio_prio_sms = 0 '\000', pdp_list = {next = 0x8093390, prev = 0x8093390}, llme = 0x8095630, tlli = 3822966132, tlli_new = 3517269223, nsei = 1801, bvci = 1801, ctrg = 0x8096048, timer = {node = {rb_parent_color = 3082574944, rb_right = 0x0, rb_left = 0x0}, list = {next = 0x80933b8, prev = 0x80933b8}, timeout = { tv_sec = 1375260414, tv_usec = 864196}, active = 1, cb = 0x804bfd0 <mmctx_timer_cb>, data = 0x80932d0}, T = 3350, num_T_exp = 0, t3350_mode = GMM_T3350_MODE_RAU, t3370_id_type = 1 '\001'} (gdb) p msgb_tlli(msg) No symbol "msgb_tlli" in current context. (gdb) frame 5 #5 0x0804a2f6 in gsm48_gmm_sendmsg (msg=0x80963a8, command=0, mm=0x80932d0) at gprs_gmm.c:241 241 return gprs_llc_tx_ui(msg, GPRS_SAPI_GMM, command, mm); (gdb) frame 6 #6 0x0804b687 in gsm48_tx_gmm_ra_upd_ack (mm=0x80932d0) at gprs_gmm.c:851 851 return gsm48_gmm_sendmsg(msg, 0, mm); (gdb) p msgb_tlli(msg) $5 = 3822966132 (gdb) p mmctx->tlli No symbol "mmctx" in current context. (gdb) p mm->tlli $6 = 3822966132 (gdb) p mm->tlli_new $7 = 3517269223 (gdb) p mm->llme->tlli $8 = 3517269223 (gdb) p mm->llme->tlli_old There is no member named tlli_old. (gdb) p mm->llme->old_tlli $9 = 4222716219 (gdb) bt #0 0xb7fde424 in __kernel_vsyscall () #1 0xb797f83f in __GI_raise (sig=sig@entry=6) at ../nptl/sysdeps/unix/sysv/linux/raise.c:56 #2 0xb7982cf3 in __GI_abort () at abort.c:90 #3 0x0805224f in _bssgp_tx_dl_ud (mmctx=0x80932d0, msg=0x80963a8) at gprs_llc.c:120 #4 gprs_llc_tx_ui (msg=0x80963a8, sapi=1 '\001', command=0, mmctx=0x80932d0) at gprs_llc.c:496 #5 0x0804a2f6 in gsm48_gmm_sendmsg (msg=0x80963a8, command=0, mm=0x80932d0) at gprs_gmm.c:241 #6 0x0804b687 in gsm48_tx_gmm_ra_upd_ack (mm=0x80932d0) at gprs_gmm.c:851 #7 0x0804bc35 in gsm48_rx_gmm_ra_upd_req (mmctx=0x80932d0, msg=0x8091ce8, llme=0x8095630) at gprs_gmm.c:1004 #8 0x0804bd4e in gsm0408_rcv_gmm (mmctx=0x80932d0, msg=0x8091ce8, llme=0x8095630) at gprs_gmm.c:1036 #9 0x0804d20c in gsm0408_gprs_rcvmsg (msg=msg@entry=0x8091ce8, llme=0x8095630) at gprs_gmm.c:1566 #10 0x08052be4 in gprs_llc_rcvmsg (msg=0x8091ce8, tv=0xbfffdcb0) at gprs_llc.c:882 #11 0x0804f5a4 in bssgp_prim_cb (oph=oph@entry=0xbfffdc8c, ctx=ctx@entry=0x0) at sgsn_main.c:114 #12 0xb7b626b8 in bssgp_rx_ul_ud (tp=0xbfffdcb0, msg=0x8091ce8, ctx=<optimized out>) at gprs_bssgp.c:398 #13 bssgp_rx_ptp (bctx=0x8091a08, tp=0xbfffdcb0, msg=0x8091ce8) at gprs_bssgp.c:820 #14 bssgp_rcvmsg (msg=0x8091ce8) at gprs_bssgp.c:1016 #15 0x0804f4f1 in sgsn_ns_cb (event=GPRS_NS_EVT_UNIT_DATA, nsvc=0x8090740, msg=0x8091ce8, bvci=1801) at sgsn_main.c:92 #16 0xb7b5ecf7 in gprs_ns_rx_unitdata (msg=0x8091ce8, nsvc=0x8090740) at gprs_ns.c:616 #17 gprs_ns_rcvmsg (nsi=nsi@entry=0x807fd38, msg=msg@entry=0x8091ce8, saddr=saddr@entry=0xbfffedc0, ll=ll@entry=GPRS_NS_LL_UDP) at gprs_ns.c:841 #18 0xb7b5f311 in handle_nsip_read (bfd=0x807fd58) at gprs_ns.c:991 #19 nsip_fd_cb (bfd=0x807fd58, what=1) at gprs_ns.c:1024 #20 0xb7bb2ed2 in osmo_select_main (polling=0) at select.c:158 #21 0x0804fbcd in main (argc=3, argv=0xbffff234) at sgsn_main.c:369 (gdb) frame 3 #3 0x0805224f in _bssgp_tx_dl_ud (mmctx=0x80932d0, msg=0x80963a8) at gprs_llc.c:120 120 OSMO_ASSERT(msgb_tlli(msg) == mmctx->llme->tlli (gdb) p msgb_tlli(msg) No symbol "msgb_tlli" in current context. (gdb) frame 4 #4 gprs_llc_tx_ui (msg=0x80963a8, sapi=1 '\001', command=0, mmctx=0x80932d0) at gprs_llc.c:496 496 return _bssgp_tx_dl_ud(msg, mmctx); (gdb) p msgb_tlli(msg) No symbol "msgb_tlli" in current context. (gdb) frame 5 #5 0x0804a2f6 in gsm48_gmm_sendmsg (msg=0x80963a8, command=0, mm=0x80932d0) at gprs_gmm.c:241 241 return gprs_llc_tx_ui(msg, GPRS_SAPI_GMM, command, mm); (gdb) p msgb_tlli(msg) $10 = 3822966132 (gdb) p mm->tlli $11 = 3822966132 (gdb) p mm->tlli_new $12 = 3517269223 (gdb) p mm->tlli_new $13 = 3517269223 (gdb) p mm->llme->tlli $14 = 3517269223 (gdb) p mm->llme->tlli_old There is no member named tlli_old. (gdb) p mm->llme->old_tlli $15 = 4222716219 (gdb) p mm->llme->state $16 = GPRS_LLMS_ASSIGNED (gdb) q
2013-07-31gprs_llc: Assert that we send frames with either tlli or old_tlliHolger Hans Peter Freyther1-15/+21
In case we have access to the context verify that the selected msgb_tlli is either the old_tlli or the tlli in either local or foreign format. It is wrong to use any other TLLI.
2013-07-31gprs_llc: Work on finding the right LLE/LLME in case of routing area updateHolger Hans Peter Freyther1-65/+94
Attempt to solve what f0901f0067e363c0ced6254db1b45a9771640412 tried to solve without breaking the case of someone with a foreign TLLI from a different network. Lookup with the foreign TLLI converted to a local one in case we did not find the TLLI and only then create a LLE/LLME on the fly for the RX path.
2013-07-31gprs_llc: Lookup lle based on the real TLLIHolger Hans Peter Freyther1-3/+1
During the GPRS Attach procedure we might have a foreign tlli and in the RX create a LLME on the fly for this tlli. The GMM GPRS Attach handling code will then assign a new TLLI and keep the foreign tlli as the llme->old_tlli. When the GMM is sending the identity request the msgb_tlli will point to the foreign tlli. The GPRS LLC code will then try to find that foreign tlli but due the conversion this will not be found. Instead a new ad-hoc LLE/LLME will be created on the fly for each message (this means there are duplicate LLE/LLMEs in the list). Make the code more strict and remove the tlli_foreign2local change from the look-up routine. This will make the GPRS LLC code find the right LLE/LLME and the N(U) will be handled correctly. This partially reverts: f0901f0067e363c0ced6254db1b45a9771640412 Addresses: <0012> gprs_llc.c:773 LLC RX: unknown TLLI 0xadf11820, creating LLME on the fly ... <0012> gprs_llc.c:357 LLC TX: unknown TLLI 0xedf11820, creating LLME on the fly Reproducable: Use pcu_emu (gprs attach) and observe with wireshark.
2013-07-31gprs_sgsn: In case of a Activate PDP Context timeout we should free pdpHolger Hans Peter Freyther1-0/+9
In case of a failure this method didn't set the pctx->lib back to NULL. In case of a timeout the callback will be made with pdp=NULL and this would lead to leaking the PDP context. Check for the case of having a pctx->lib != pdp and free it. This resolves: <000f> gprs_sgsn.c:259 freeing PDP context that still has a libgtp handle attached to it, this shouldn't happen!
2013-07-28gprs: Fix a typo in the commentHolger Hans Peter Freyther1-1/+1
2013-07-27tests: TestCase.assertGreater is not available on Python 2.5Holger Hans Peter Freyther1-2/+2
The jenkins build node has Python 2.5.X installed and the assertGreater method is not available. Use assert_ until we can use newer versions of Python.
2013-07-27expiration: Allow to disable the periodic location updating procedureHolger Hans Peter Freyther5-28/+101
Disable the periodic LU using "no periodic location update" VTY command. In that case set the expire_lu to 0 which will then be translated to a NULL in the database layer. This leads to a bit of copy and paste in the db_sync_subscriber method but I don't see how we could easily use 'datetime(%i, 'unixepoch')' and 'NULL' at the same time. Change the query to find expired queries to check for NOT NULL and the time being in the past. This means if there are still old subscribers in the database they might not be expired. One would need to execute a query like "UPATE Subscriber SET expire_lu = 0 WHERE expire_lu is null". The same applies when disabling the periodic LU. One would need to update the database by hand. Manual tests executed/passed: 1.) periodic LU enabled: * use gst LUTest.st to do a LU * UPDATE Subscriber SET expire_lu=datetime('now'); * observe the subscriber being expired (it was) 2.) periodic LU disabled: * use gst LUTest.st to do a LU * verify that the expire_lu is NULL in the database
2013-07-27expiration: Speculative fixes for the periodic expiring handlingHolger Hans Peter Freyther6-17/+66
We were expiring subscribers during active calls. This is because the T3212 is stopped under certain conditions but we didn't stop that timer at all. Remember if T3212 timer was stopped due something done by NITB and update the expiration time at the end of the radio connection, as the phone should restart it when returning to MM Idle. It is a bit difficult to decide when we should set the flag. E.g. in a CM Service Request we don't know if we accept the service and during a LU we already send MM messages before we accept or reject the subscriber. The easiest is to set the flag when receiving a paging response on known subscribers and at the end of the authentication process. Do not expire a subscriber that has an active connection that is marked with the flag, e.g. we would still expire a subscriber that is being paged. Manual tests executed/passed: * gst LUTest.st verified that a expiration date was set * gst SMSTest.st (doing another LU but forcing a timeout on the SMS sending). Verified that the expire_lu was updated.