[OpenSIPS-Users] ul_dump brief is not accepted, opensips 3.4
Hi, 1, looks like ul_dump brief params is not accepted OS: Ubuntu 22 opensips -V version: opensips 3.4.0-rc1 (x86_64/linux) flags: STATS: On, DISABLE_NAGLE, USE_MCAST, SHM_MMAP, PKG_MALLOC, Q_MALLOC, F_MALLOC, HP_MALLOC, DBG_MALLOC, CC_O0, FAST_LOCK-ADAPTIVE_WAIT ADAPTIVE_WAIT_LOOPS=1024, MAX_RECV_BUFFER_SIZE 262144, MAX_LISTEN 16, MAX_URI_SIZE 1024, BUF_SIZE 65535 poll method support: poll, epoll, sigio_rt, select. git revision: c64766a85 main.c compiled on with gcc 11 apt policy opensips-cli opensips-cli: Installed: 0.1~20230512~161b5ea-1 Candidate: 0.1~20230512~161b5ea-1 Version table: *** 0.1~20230512~161b5ea-1 500 500 https://apt.opensips.org jammy/cli-nightly amd64 Packages opensips-cli --version OpenSIPS CLI 0.2.0 opensips-cli -x mi ul_dump brief ERROR: command 'ul_dump' returned: -32602: Invalid params (Bad type for parameter 'brief') 2, When doing dr_reload through httpd, the load of the HTTPD ps will increase and other services rely on it like opensips prometheus will stuck and return nothing until dr_reload is completed curl -s -X POST -H 'Content-Type: application/json' {server-IP}:/mi -d'{"jsonrpc": "2.0","id": 1,"method": "dr_reload"}' | json_pp curl 127.0.0.1:/metrics Is it possible to increase the worker number of HTTPD ps? Regards, William Jin ___ Users mailing list Users@lists.opensips.org http://lists.opensips.org/cgi-bin/mailman/listinfo/users
Re: [OpenSIPS-Users] opensips 3.2.2 Segfault on debian 11 bullseye
Yes, I've updated to the latest version and can confirm the crash no longer happens. Thanks. -- Regards, William Jin From: Liviu Chircu Sent: Wednesday, 29 September 2021 8:01 PM To: William Jin ; OpenSIPS users mailling list Subject: Re: [OpenSIPS-Users] opensips 3.2.2 Segfault on debian 11 bullseye On 17.09.2021 07:42, William Jin wrote: Sep 17 14:33:23 ip-10-20-12-22 /usr/sbin/opensips[188225]: CRITICAL:core:sig_usr: segfault in process pid: 188225, id: 15 Hi William, Not sure if you noticed the identical thread below (with James Fox, Rob Dyck and Max) but this crash has been fixed in latest 3.2! Cheers, -- Liviu Chircu www.twitter.com/liviuchircu<http://www.twitter.com/liviuchircu> | www.opensips-solutions.com<http://www.opensips-solutions.com> OpenSIPS eBootcamp 2021 | www.opensips.org/training<http://www.opensips.org/training> ___ Users mailing list Users@lists.opensips.org http://lists.opensips.org/cgi-bin/mailman/listinfo/users
Re: [OpenSIPS-Users] opensips 3.2.2 Segfault on debian 11 bullseye
And I just enabled log_level=4 It connects to db (mysql 8.0.26-1debian11) and converting password string to val, then segfault Sep 17 14:33:23 ip-10-20-12-22 /usr/sbin/opensips[188225]: DBG:db_mysql:db_mysql_do_prepared_query: conn=0x7f6d33facd80 (tail=140107000235208) MC=0x7f6d33fac560 Sep 17 14:33:23 ip-10-20-12-22 /usr/sbin/opensips[188225]: DBG:db_mysql:db_mysql_do_prepared_query: new query=|select password,rpid from subscriber where username=?| Sep 17 14:33:23 ip-10-20-12-22 /usr/sbin/opensips[188225]: DBG:db_mysql:re_init_statement: query is , ptr=(nil) Sep 17 14:33:23 ip-10-20-12-22 /usr/sbin/opensips[188225]: DBG:db_mysql:mysql_raise_event: MySQL status has not changed: connected Sep 17 14:33:23 ip-10-20-12-22 /usr/sbin/opensips[188225]: DBG:db_mysql:db_mysql_do_prepared_query: new statement(0x7f6d33faf978) on connection: (0x7f6d33facd80) 0x7f6d33fac4c8 Sep 17 14:33:23 ip-10-20-12-22 /usr/sbin/opensips[188225]: DBG:db_mysql:db_mysql_do_prepared_query: set values for the statement run Sep 17 14:33:23 ip-10-20-12-22 /usr/sbin/opensips[188225]: DBG:db_mysql:db_mysql_val2bind: added val (0): len=10; type=254; is_null=0 Sep 17 14:33:23 ip-10-20-12-22 /usr/sbin/opensips[188225]: DBG:db_mysql:db_mysql_do_prepared_query: doing BIND_PARAM in... Sep 17 14:33:23 ip-10-20-12-22 /usr/sbin/opensips[188225]: DBG:db_mysql:mysql_raise_event: MySQL status has not changed: connected Sep 17 14:33:23 ip-10-20-12-22 /usr/sbin/opensips[188225]: DBG:db_mysql:db_mysql_do_prepared_query: prepared statement has 2 columns in result Sep 17 14:33:23 ip-10-20-12-22 /usr/sbin/opensips[188225]: DBG:db_mysql:db_mysql_do_prepared_query: doing to BIND_PARAM out ... Sep 17 14:33:23 ip-10-20-12-22 /usr/sbin/opensips[188225]: DBG:core:db_new_result: allocate 48 bytes for result set at 0x7f6d33fadf68 Sep 17 14:33:23 ip-10-20-12-22 /usr/sbin/opensips[188225]: DBG:db_mysql:db_mysql_get_columns: 2 columns returned from the query Sep 17 14:33:23 ip-10-20-12-22 /usr/sbin/opensips[188225]: DBG:core:db_allocate_columns: allocate 56 bytes for result columns at 0x7f6d33faf060 Sep 17 14:33:23 ip-10-20-12-22 /usr/sbin/opensips[188225]: DBG:db_mysql:db_mysql_get_columns: RES_NAMES(0x7f6d33faf070)[0]=[password] Sep 17 14:33:23 ip-10-20-12-22 /usr/sbin/opensips[188225]: DBG:db_mysql:db_mysql_get_columns: use DB_STRING result type Sep 17 14:33:23 ip-10-20-12-22 /usr/sbin/opensips[188225]: DBG:db_mysql:db_mysql_get_columns: RES_NAMES(0x7f6d33faf080)[1]=[rpid] Sep 17 14:33:23 ip-10-20-12-22 /usr/sbin/opensips[188225]: DBG:db_mysql:db_mysql_get_columns: use DB_STRING result type Sep 17 14:33:23 ip-10-20-12-22 /usr/sbin/opensips[188225]: DBG:core:db_allocate_rows: allocate 80 bytes for result rows and values at 0x7f6d33fb0450 Sep 17 14:33:23 ip-10-20-12-22 /usr/sbin/opensips[188225]: DBG:db_mysql:db_mysql_str2val: converting STRING [my@password] Sep 17 14:33:23 ip-10-20-12-22 /usr/sbin/opensips[188225]: CRITICAL:core:sig_usr: segfault in process pid: 188225, id: 15 -- Regards, William Jin From: Users on behalf of William Jin Sent: Friday, 17 September 2021 2:19 PM To: Liviu Chircu ; OpenSIPS users mailling list Subject: Re: [OpenSIPS-Users] opensips 3.2.2 Segfault on debian 11 bullseye And just further investigation, looks like it's crashed only after I register my test user. (which can register on debian 10 before upgraded to debian 11 using zoiper softphone) And the backtrack also indicate the crash from auth_db?? -- Regards, William Jin From: Users on behalf of William Jin Sent: Friday, 17 September 2021 9:12 AM To: Liviu Chircu ; OpenSIPS users mailling list Subject: Re: [OpenSIPS-Users] opensips 3.2.2 Segfault on debian 11 bullseye Hi Liviu, I attached the backtrace in the init email. I've attached it again. -- Regards, William Jin From: Liviu Chircu Sent: Thursday, 16 September 2021 5:12 PM To: OpenSIPS users mailling list ; William Jin Subject: Re: [OpenSIPS-Users] opensips 3.2.2 Segfault on debian 11 bullseye On 16.09.2021 02:16, William Jin wrote: > Anyone has any chance to look into this? Hi William, There is not enough info to look into anything, really: * how are you reproducing this crash? can you share a basic opensips.cfg that causes it? * are there any useful logs _before_ those "signal 11" logs? * can you obtain a backtrace out of this crash? See this guide [1] for some help here [1]: https://www.opensips.org/Documentation/TroubleShooting-Crash Best, -- Liviu Chircu www.twitter.com/liviuchircu<http://www.twitter.com/liviuchircu> | www.opensips-solutions.com<http://www.opensips-solutions.com> ___ Users mailing list Users@lists.opensips.org http://lists.opensips.org/cgi-bin/mailman/listinfo/users
Re: [OpenSIPS-Users] opensips 3.2.2 Segfault on debian 11 bullseye
And just further investigation, looks like it's crashed only after I register my test user. (which can register on debian 10 before upgraded to debian 11 using zoiper softphone) And the backtrack also indicate the crash from auth_db?? -- Regards, William Jin From: Users on behalf of William Jin Sent: Friday, 17 September 2021 9:12 AM To: Liviu Chircu ; OpenSIPS users mailling list Subject: Re: [OpenSIPS-Users] opensips 3.2.2 Segfault on debian 11 bullseye Hi Liviu, I attached the backtrace in the init email. I've attached it again. -- Regards, William Jin From: Liviu Chircu Sent: Thursday, 16 September 2021 5:12 PM To: OpenSIPS users mailling list ; William Jin Subject: Re: [OpenSIPS-Users] opensips 3.2.2 Segfault on debian 11 bullseye On 16.09.2021 02:16, William Jin wrote: > Anyone has any chance to look into this? Hi William, There is not enough info to look into anything, really: * how are you reproducing this crash? can you share a basic opensips.cfg that causes it? * are there any useful logs _before_ those "signal 11" logs? * can you obtain a backtrace out of this crash? See this guide [1] for some help here [1]: https://www.opensips.org/Documentation/TroubleShooting-Crash Best, -- Liviu Chircu www.twitter.com/liviuchircu<http://www.twitter.com/liviuchircu> | www.opensips-solutions.com<http://www.opensips-solutions.com> ___ Users mailing list Users@lists.opensips.org http://lists.opensips.org/cgi-bin/mailman/listinfo/users
Re: [OpenSIPS-Users] opensips 3.2.2 Segfault on debian 11 bullseye
Hi Liviu, I attached the backtrace in the init email. I've attached it again. -- Regards, William Jin From: Liviu Chircu Sent: Thursday, 16 September 2021 5:12 PM To: OpenSIPS users mailling list ; William Jin Subject: Re: [OpenSIPS-Users] opensips 3.2.2 Segfault on debian 11 bullseye On 16.09.2021 02:16, William Jin wrote: > Anyone has any chance to look into this? Hi William, There is not enough info to look into anything, really: * how are you reproducing this crash? can you share a basic opensips.cfg that causes it? * are there any useful logs _before_ those "signal 11" logs? * can you obtain a backtrace out of this crash? See this guide [1] for some help here [1]: https://www.opensips.org/Documentation/TroubleShooting-Crash Best, -- Liviu Chircu www.twitter.com/liviuchircu<http://www.twitter.com/liviuchircu> | www.opensips-solutions.com<http://www.opensips-solutions.com> gdb /usr/sbin/opensips /tmp/core GNU gdb (Debian 10.1-1.7) 10.1.90.20210103-git Copyright (C) 2021 Free Software Foundation, Inc. License GPLv3+: GNU GPL version 3 or later <http://gnu.org/licenses/gpl.html> This is free software: you are free to change and redistribute it. There is NO WARRANTY, to the extent permitted by law. Type "show copying" and "show warranty" for details. This GDB was configured as "x86_64-linux-gnu". Type "show configuration" for configuration details. For bug reporting instructions, please see: <https://www.gnu.org/software/gdb/bugs/>. Find the GDB manual and other documentation resources online at: <http://www.gnu.org/software/gdb/documentation/>. For help, type "help". Type "apropos word" to search for commands related to "word"... Reading symbols from /usr/sbin/opensips... Reading symbols from /usr/lib/debug/.build-id/68/43b0716a68e5fab40b5fba966d9bce8d0e224f.debug... warning: Can't open file /dev/zero (deleted) during file-backed mapping note processing [New LWP 10586] [Thread debugging using libthread_db enabled] Using host libthread_db library "/lib/x86_64-linux-gnu/libthread_db.so.1". Core was generated by `/usr/sbin/opensips -P /run/opensips/opensips.pid -f /etc/opensips/opensips.cfg'. Program terminated with signal SIGSEGV, Segmentation fault. #0 0x559ddc5502a0 in memcpy (__len=64, __src=0xaf01ff, __dest=0x7ffd97072be8) at /usr/include/x86_64-linux-gnu/bits/string_fortified.h:34 34return __builtin___memcpy_chk (__dest, __src, __len, __bos0 (__dest)); (gdb) bt full #0 0x559ddc5502a0 in memcpy (__len=64, __src=0xaf01ff, __dest=0x7ffd97072be8) at /usr/include/x86_64-linux-gnu/bits/string_fortified.h:34 No locals. #1 MD5_memcpy (len=64, input=0xaf01ff , output=0x7ffd97072be8 "\240-\a\227\375\177") at md5.c:329 No locals. #2 MD5Update (context=0x7ffd97072bd0, input=0xaf01ff , inputLen=127) at md5.c:152 i = index = 0 partLen = 64 #3 0x7f67bad9a788 in digest_calc_HA1 (crd=0x7ffd97072d60, sess_key=0x7ffd97072de0) at dauth_calc_md5.c:48 Md5Ctx = {state = {1732584193, 4023233417, 2562383102, 271733878}, count = {1016, 0}, buffer = "\240-\a\227\375\177\000\000(-\a\227\375\177\000\000\000\305L\321\027`\200P\220-\a\227\375\177\000\000\200\355\003\277g\177\000\000(-\a\227\375\177\000\000\310\344\003\277g\177\000\000(-\a\227\375\177\000"} HA1 = "(-\a\227\375\177\000\000z\247\244\272g\177\000" #4 0x7f67bad9476d in auth_calc_HA1 (params=0x7ffd97072d40, sess_key=0x7ffd97072de0) at api.c:341 digest_calc = 0x7f67bada2800 __FUNCTION__ = "auth_calc_HA1" #5 0x7f67baa8e936 in get_ha1 (res=0x7ffd97072d28, _ha1=0x7ffd97072de0, _table=, _domain=0x7ffd97072f88, digest=0x7f67bf040f58) at authorize.c:169 cprms = {use_hashed = 0, alg = ALG_MD5, creds = {open = 0x7ffd97072d60, ha1 = 0x7ffd97072d60}, nonce = 0x7f67bf040f98, cnonce = 0x7f67bf040fe0} vals = {{type = DB_STR, nul = 0, free = 0, val = {int_val = -594864196, bigint_val = 94136498329532, double_val = 4.6509609844413654e-310, time_val = 94136498329532, string_val = 0x559ddc8b17bc "1122334455\",realm=\"sip1.mydomain.com\",nonce=\"KprTqedqhYxO7vVNGoDdw6U2mvUQ1v6Jd7liqtOCO94A\",uri=\"sip:sip1.mydomain.com;transport=UDP\",response=\"46f25ec4472d495ed41f46677a1eb647\",algorithm=MD5"..., str_val = { s = 0x559ddc8b17bc "1122334455\",realm=\"sip1.mydomain.com\",nonce=\"KprTqedqhYxO7vVNGoDdw6U2mvUQ1v6Jd7liqtOCO94A\",uri=\"sip:sip1.mydomain.com;transport=UDP\",response=\"46f25ec4472d495ed41f46677a1eb647\",algorithm=MD5"..., len = 10}, blob_val = { s = 0x559ddc8b17bc "1122334455\",realm=\"sip1.mydomain.com\",nonce=\"KprTqedqhYxO
Re: [OpenSIPS-Users] opensips 3.2.2 Segfault on debian 11 bullseye
Anyone has any chance to look into this? -- Regards, William Jin From: Users on behalf of William Jin Sent: Thursday, 9 September 2021 12:32 PM To: users@lists.opensips.org Subject: [OpenSIPS-Users] opensips 3.2.2 Segfault on debian 11 bullseye We currently upgraded our Debian server from 10(buster) to 11(bullseye), After the upgrade, the Opensips keep crashing with a segfault. Sep 9 10:35:04 ip-10-20-12-22 /usr/sbin/opensips[8700]: CRITICAL:core:sig_usr: segfault in process pid: 8700, id: 16 Sep 9 10:35:04 ip-10-20-12-22 /usr/sbin/opensips[8683]: INFO:core:handle_sigs: child process 8700 exited by a signal 11 Sep 9 10:35:04 ip-10-20-12-22 /usr/sbin/opensips[8683]: INFO:core:handle_sigs: core was generated Sep 9 10:35:04 ip-10-20-12-22 /usr/sbin/opensips[8683]: INFO:core:handle_sigs: terminating due to SIGCHLD We are using the apt repo opensips, tried both stable and nightly version, same result. opensips -V version: opensips 3.2.2 (x86_64/linux) flags: STATS: On, DISABLE_NAGLE, USE_MCAST, SHM_MMAP, PKG_MALLOC, Q_MALLOC, F_MALLOC, HP_MALLOC, DBG_MALLOC, FAST_LOCK-ADAPTIVE_WAIT ADAPTIVE_WAIT_LOOPS=1024, MAX_RECV_BUFFER_SIZE 262144, MAX_LISTEN 16, MAX_URI_SIZE 1024, BUF_SIZE 65535 poll method support: poll, epoll, sigio_rt, select. git revision: dc45f112e main.c compiled on with gcc 10 DBG attached PS: I've replaced my number to 1122334455, domain to sip1.mydomain.com -- Regards, William Jin ___ Users mailing list Users@lists.opensips.org http://lists.opensips.org/cgi-bin/mailman/listinfo/users
[OpenSIPS-Users] opensips 3.2.2 Segfault on debian 11 bullseye
We currently upgraded our Debian server from 10(buster) to 11(bullseye), After the upgrade, the Opensips keep crashing with a segfault. Sep 9 10:35:04 ip-10-20-12-22 /usr/sbin/opensips[8700]: CRITICAL:core:sig_usr: segfault in process pid: 8700, id: 16 Sep 9 10:35:04 ip-10-20-12-22 /usr/sbin/opensips[8683]: INFO:core:handle_sigs: child process 8700 exited by a signal 11 Sep 9 10:35:04 ip-10-20-12-22 /usr/sbin/opensips[8683]: INFO:core:handle_sigs: core was generated Sep 9 10:35:04 ip-10-20-12-22 /usr/sbin/opensips[8683]: INFO:core:handle_sigs: terminating due to SIGCHLD We are using the apt repo opensips, tried both stable and nightly version, same result. opensips -V version: opensips 3.2.2 (x86_64/linux) flags: STATS: On, DISABLE_NAGLE, USE_MCAST, SHM_MMAP, PKG_MALLOC, Q_MALLOC, F_MALLOC, HP_MALLOC, DBG_MALLOC, FAST_LOCK-ADAPTIVE_WAIT ADAPTIVE_WAIT_LOOPS=1024, MAX_RECV_BUFFER_SIZE 262144, MAX_LISTEN 16, MAX_URI_SIZE 1024, BUF_SIZE 65535 poll method support: poll, epoll, sigio_rt, select. git revision: dc45f112e main.c compiled on with gcc 10 DBG attached PS: I've replaced my number to 1122334455, domain to sip1.mydomain.com -- Regards, William Jin gdb /usr/sbin/opensips /tmp/core GNU gdb (Debian 10.1-1.7) 10.1.90.20210103-git Copyright (C) 2021 Free Software Foundation, Inc. License GPLv3+: GNU GPL version 3 or later <http://gnu.org/licenses/gpl.html> This is free software: you are free to change and redistribute it. There is NO WARRANTY, to the extent permitted by law. Type "show copying" and "show warranty" for details. This GDB was configured as "x86_64-linux-gnu". Type "show configuration" for configuration details. For bug reporting instructions, please see: <https://www.gnu.org/software/gdb/bugs/>. Find the GDB manual and other documentation resources online at: <http://www.gnu.org/software/gdb/documentation/>. For help, type "help". Type "apropos word" to search for commands related to "word"... Reading symbols from /usr/sbin/opensips... Reading symbols from /usr/lib/debug/.build-id/68/43b0716a68e5fab40b5fba966d9bce8d0e224f.debug... warning: Can't open file /dev/zero (deleted) during file-backed mapping note processing [New LWP 10586] [Thread debugging using libthread_db enabled] Using host libthread_db library "/lib/x86_64-linux-gnu/libthread_db.so.1". Core was generated by `/usr/sbin/opensips -P /run/opensips/opensips.pid -f /etc/opensips/opensips.cfg'. Program terminated with signal SIGSEGV, Segmentation fault. #0 0x559ddc5502a0 in memcpy (__len=64, __src=0xaf01ff, __dest=0x7ffd97072be8) at /usr/include/x86_64-linux-gnu/bits/string_fortified.h:34 34return __builtin___memcpy_chk (__dest, __src, __len, __bos0 (__dest)); (gdb) bt full #0 0x559ddc5502a0 in memcpy (__len=64, __src=0xaf01ff, __dest=0x7ffd97072be8) at /usr/include/x86_64-linux-gnu/bits/string_fortified.h:34 No locals. #1 MD5_memcpy (len=64, input=0xaf01ff , output=0x7ffd97072be8 "\240-\a\227\375\177") at md5.c:329 No locals. #2 MD5Update (context=0x7ffd97072bd0, input=0xaf01ff , inputLen=127) at md5.c:152 i = index = 0 partLen = 64 #3 0x7f67bad9a788 in digest_calc_HA1 (crd=0x7ffd97072d60, sess_key=0x7ffd97072de0) at dauth_calc_md5.c:48 Md5Ctx = {state = {1732584193, 4023233417, 2562383102, 271733878}, count = {1016, 0}, buffer = "\240-\a\227\375\177\000\000(-\a\227\375\177\000\000\000\305L\321\027`\200P\220-\a\227\375\177\000\000\200\355\003\277g\177\000\000(-\a\227\375\177\000\000\310\344\003\277g\177\000\000(-\a\227\375\177\000"} HA1 = "(-\a\227\375\177\000\000z\247\244\272g\177\000" #4 0x7f67bad9476d in auth_calc_HA1 (params=0x7ffd97072d40, sess_key=0x7ffd97072de0) at api.c:341 digest_calc = 0x7f67bada2800 __FUNCTION__ = "auth_calc_HA1" #5 0x7f67baa8e936 in get_ha1 (res=0x7ffd97072d28, _ha1=0x7ffd97072de0, _table=, _domain=0x7ffd97072f88, digest=0x7f67bf040f58) at authorize.c:169 cprms = {use_hashed = 0, alg = ALG_MD5, creds = {open = 0x7ffd97072d60, ha1 = 0x7ffd97072d60}, nonce = 0x7f67bf040f98, cnonce = 0x7f67bf040fe0} vals = {{type = DB_STR, nul = 0, free = 0, val = {int_val = -594864196, bigint_val = 94136498329532, double_val = 4.6509609844413654e-310, time_val = 94136498329532, string_val = 0x559ddc8b17bc "1122334455\",realm=\"sip1.mydomain.com\",nonce=\"KprTqedqhYxO7vVNGoDdw6U2mvUQ1v6Jd7liqtOCO94A\",uri=\"sip:sip1.mydomain.com;transport=UDP\",response=\"46f25ec4472d495ed41f46677a1eb647\",algorithm=MD5"..., str_val = { s = 0x559ddc8b17bc "1122334455\",realm=\"sip1.mydomain.com\",nonce=\"KprTqedqhYxO7vVNGoDdw6U2mvUQ1v6Jd7liqtOCO94A\",uri=\"sip:sip1.mydomain.com;transport=UDP\",respo
Re: [OpenSIPS-Users] Usrloc can't be sync after upgrade to 2.4.8
Found some extra error logs that might be related to this. checking how we can fix this /usr/sbin/opensips[1983]: ERROR:proto_bin:add_write_chunk: We have reached the limit of max async postponed chunks /usr/sbin/opensips[1983]: ERROR:proto_bin:proto_bin_send: failed to send /usr/sbin/opensips[1983]: ERROR:clusterer:msg_send: send() to 10.2.2.16:5566 for proto bin/7 failed -- Regards, William Jin From: William Jin Sent: Thursday, 2 July 2020 11:24 AM To: OpenSIPS users mailling list Subject: Re: [OpenSIPS-Users] Usrloc can't be sync after upgrade to 2.4.8 Hi, Thanks for the info. I have upgraded both nodes to latest 2.4.8, and restarted opensips on both nodes so it should start from empty. The error "INFO:usrloc:receive_binary_packets: discarding , ver 1: need ver 2" is gone. However, the 2nd node usrloc is still not synced with 1st node(seed), even manually issue the command opensipsctl fifo ul_cluster_sync. my config: (on both node) # - usrloc params - modparam("usrloc", "cluster_mode", "full-sharing") modparam("usrloc", "nat_bflag", "NAT") modparam("usrloc", "location_cluster", 1) modparam("usrloc", "restart_persistency", "sync-from-cluster") The clusterer table: id | cluster_id | node_id |url | state | no_ping_retries | priority | sip_addr | description | flags ++-++---+-+--+--+-+--- 2 | 1 | 2 | bin:10.2.2.16:5566 | 1 | 3 | 50 | | node2 | 1 | 1 | 1 | bin:10.2.2.15:5566 | 1 | 3 | 50 | | node1 | seed >From node1 log: Jul 2 10:55:56 /usr/sbin/opensips[1983]: INFO:clusterer:handle_internal_msg: Node [2] is UP Jul 2 10:55:56 /usr/sbin/opensips[1983]: INFO:clusterer:handle_sync_request: Received sync request for capability 'dialog-dlg-repl' from node 2, cluster 1 Jul 2 10:55:56 /usr/sbin/opensips[1983]: INFO:clusterer:send_sync_repl: Sent all sync packets for capability 'dialog-dlg-repl' to node 2, cluster 1 Jul 2 10:55:56 /usr/sbin/opensips[1983]: INFO:clusterer:handle_sync_request: Received sync request for capability 'usrloc-contact-repl' from node 2, cluster 1 Jul 2 10:55:56 /usr/sbin/opensips[1983]: ERROR:clusterer:msg_send: send() to 10.2.2.16:5566 for proto bin/7 failed Jul 2 10:55:56 /usr/sbin/opensips[1983]: ERROR:clusterer:msg_send_retry: msg_send() to node [2] failed Jul 2 10:55:56 /usr/sbin/opensips[1983]: ERROR:clusterer:cl_sync_chunk_start: Failed to send sync packet ... Jul 2 10:55:56 /usr/sbin/opensips[1983]: ERROR:clusterer:cl_sync_chunk_start: Failed to send sync packet Jul 2 10:55:56 /usr/sbin/opensips[1983]: ERROR:clusterer:send_sync_repl: Failed to send sync packet, rc=-1 Jul 2 10:55:56 /usr/sbin/opensips[1983]: ERROR:clusterer:send_sync_repl: Failed to send sync end message Jul 2 10:55:56 /usr/sbin/opensips[1983]: INFO:clusterer:handle_internal_msg: Node [2] is UP Node2 log: Jul 2 10:55:56 /usr/sbin/opensips[13782]: INFO:clusterer:handle_internal_msg: Node [1] is UP Jul 2 10:55:56 /usr/sbin/opensips[13782]: INFO:clusterer:send_sync_req: Sent sync request for capability 'dialog-dlg-repl' to node 1, cluster 1 Jul 2 10:55:56 /usr/sbin/opensips[13782]: INFO:clusterer:send_sync_req: Sent sync request for capability 'usrloc-contact-repl' to node 1, cluster 1 Jul 2 10:55:56 /usr/sbin/opensips[13782]: INFO:clusterer:handle_sync_packet: Received all sync packets for capability 'dialog-dlg-repl' in cluster 1 Jul 2 10:55:56 /usr/sbin/opensips[13793]: INFO:core:probe_max_sock_buff: using snd buffer of 512 kb Jul 2 10:55:56 /usr/sbin/opensips[13793]: INFO:core:init_sock_keepalive: TCP keepalive enabled on socket 184 we have around 30,000 end devices(UDP), and I found the https://github.com/OpenSIPS/opensips/issues/1976, not sure if it's related Any idea? Do I need to increase the ping_timeout and give it a try? -- Regards, William Jin From: Users on behalf of Liviu Chircu Sent: Wednesday, 1 July 2020 4:48 PM To: OpenSIPS users mailling list Subject: Re: [OpenSIPS-Users] Usrloc can't be sync after upgrade to 2.4.8 On 01.07.2020 09:39, Liviu Chircu wrote: > then re-register them manually with the "ul_add" [2] command that and > some Python scripting logic, for example. Sorry, I meant bash, not Python. Not that I have anything against Python, but the former script (basically a simple loop) is much quicker to write in order to achieve your purpose. -- Liviu Chircu www.twitter.com/liviuchircu<http://www.twitter.com/liviuchircu> | www.opensips-solutions.com<http://www.opensips-solutions.com> ___ Users mailing list Users@lists.opensips.org ht
Re: [OpenSIPS-Users] Usrloc can't be sync after upgrade to 2.4.8
Hi, Thanks for the info. I have upgraded both nodes to latest 2.4.8, and restarted opensips on both nodes so it should start from empty. The error "INFO:usrloc:receive_binary_packets: discarding , ver 1: need ver 2" is gone. However, the 2nd node usrloc is still not synced with 1st node(seed), even manually issue the command opensipsctl fifo ul_cluster_sync. my config: (on both node) # - usrloc params - modparam("usrloc", "cluster_mode", "full-sharing") modparam("usrloc", "nat_bflag", "NAT") modparam("usrloc", "location_cluster", 1) modparam("usrloc", "restart_persistency", "sync-from-cluster") The clusterer table: id | cluster_id | node_id |url | state | no_ping_retries | priority | sip_addr | description | flags ++-++---+-+--+--+-+--- 2 | 1 | 2 | bin:10.2.2.16:5566 | 1 | 3 | 50 | | node2 | 1 | 1 | 1 | bin:10.2.2.15:5566 | 1 | 3 | 50 | | node1 | seed >From node1 log: Jul 2 10:55:56 /usr/sbin/opensips[1983]: INFO:clusterer:handle_internal_msg: Node [2] is UP Jul 2 10:55:56 /usr/sbin/opensips[1983]: INFO:clusterer:handle_sync_request: Received sync request for capability 'dialog-dlg-repl' from node 2, cluster 1 Jul 2 10:55:56 /usr/sbin/opensips[1983]: INFO:clusterer:send_sync_repl: Sent all sync packets for capability 'dialog-dlg-repl' to node 2, cluster 1 Jul 2 10:55:56 /usr/sbin/opensips[1983]: INFO:clusterer:handle_sync_request: Received sync request for capability 'usrloc-contact-repl' from node 2, cluster 1 Jul 2 10:55:56 /usr/sbin/opensips[1983]: ERROR:clusterer:msg_send: send() to 10.2.2.16:5566 for proto bin/7 failed Jul 2 10:55:56 /usr/sbin/opensips[1983]: ERROR:clusterer:msg_send_retry: msg_send() to node [2] failed Jul 2 10:55:56 /usr/sbin/opensips[1983]: ERROR:clusterer:cl_sync_chunk_start: Failed to send sync packet ... Jul 2 10:55:56 /usr/sbin/opensips[1983]: ERROR:clusterer:cl_sync_chunk_start: Failed to send sync packet Jul 2 10:55:56 /usr/sbin/opensips[1983]: ERROR:clusterer:send_sync_repl: Failed to send sync packet, rc=-1 Jul 2 10:55:56 /usr/sbin/opensips[1983]: ERROR:clusterer:send_sync_repl: Failed to send sync end message Jul 2 10:55:56 /usr/sbin/opensips[1983]: INFO:clusterer:handle_internal_msg: Node [2] is UP Node2 log: Jul 2 10:55:56 /usr/sbin/opensips[13782]: INFO:clusterer:handle_internal_msg: Node [1] is UP Jul 2 10:55:56 /usr/sbin/opensips[13782]: INFO:clusterer:send_sync_req: Sent sync request for capability 'dialog-dlg-repl' to node 1, cluster 1 Jul 2 10:55:56 /usr/sbin/opensips[13782]: INFO:clusterer:send_sync_req: Sent sync request for capability 'usrloc-contact-repl' to node 1, cluster 1 Jul 2 10:55:56 /usr/sbin/opensips[13782]: INFO:clusterer:handle_sync_packet: Received all sync packets for capability 'dialog-dlg-repl' in cluster 1 Jul 2 10:55:56 /usr/sbin/opensips[13793]: INFO:core:probe_max_sock_buff: using snd buffer of 512 kb Jul 2 10:55:56 /usr/sbin/opensips[13793]: INFO:core:init_sock_keepalive: TCP keepalive enabled on socket 184 we have around 30,000 end devices(UDP), and I found the https://github.com/OpenSIPS/opensips/issues/1976, not sure if it's related Any idea? Do I need to increase the ping_timeout and give it a try? -- Regards, William Jin From: Users on behalf of Liviu Chircu Sent: Wednesday, 1 July 2020 4:48 PM To: OpenSIPS users mailling list Subject: Re: [OpenSIPS-Users] Usrloc can't be sync after upgrade to 2.4.8 On 01.07.2020 09:39, Liviu Chircu wrote: > then re-register them manually with the "ul_add" [2] command that and > some Python scripting logic, for example. Sorry, I meant bash, not Python. Not that I have anything against Python, but the former script (basically a simple loop) is much quicker to write in order to achieve your purpose. -- Liviu Chircu www.twitter.com/liviuchircu<http://www.twitter.com/liviuchircu> | www.opensips-solutions.com<http://www.opensips-solutions.com> ___ Users mailing list Users@lists.opensips.org http://lists.opensips.org/cgi-bin/mailman/listinfo/users ___ Users mailing list Users@lists.opensips.org http://lists.opensips.org/cgi-bin/mailman/listinfo/users
[OpenSIPS-Users] Usrloc can't be sync after upgrade to 2.4.8
Hi, We were using opensips 2.4.6 from apt repo, and we upgraded to 2.4.8 today. However, after the upgrade, the usrloc-contact-repl can not be synced between the 2 nodes within 1 cluster. Even we issue the ul_cluster_sync command. opensipsctl fifo clusterer_list_cap Cluster:: 1 Capability:: usrloc-contact-repl State=not synced And we found the below log INFO:usrloc:receive_binary_packets: discarding , ver 1: need ver 2 Is there anything I missed? where should I change the version to 2? -- Regards, William Jin ___ Users mailing list Users@lists.opensips.org http://lists.opensips.org/cgi-bin/mailman/listinfo/users
Re: [OpenSIPS-Users] question about rtpengine_manage() in failure_route
Hi Razvan, Many thanks for your advice. Yes, I can confirm the branch_route did the trick. -- Regards, William Jin From: Users on behalf of Răzvan Crainea Sent: Wednesday, 10 June 2020 5:59 PM To: users@lists.opensips.org Subject: Re: [OpenSIPS-Users] question about rtpengine_manage() in failure_route Hi, William! Make sure you call rtpengine_manage() in branch route, not in the request route, otherwise all the changes will be inherited throughout all the branches further created. Best regards, Răzvan On 6/9/20 7:04 AM, William Jin wrote: Below is a short config example. route { ... setflag(CallFWD); rtpengine_manage(); #say the first call attempt need rtpengine t_on_failure("handle_failure"); r_relay(); ... } failure_route[handle_failure] { ... if (isflagset(CallFWD)){route(handle_callfwd);} ... } route[handle_callfwd]{ ... xlog("L_INFO","RB=$rb(application/sdp)"); if (t_relay()){ xlog("L_INFO","Stateful relay done. RB=$rb(application/sdp)"); } ... } Looks like the t_relay in the handle_callfwd will inherit the SDP info that rtpengine_manage entered in the first attempt. Also, another interesting fact is that the SDP info is added while the t_relay is called. Looks like the t_relay is using its data in memory and re-write the SDP. the $rb(application/sdp) is not changed until the t_relay is called. How can I, for example, remove the rtpengine related SDP for the second INVITE generated by the failure_route? I tried rtpengine_delete() in handle_callfwd or rtpengine_manage() in the failure_route which suppose to do rtpengine_delete also, but they don't work as expected. Thanks in advance. -- Regards, William Jin From: Users <mailto:users-boun...@lists.opensips.org> on behalf of William Jin <mailto:willi...@exetel.com.au> Sent: Tuesday, 9 June 2020 9:09 AM To: OpenSIPS users mailling list <mailto:users@lists.opensips.org> Subject: [OpenSIPS-Users] question about rtpengine_manage() in failure_route Hi All May I know how can I rewrite the SDP (rtpengine) in the failure route? The scenario is we use rtpengine_manage() in the first call attempt, if it fails, it uses failure_route, however, we want to change the SDP info. For example, the call's first attempt is to an ipv6 UAC, when failed, we try ipv4 UAC. We need to change the rtpengine address-family to IP4 so the c= line can follow with an IPv4 address. We tried to use rtpengine_manage("address-family=IP4"), but looks like the SDP still not changed. Does anyone have any idea about this? Or is there any other way to achieve this? -- Regards, William Jin ___ Users mailing list Users@lists.opensips.org<mailto:Users@lists.opensips.org> http://lists.opensips.org/cgi-bin/mailman/listinfo/users ___ Users mailing list Users@lists.opensips.org http://lists.opensips.org/cgi-bin/mailman/listinfo/users
Re: [OpenSIPS-Users] question about rtpengine_manage() in failure_route
Below is a short config example. route { ... setflag(CallFWD); rtpengine_manage(); #say the first call attempt need rtpengine t_on_failure("handle_failure"); r_relay(); ... } failure_route[handle_failure] { ... if (isflagset(CallFWD)){route(handle_callfwd);} ... } route[handle_callfwd]{ ... xlog("L_INFO","RB=$rb(application/sdp)"); if (t_relay()){ xlog("L_INFO","Stateful relay done. RB=$rb(application/sdp)"); } ... } Looks like the t_relay in the handle_callfwd will inherit the SDP info that rtpengine_manage entered in the first attempt. Also, another interesting fact is that the SDP info is added while the t_relay is called. Looks like the t_relay is using its data in memory and re-write the SDP. the $rb(application/sdp) is not changed until the t_relay is called. How can I, for example, remove the rtpengine related SDP for the second INVITE generated by the failure_route? I tried rtpengine_delete() in handle_callfwd or rtpengine_manage() in the failure_route which suppose to do rtpengine_delete also, but they don't work as expected. Thanks in advance. -- Regards, William Jin ________ From: Users on behalf of William Jin Sent: Tuesday, 9 June 2020 9:09 AM To: OpenSIPS users mailling list Subject: [OpenSIPS-Users] question about rtpengine_manage() in failure_route Hi All May I know how can I rewrite the SDP (rtpengine) in the failure route? The scenario is we use rtpengine_manage() in the first call attempt, if it fails, it uses failure_route, however, we want to change the SDP info. For example, the call's first attempt is to an ipv6 UAC, when failed, we try ipv4 UAC. We need to change the rtpengine address-family to IP4 so the c= line can follow with an IPv4 address. We tried to use rtpengine_manage("address-family=IP4"), but looks like the SDP still not changed. Does anyone have any idea about this? Or is there any other way to achieve this? -- Regards, William Jin ___ Users mailing list Users@lists.opensips.org http://lists.opensips.org/cgi-bin/mailman/listinfo/users
[OpenSIPS-Users] question about rtpengine_manage() in failure_route
Hi All May I know how can I rewrite the SDP (rtpengine) in the failure route? The scenario is we use rtpengine_manage() in the first call attempt, if it fails, it uses failure_route, however, we want to change the SDP info. For example, the call's first attempt is to an ipv6 UAC, when failed, we try ipv4 UAC. We need to change the rtpengine address-family to IP4 so the c= line can follow with an IPv4 address. We tried to use rtpengine_manage("address-family=IP4"), but looks like the SDP still not changed. Does anyone have any idea about this? Or is there any other way to achieve this? -- Regards, William Jin ___ Users mailing list Users@lists.opensips.org http://lists.opensips.org/cgi-bin/mailman/listinfo/users
Re: [OpenSIPS-Users] Question about IP Transit
OK, I think I figured it out myself. Have to use force_send_socket(udp:ipv4 address); -- Regards, William Jin From: Users on behalf of William Jin Sent: Thursday, 4 June 2020 1:35 PM To: OpenSIPS users mailling list Subject: Re: [OpenSIPS-Users] Question about IP Transit Hi, Does anyone have any idea? I can send the sip trace if needed. -- Regards, William Jin From: Users on behalf of William Jin Sent: Tuesday, 2 June 2020 2:48 PM To: OpenSIPS users mailling list Subject: Re: [OpenSIPS-Users] Question about IP Transit Just additional info: A(v4) => opensips(v4/v6) => B(v6) => C(v6) - also working Does that mean if the B number is V6, I need to send the invite to the v6 IP of the opensips? Then the server should handle the ip transit from v6 to v4 again? -- Regards, William Jin From: Users on behalf of William Jin Sent: Tuesday, 2 June 2020 2:07 PM To: OpenSIPS users mailling list Subject: [OpenSIPS-Users] Question about IP Transit Hi We are using opensips 2.4.6 at the moment. It's the apt standard(not nightly) repo of debian opensips -V version: opensips 2.4.6 (x86_64/linux) flags: STATS: On, DISABLE_NAGLE, USE_MCAST, SHM_MMAP, PKG_MALLOC, F_MALLOC, FAST_LOCK-ADAPTIVE_WAIT ADAPTIVE_WAIT_LOOPS=1024, MAX_RECV_BUFFER_SIZE 262144, MAX_LISTEN 16, MAX_URI_SIZE 1024, BUF_SIZE 65535 poll method support: poll, epoll, sigio_rt, select. main.c compiled on with gcc 6.3.0 The situation is: Our sip server has both IPv4 and IPv6 addresses, it handles IP transit pretty well. Say if there is an incoming call from the gateway A (asterisk IPv4), it can do the IP transit to the destination number registered in IPv6 A(v4) <=> opensips(v4<=>v6) <=> B(v6) We are using $T_fr_inv_timeout to do call forwarding if no answer. The issue here is: when the B has a call forwarding setting and if the destination number, let's say C is v4, then the t_relay() of the INVITE after $T_fr_inv_timeout seconds results in fail with $retcode -6, I checked the doc it says "generic send failed". I am not too sure what does that means. If both B and C are in v4 only env, then it works without any issue. A(v4) => opensips(v4/v6) => B(v6) - working A(v4) => opensips(v4/v6) => B(v6) => C(v4) - not working A(v4) => opensips(v4/v6) => B(v4) => C(v4) - working Could you please shed some light on it? -- Regards, William Jin ___ Users mailing list Users@lists.opensips.org http://lists.opensips.org/cgi-bin/mailman/listinfo/users
Re: [OpenSIPS-Users] Question about IP Transit
Hi, Does anyone have any idea? I can send the sip trace if needed. -- Regards, William Jin From: Users on behalf of William Jin Sent: Tuesday, 2 June 2020 2:48 PM To: OpenSIPS users mailling list Subject: Re: [OpenSIPS-Users] Question about IP Transit Just additional info: A(v4) => opensips(v4/v6) => B(v6) => C(v6) - also working Does that mean if the B number is V6, I need to send the invite to the v6 IP of the opensips? Then the server should handle the ip transit from v6 to v4 again? -- Regards, William Jin From: Users on behalf of William Jin Sent: Tuesday, 2 June 2020 2:07 PM To: OpenSIPS users mailling list Subject: [OpenSIPS-Users] Question about IP Transit Hi We are using opensips 2.4.6 at the moment. It's the apt standard(not nightly) repo of debian opensips -V version: opensips 2.4.6 (x86_64/linux) flags: STATS: On, DISABLE_NAGLE, USE_MCAST, SHM_MMAP, PKG_MALLOC, F_MALLOC, FAST_LOCK-ADAPTIVE_WAIT ADAPTIVE_WAIT_LOOPS=1024, MAX_RECV_BUFFER_SIZE 262144, MAX_LISTEN 16, MAX_URI_SIZE 1024, BUF_SIZE 65535 poll method support: poll, epoll, sigio_rt, select. main.c compiled on with gcc 6.3.0 The situation is: Our sip server has both IPv4 and IPv6 addresses, it handles IP transit pretty well. Say if there is an incoming call from the gateway A (asterisk IPv4), it can do the IP transit to the destination number registered in IPv6 A(v4) <=> opensips(v4<=>v6) <=> B(v6) We are using $T_fr_inv_timeout to do call forwarding if no answer. The issue here is: when the B has a call forwarding setting and if the destination number, let's say C is v4, then the t_relay() of the INVITE after $T_fr_inv_timeout seconds results in fail with $retcode -6, I checked the doc it says "generic send failed". I am not too sure what does that means. If both B and C are in v4 only env, then it works without any issue. A(v4) => opensips(v4/v6) => B(v6) - working A(v4) => opensips(v4/v6) => B(v6) => C(v4) - not working A(v4) => opensips(v4/v6) => B(v4) => C(v4) - working Could you please shed some light on it? -- Regards, William Jin ___ Users mailing list Users@lists.opensips.org http://lists.opensips.org/cgi-bin/mailman/listinfo/users
Re: [OpenSIPS-Users] Question about IP Transit
Just additional info: A(v4) => opensips(v4/v6) => B(v6) => C(v6) - also working Does that mean if the B number is V6, I need to send the invite to the v6 IP of the opensips? Then the server should handle the ip transit from v6 to v4 again? -- Regards, William Jin From: Users on behalf of William Jin Sent: Tuesday, 2 June 2020 2:07 PM To: OpenSIPS users mailling list Subject: [OpenSIPS-Users] Question about IP Transit Hi We are using opensips 2.4.6 at the moment. It's the apt standard(not nightly) repo of debian opensips -V version: opensips 2.4.6 (x86_64/linux) flags: STATS: On, DISABLE_NAGLE, USE_MCAST, SHM_MMAP, PKG_MALLOC, F_MALLOC, FAST_LOCK-ADAPTIVE_WAIT ADAPTIVE_WAIT_LOOPS=1024, MAX_RECV_BUFFER_SIZE 262144, MAX_LISTEN 16, MAX_URI_SIZE 1024, BUF_SIZE 65535 poll method support: poll, epoll, sigio_rt, select. main.c compiled on with gcc 6.3.0 The situation is: Our sip server has both IPv4 and IPv6 addresses, it handles IP transit pretty well. Say if there is an incoming call from the gateway A (asterisk IPv4), it can do the IP transit to the destination number registered in IPv6 A(v4) <=> opensips(v4<=>v6) <=> B(v6) We are using $T_fr_inv_timeout to do call forwarding if no answer. The issue here is: when the B has a call forwarding setting and if the destination number, let's say C is v4, then the t_relay() of the INVITE after $T_fr_inv_timeout seconds results in fail with $retcode -6, I checked the doc it says "generic send failed". I am not too sure what does that means. If both B and C are in v4 only env, then it works without any issue. A(v4) => opensips(v4/v6) => B(v6) - working A(v4) => opensips(v4/v6) => B(v6) => C(v4) - not working A(v4) => opensips(v4/v6) => B(v4) => C(v4) - working Could you please shed some light on it? -- Regards, William Jin ___ Users mailing list Users@lists.opensips.org http://lists.opensips.org/cgi-bin/mailman/listinfo/users
[OpenSIPS-Users] Question about IP Transit
Hi We are using opensips 2.4.6 at the moment. It's the apt standard(not nightly) repo of debian opensips -V version: opensips 2.4.6 (x86_64/linux) flags: STATS: On, DISABLE_NAGLE, USE_MCAST, SHM_MMAP, PKG_MALLOC, F_MALLOC, FAST_LOCK-ADAPTIVE_WAIT ADAPTIVE_WAIT_LOOPS=1024, MAX_RECV_BUFFER_SIZE 262144, MAX_LISTEN 16, MAX_URI_SIZE 1024, BUF_SIZE 65535 poll method support: poll, epoll, sigio_rt, select. main.c compiled on with gcc 6.3.0 The situation is: Our sip server has both IPv4 and IPv6 addresses, it handles IP transit pretty well. Say if there is an incoming call from the gateway A (asterisk IPv4), it can do the IP transit to the destination number registered in IPv6 A(v4) <=> opensips(v4<=>v6) <=> B(v6) We are using $T_fr_inv_timeout to do call forwarding if no answer. The issue here is: when the B has a call forwarding setting and if the destination number, let's say C is v4, then the t_relay() of the INVITE after $T_fr_inv_timeout seconds results in fail with $retcode -6, I checked the doc it says "generic send failed". I am not too sure what does that means. If both B and C are in v4 only env, then it works without any issue. A(v4) => opensips(v4/v6) => B(v6) - working A(v4) => opensips(v4/v6) => B(v6) => C(v4) - not working A(v4) => opensips(v4/v6) => B(v4) => C(v4) - working Could you please shed some light on it? -- Regards, William Jin ___ Users mailing list Users@lists.opensips.org http://lists.opensips.org/cgi-bin/mailman/listinfo/users
Re: [OpenSIPS-Users] opensips 3.0.2 100% CPU after enable tls?
ok, thanks. I can confirm the nightly build works without any issue. -- Regards, William Jin From: Bogdan-Andrei Iancu Sent: Tuesday, 28 April 2020 5:19 PM To: William Jin ; OpenSIPS users mailling list Subject: Re: [OpenSIPS-Users] opensips 3.0.2 100% CPU after enable tls? Similarly, try the nightly build Regards, Bogdan-Andrei Iancu OpenSIPS Founder and Developer https://www.opensips-solutions.com On 4/28/20 1:28 AM, William Jin wrote: Ok, I will try that and let you know the result And BTW, does the 2.4.7 (release) also has this issue? -- Regards, William Jin From: Bogdan-Andrei Iancu <mailto:bog...@opensips.org> Sent: Monday, 27 April 2020 7:10 PM To: William Jin <mailto:willi...@exetel.com.au>; OpenSIPS users mailling list <mailto:users@lists.opensips.org> Subject: Re: [OpenSIPS-Users] opensips 3.0.2 100% CPU after enable tls? Hi William, Please use the nightly builds for 3.0 - there is a fix which didn't make it into the release package. Regards, Bogdan-Andrei Iancu OpenSIPS Founder and Developer https://www.opensips-solutions.com On 4/25/20 12:12 AM, William Jin wrote: It's the release. -- Regards, William Jin From: Bogdan-Andrei Iancu <mailto:bog...@opensips.org> Sent: Friday, 24 April 2020 6:14 PM To: William Jin <mailto:willi...@exetel.com.au>; OpenSIPS users mailling list <mailto:users@lists.opensips.org> Subject: Re: [OpenSIPS-Users] opensips 3.0.2 100% CPU after enable tls? And it is the release or nightly build ? Regards, Bogdan-Andrei Iancu OpenSIPS Founder and Developer https://www.opensips-solutions.com On 4/24/20 1:22 AM, William Jin wrote: I am using the debian apt repo, not from git. -- Regards, William Jin From: Bogdan-Andrei Iancu <mailto:bog...@opensips.org> Sent: Thursday, 23 April 2020 6:49 PM To: OpenSIPS users mailling list <mailto:users@lists.opensips.org>; William Jin <mailto:willi...@exetel.com.au> Subject: Re: [OpenSIPS-Users] opensips 3.0.2 100% CPU after enable tls? Hi William, What GIT revision of OpenSIPS do you use? (this is exposed by the "opensips -V") Regards, Bogdan-Andrei Iancu OpenSIPS Founder and Developer https://www.opensips-solutions.com On 4/23/20 7:04 AM, William Jin wrote: Hi, Linux platform: Debian 9 (stretch) opensips -V version: opensips 3.0.2 (x86_64/linux) flags: STATS: On, DISABLE_NAGLE, USE_MCAST, SHM_MMAP, PKG_MALLOC, Q_MALLOC, F_MALLOC, HP_MALLOC, DBG_MALLOC, FAST_LOCK-ADAPTIVE_WAIT ADAPTIVE_WAIT_LOOPS=1024, MAX_RECV_BUFFER_SIZE 262144, MAX_LISTEN 16, MAX_URI_SIZE 1024, BUF_SIZE 65535 poll method support: poll, epoll, sigio_rt, select. main.c compiled on with gcc 6.3.0 related config: listen = tls:xxx.xxx.xxx.xxx:5061 anycast TLS loadmodule "tls_mgm.so" loadmodule "proto_tls.so" modparam("tls_mgm", "server_domain", "sip1") modparam("tls_mgm", "match_ip_address", "[sip1]xx.xx.xx.xx:5061") modparam("tls_mgm", "match_sip_domain", "[sip1]xxx.xxx.example.com") modparam("tls_mgm", "verify_cert", "[sip1]1") modparam("tls_mgm", "require_cert", "[sip1]0") modparam("tls_mgm", "tls_method", "[sip1]SSLv23") modparam("tls_mgm", "ciphers_list", "[sip1]AES256-GCM-SHA384,AES256-SHA256,AES256-SHA,CAMELLIA256-SHA,AES128-SHA,SEED-SHA,CAMELLIA128-SHA,RC4-SHA,DES-CBC3-SHA") modparam("tls_mgm", "certificate", "[sip1]/etc/opensips/tls/mycerts/selfsignedcert.pem") modparam("tls_mgm", "private_key", "[sip1]/etc/opensips/tls/mycerts/unsecuredkey.pem") opensips-cli -x trap {pid} result attached Can someone shed some light on it? Thanks. -- Regards, William Jin ___ Users mailing list Users@lists.opensips.org<mailto:Users@lists.opensips.org> http://lists.opensips.org/cgi-bin/mailman/listinfo/users ___ Users mailing list Users@lists.opensips.org http://lists.opensips.org/cgi-bin/mailman/listinfo/users
Re: [OpenSIPS-Users] opensips 3.0.2 100% CPU after enable tls?
Ok, I will try that and let you know the result And BTW, does the 2.4.7 (release) also has this issue? -- Regards, William Jin From: Bogdan-Andrei Iancu Sent: Monday, 27 April 2020 7:10 PM To: William Jin ; OpenSIPS users mailling list Subject: Re: [OpenSIPS-Users] opensips 3.0.2 100% CPU after enable tls? Hi William, Please use the nightly builds for 3.0 - there is a fix which didn't make it into the release package. Regards, Bogdan-Andrei Iancu OpenSIPS Founder and Developer https://www.opensips-solutions.com On 4/25/20 12:12 AM, William Jin wrote: It's the release. -- Regards, William Jin From: Bogdan-Andrei Iancu <mailto:bog...@opensips.org> Sent: Friday, 24 April 2020 6:14 PM To: William Jin <mailto:willi...@exetel.com.au>; OpenSIPS users mailling list <mailto:users@lists.opensips.org> Subject: Re: [OpenSIPS-Users] opensips 3.0.2 100% CPU after enable tls? And it is the release or nightly build ? Regards, Bogdan-Andrei Iancu OpenSIPS Founder and Developer https://www.opensips-solutions.com On 4/24/20 1:22 AM, William Jin wrote: I am using the debian apt repo, not from git. -- Regards, William Jin From: Bogdan-Andrei Iancu <mailto:bog...@opensips.org> Sent: Thursday, 23 April 2020 6:49 PM To: OpenSIPS users mailling list <mailto:users@lists.opensips.org>; William Jin <mailto:willi...@exetel.com.au> Subject: Re: [OpenSIPS-Users] opensips 3.0.2 100% CPU after enable tls? Hi William, What GIT revision of OpenSIPS do you use? (this is exposed by the "opensips -V") Regards, Bogdan-Andrei Iancu OpenSIPS Founder and Developer https://www.opensips-solutions.com On 4/23/20 7:04 AM, William Jin wrote: Hi, Linux platform: Debian 9 (stretch) opensips -V version: opensips 3.0.2 (x86_64/linux) flags: STATS: On, DISABLE_NAGLE, USE_MCAST, SHM_MMAP, PKG_MALLOC, Q_MALLOC, F_MALLOC, HP_MALLOC, DBG_MALLOC, FAST_LOCK-ADAPTIVE_WAIT ADAPTIVE_WAIT_LOOPS=1024, MAX_RECV_BUFFER_SIZE 262144, MAX_LISTEN 16, MAX_URI_SIZE 1024, BUF_SIZE 65535 poll method support: poll, epoll, sigio_rt, select. main.c compiled on with gcc 6.3.0 related config: listen = tls:xxx.xxx.xxx.xxx:5061 anycast TLS loadmodule "tls_mgm.so" loadmodule "proto_tls.so" modparam("tls_mgm", "server_domain", "sip1") modparam("tls_mgm", "match_ip_address", "[sip1]xx.xx.xx.xx:5061") modparam("tls_mgm", "match_sip_domain", "[sip1]xxx.xxx.example.com") modparam("tls_mgm", "verify_cert", "[sip1]1") modparam("tls_mgm", "require_cert", "[sip1]0") modparam("tls_mgm", "tls_method", "[sip1]SSLv23") modparam("tls_mgm", "ciphers_list", "[sip1]AES256-GCM-SHA384,AES256-SHA256,AES256-SHA,CAMELLIA256-SHA,AES128-SHA,SEED-SHA,CAMELLIA128-SHA,RC4-SHA,DES-CBC3-SHA") modparam("tls_mgm", "certificate", "[sip1]/etc/opensips/tls/mycerts/selfsignedcert.pem") modparam("tls_mgm", "private_key", "[sip1]/etc/opensips/tls/mycerts/unsecuredkey.pem") opensips-cli -x trap {pid} result attached Can someone shed some light on it? Thanks. -- Regards, William Jin ___ Users mailing list Users@lists.opensips.org<mailto:Users@lists.opensips.org> http://lists.opensips.org/cgi-bin/mailman/listinfo/users ___ Users mailing list Users@lists.opensips.org http://lists.opensips.org/cgi-bin/mailman/listinfo/users
Re: [OpenSIPS-Users] opensips 3.0.2 100% CPU after enable tls?
It's the release. -- Regards, William Jin From: Bogdan-Andrei Iancu Sent: Friday, 24 April 2020 6:14 PM To: William Jin ; OpenSIPS users mailling list Subject: Re: [OpenSIPS-Users] opensips 3.0.2 100% CPU after enable tls? And it is the release or nightly build ? Regards, Bogdan-Andrei Iancu OpenSIPS Founder and Developer https://www.opensips-solutions.com On 4/24/20 1:22 AM, William Jin wrote: I am using the debian apt repo, not from git. -- Regards, William Jin From: Bogdan-Andrei Iancu <mailto:bog...@opensips.org> Sent: Thursday, 23 April 2020 6:49 PM To: OpenSIPS users mailling list <mailto:users@lists.opensips.org>; William Jin <mailto:willi...@exetel.com.au> Subject: Re: [OpenSIPS-Users] opensips 3.0.2 100% CPU after enable tls? Hi William, What GIT revision of OpenSIPS do you use? (this is exposed by the "opensips -V") Regards, Bogdan-Andrei Iancu OpenSIPS Founder and Developer https://www.opensips-solutions.com On 4/23/20 7:04 AM, William Jin wrote: Hi, Linux platform: Debian 9 (stretch) opensips -V version: opensips 3.0.2 (x86_64/linux) flags: STATS: On, DISABLE_NAGLE, USE_MCAST, SHM_MMAP, PKG_MALLOC, Q_MALLOC, F_MALLOC, HP_MALLOC, DBG_MALLOC, FAST_LOCK-ADAPTIVE_WAIT ADAPTIVE_WAIT_LOOPS=1024, MAX_RECV_BUFFER_SIZE 262144, MAX_LISTEN 16, MAX_URI_SIZE 1024, BUF_SIZE 65535 poll method support: poll, epoll, sigio_rt, select. main.c compiled on with gcc 6.3.0 related config: listen = tls:xxx.xxx.xxx.xxx:5061 anycast TLS loadmodule "tls_mgm.so" loadmodule "proto_tls.so" modparam("tls_mgm", "server_domain", "sip1") modparam("tls_mgm", "match_ip_address", "[sip1]xx.xx.xx.xx:5061") modparam("tls_mgm", "match_sip_domain", "[sip1]xxx.xxx.example.com") modparam("tls_mgm", "verify_cert", "[sip1]1") modparam("tls_mgm", "require_cert", "[sip1]0") modparam("tls_mgm", "tls_method", "[sip1]SSLv23") modparam("tls_mgm", "ciphers_list", "[sip1]AES256-GCM-SHA384,AES256-SHA256,AES256-SHA,CAMELLIA256-SHA,AES128-SHA,SEED-SHA,CAMELLIA128-SHA,RC4-SHA,DES-CBC3-SHA") modparam("tls_mgm", "certificate", "[sip1]/etc/opensips/tls/mycerts/selfsignedcert.pem") modparam("tls_mgm", "private_key", "[sip1]/etc/opensips/tls/mycerts/unsecuredkey.pem") opensips-cli -x trap {pid} result attached Can someone shed some light on it? Thanks. -- Regards, William Jin ___ Users mailing list Users@lists.opensips.org<mailto:Users@lists.opensips.org> http://lists.opensips.org/cgi-bin/mailman/listinfo/users ___ Users mailing list Users@lists.opensips.org http://lists.opensips.org/cgi-bin/mailman/listinfo/users
Re: [OpenSIPS-Users] opensips 3.0.2 100% CPU after enable tls?
I am using the debian apt repo, not from git. -- Regards, William Jin From: Bogdan-Andrei Iancu Sent: Thursday, 23 April 2020 6:49 PM To: OpenSIPS users mailling list ; William Jin Subject: Re: [OpenSIPS-Users] opensips 3.0.2 100% CPU after enable tls? Hi William, What GIT revision of OpenSIPS do you use? (this is exposed by the "opensips -V") Regards, Bogdan-Andrei Iancu OpenSIPS Founder and Developer https://www.opensips-solutions.com On 4/23/20 7:04 AM, William Jin wrote: Hi, Linux platform: Debian 9 (stretch) opensips -V version: opensips 3.0.2 (x86_64/linux) flags: STATS: On, DISABLE_NAGLE, USE_MCAST, SHM_MMAP, PKG_MALLOC, Q_MALLOC, F_MALLOC, HP_MALLOC, DBG_MALLOC, FAST_LOCK-ADAPTIVE_WAIT ADAPTIVE_WAIT_LOOPS=1024, MAX_RECV_BUFFER_SIZE 262144, MAX_LISTEN 16, MAX_URI_SIZE 1024, BUF_SIZE 65535 poll method support: poll, epoll, sigio_rt, select. main.c compiled on with gcc 6.3.0 related config: listen = tls:xxx.xxx.xxx.xxx:5061 anycast TLS loadmodule "tls_mgm.so" loadmodule "proto_tls.so" modparam("tls_mgm", "server_domain", "sip1") modparam("tls_mgm", "match_ip_address", "[sip1]xx.xx.xx.xx:5061") modparam("tls_mgm", "match_sip_domain", "[sip1]xxx.xxx.example.com") modparam("tls_mgm", "verify_cert", "[sip1]1") modparam("tls_mgm", "require_cert", "[sip1]0") modparam("tls_mgm", "tls_method", "[sip1]SSLv23") modparam("tls_mgm", "ciphers_list", "[sip1]AES256-GCM-SHA384,AES256-SHA256,AES256-SHA,CAMELLIA256-SHA,AES128-SHA,SEED-SHA,CAMELLIA128-SHA,RC4-SHA,DES-CBC3-SHA") modparam("tls_mgm", "certificate", "[sip1]/etc/opensips/tls/mycerts/selfsignedcert.pem") modparam("tls_mgm", "private_key", "[sip1]/etc/opensips/tls/mycerts/unsecuredkey.pem") opensips-cli -x trap {pid} result attached Can someone shed some light on it? Thanks. -- Regards, William Jin ___ Users mailing list Users@lists.opensips.org<mailto:Users@lists.opensips.org> http://lists.opensips.org/cgi-bin/mailman/listinfo/users ___ Users mailing list Users@lists.opensips.org http://lists.opensips.org/cgi-bin/mailman/listinfo/users
[OpenSIPS-Users] opensips 3.0.2 100% CPU after enable tls?
Hi, Linux platform: Debian 9 (stretch) opensips -V version: opensips 3.0.2 (x86_64/linux) flags: STATS: On, DISABLE_NAGLE, USE_MCAST, SHM_MMAP, PKG_MALLOC, Q_MALLOC, F_MALLOC, HP_MALLOC, DBG_MALLOC, FAST_LOCK-ADAPTIVE_WAIT ADAPTIVE_WAIT_LOOPS=1024, MAX_RECV_BUFFER_SIZE 262144, MAX_LISTEN 16, MAX_URI_SIZE 1024, BUF_SIZE 65535 poll method support: poll, epoll, sigio_rt, select. main.c compiled on with gcc 6.3.0 related config: listen = tls:xxx.xxx.xxx.xxx:5061 anycast TLS loadmodule "tls_mgm.so" loadmodule "proto_tls.so" modparam("tls_mgm", "server_domain", "sip1") modparam("tls_mgm", "match_ip_address", "[sip1]xx.xx.xx.xx:5061") modparam("tls_mgm", "match_sip_domain", "[sip1]xxx.xxx.example.com") modparam("tls_mgm", "verify_cert", "[sip1]1") modparam("tls_mgm", "require_cert", "[sip1]0") modparam("tls_mgm", "tls_method", "[sip1]SSLv23") modparam("tls_mgm", "ciphers_list", "[sip1]AES256-GCM-SHA384,AES256-SHA256,AES256-SHA,CAMELLIA256-SHA,AES128-SHA,SEED-SHA,CAMELLIA128-SHA,RC4-SHA,DES-CBC3-SHA") modparam("tls_mgm", "certificate", "[sip1]/etc/opensips/tls/mycerts/selfsignedcert.pem") modparam("tls_mgm", "private_key", "[sip1]/etc/opensips/tls/mycerts/unsecuredkey.pem") opensips-cli -x trap {pid} result attached Can someone shed some light on it? Thanks. -- Regards, William Jin gdb_opensips_20200423_135258 Description: gdb_opensips_20200423_135258 ___ Users mailing list Users@lists.opensips.org http://lists.opensips.org/cgi-bin/mailman/listinfo/users
[OpenSIPS-Users] opensips 2.4.6 crash dns_cache module with CNAME record
Hi, We are using Opensips 2.4.6(apt repo) on Debian stretch opensips -V version: opensips 2.4.6 (x86_64/linux) flags: STATS: On, DISABLE_NAGLE, USE_MCAST, SHM_MMAP, PKG_MALLOC, F_MALLOC, FAST_LOCK-ADAPTIVE_WAIT ADAPTIVE_WAIT_LOOPS=1024, MAX_RECV_BUFFER_SIZE 262144, MAX_LISTEN 16, MAX_URI_SIZE 1024, BUF_SIZE 65535 poll method support: poll, epoll, sigio_rt, select. main.c compiled on with gcc 6.3.0 We started to use dns_cache module recently. However, after we add the module, opensips crashes when it stores a particular dns_cache and we noticed below error Dec 4 03:53:17 /usr/sbin/opensips[10788]: INFO:dns_cache:put_dnscache_value: putting key [dnscache__naptr] with value [|] ttl = 7200 Dec 4 03:53:17 /usr/sbin/opensips[10788]: INFO:dns_cache:put_dnscache_value: putting key [dnscache__sip._udp._srv] with value [|] ttl = 7200 Dec 4 03:53:17 /usr/sbin/opensips[10788]: INFO:dns_cache:put_dnscache_value: putting key [dnscache__] with value [|] ttl = 7200 Dec 4 03:53:17 /usr/sbin/opensips[10788]: INFO:dns_cache:put_dnscache_value: putting key [dnscache__a] with value [#002] ttl = 1769 Dec 4 03:53:17 /usr/sbin/opensips[10788]: INFO:dns_cache:put_dnscache_value: putting key [dnscache__naptr] with value [|] ttl = 7200 Dec 4 03:53:17 /usr/sbin/opensips[10788]: INFO:dns_cache:put_dnscache_value: putting key [dnscache__sip._udp._srv] with value [|] ttl = 7200 Dec 4 03:53:17 /usr/sbin/opensips[10788]: INFO:dns_cache:put_dnscache_value: putting key [dnscache__] with value [|] ttl = 7200 Dec 4 03:53:17 /usr/sbin/opensips[10788]: INFO:dns_cache:put_dnscache_value: putting key [dnscache__a] with value [#002] ttl = 60 Dec 4 03:53:24 /usr/sbin/opensips[10789]: INFO:dns_cache:put_dnscache_value: putting key [dnscache_[CNAME_hostname]_] with value [#012] ttl = 20062 Dec 4 03:53:24 /usr/sbin/opensips[10789]: CRITICAL:core:sig_usr: segfault in process pid: 10789, id: 6 Dec 4 03:53:24 /usr/sbin/opensips[10854]: CRITICAL:core:handle_worker: dead child 6 (EOF received), pid 10789 Dec 4 03:53:24 /usr/sbin/opensips[10773]: INFO:core:handle_sigs: child process 10789 exited by a signal 11 To replicate: == dns_try_ipv6=yes ... loadmodule "cachedb_local.so" loadmodule "dns_cache.so" ... modparam("dns_cache", "cachedb_url","local://") modparam("dns_cache", "blacklist_timeout",7200) Create a hostname with record points to a CNAME without any records (E.g www.example.com<http://www.example.com> 60 IN CNAME www.example.net<http://www.example.net>., but do not give any record to www.example.net<http://www.example.net>) modparam("proto_hep", "hep_id", "[hep_dst] www.example.com;transport=udp;version=3<http://www.example.com;transport=udp;version=3>") #we are using the hostname here in proto_hep, I think it can be triggered anywhere. == >From the log, I believe if the dns query fail, it will store it for >{blacklist_timeout} seconds. However, in my case, it can get a CNAME answer, but there is no actual IPv6 IP associate with it, and it is where opensips crash happens. The ttl = 20062 also looks weird in the log, because I set it up for 60 only. We also tried to use the cachedb_memcached by setting up a memcache server, same crash. However, when we tried to use the same settings on a server without this particular CNAME record, then the issue is gone. -- Regards, William Jin [cid:image001.png@01CF28AB.CA19A270] Exetel System Adminitrator | Exetel PTY LTD Web: www.exetel.com.au<http://www.exetel.com.au/> Main : 0280301000 Direct : 0280301038 Fax : 0280302100 Disclaimer: Hey, we like to 'Get Things Done' but sometimes little things can go astray. Like emails. This email may contain confidential information. If you received it accidentally please let the sender know and delete it. No contractual obligations for pricing or any services will arise until we sign a formal written contract or formal variation to your existing contract. Exetel Pty Ltd - PRIVACY POLICY Exetel respects your privacy, and we will only reveal, discuss, or transact with the owner of the service(s) via e-mail or telephone once we are satisfied we have identified the person who is seeking information. If for any reason any other person(s) wishes to enquire on a service that they are not the owner of, we will not discuss any matters with them. Exetel's Privacy policy is not negotiable under any circumstances, except for urgent or vital situations as described in the relevant Federal legislation (Privacy Act). ___ Users mailing list Users@lists.opensips.org http://lists.opensips.org/cgi-bin/mailman/listinfo/users