- Freeradius-Devel - lists.freeradius.org

Failed to insert event?
by Josip Almasi 09 Apr '10

09 Apr '10

Hi all, today my 2.0.2 died with 'Failed to insert event' message in wait_a_bit function (main/event.c). From what I've read, this can easily be fixed by assuring delay longer than timer resolution in fr_event_insert (lib/event.c). And, that was already fixed, during 2.0.3. Right? Thanks! Regards...

2 2

GIT Log for 2010-04-08 23:33 GMT
by aland＠deployingradius.com 08 Apr '10

08 Apr '10

commit 39f2b8581c9c6bb5a4142670b39f780be69589a3 Author: Alan T. DeKok <aland(a)freeradius.org> Date: Thu Apr 8 09:59:55 2010 +0200 Make subdirs before modules Fixes "cannot build from clean directory" problem after switching to the new method of using "make" for recursing into subdirs Files changed: src/modules/Makefile | 6 ++++-- 1 files changed, 4 insertions(+), 2 deletions(-) ====================================================================== commit 397533ff05725eef2c64898c15c34ad712041f9c Author: Alan T. DeKok <aland(a)freeradius.org> Date: Wed Apr 7 01:59:57 2010 +0200 Add missing 'break' to escape '%' properly. Files changed: src/modules/rlm_sqlcounter/rlm_sqlcounter.c | 1 + 1 files changed, 1 insertions(+), 0 deletions(-) ====================================================================== commit c5da5a8f1af076c690463b5fb5010163377a2a8b Author: Alan T. DeKok <aland(a)freeradius.org> Date: Wed Apr 7 01:58:00 2010 +0200 On DHCP Discover fail, don't send a NAK Instead, just don't respond Files changed: src/modules/frs_dhcp/frs_dhcp.c | 12 ++++++++---- 1 files changed, 8 insertions(+), 4 deletions(-) ======================================================================

1 0

Re: Re: Radius server crash due to 35# bug
by Liu Linyong 08 Apr '10

08 Apr '10

Please refer to the following post. ========================= > Author: Alan DeKok > Date: 2010-04-07 19:52 +800 > To: FreeRadius developers mailing list > Subject: Re: Radius server crash due to 35# bug > > Liu Linyong wrote: > > The conclusion: > > The 35# bug can recur rather easily. > > Hmm... I've run the server at 40K packets/s for *days*, and have been > unable to reproduce it. > > > We found FreeRadius server crash sometimes with load. > ... > > Then use gdb to run 'radiusd', and do load test with LoadRunner, which sending thousands of requests per second. > > How many requests in parallel? How many threads is the server using? ====> 4K requests/sec. start_servers = 5 and max_servers = 32. But totally 9 threads were created.(That is: 3 more threads were created after initializing) Oh, I've set max_requests = 102400 in radiusd.conf. > > > FreeRadius will cost little mem(16M) and cpu(1.6%). > > After an hour or so, the server will crash. These can be repeated easily. gdb show the position: > > It looks like a race condition, where data is accessed after it's been > freed. However, tracking it down is hard. > > Alan DeKok. ================================= Liu Linyong 2010-04-08

2 1

Radius server crash due to 35# bug
by Liu Linyong 07 Apr '10

07 Apr '10

Hi, The conclusion: The 35# bug can recur rather easily. We found FreeRadius server crash sometimes with load. To test the stability of the FreeRadius architecture, I modify the two functions: rad_authenticate() / rad_accounting(), they will do very simple things, and will not use any modules. int rad_accounting(REQUEST *request) { request->reply->code = PW_ACCOUNTING_RESPONSE; return RLM_MODULE_OK; } int rad_authenticate(REQUEST *request) { VALUE_PAIR *reply_tmp = NULL; reply_tmp = pairmake( "Service-Type", "Framed-User", T_OP_EQ); if( reply_tmp ) { pairxlatmove(request, &request->reply->vps, &reply_tmp); pairfree(&reply_tmp); } if (request->reply->code == 0) { request->reply->code = PW_AUTHENTICATION_ACK; } return RLM_MODULE_OK; } Then use gdb to run 'radiusd', and do load test with LoadRunner, which sending thousands of requests per second. FreeRadius will cost little mem(16M) and cpu(1.6%). After an hour or so, the server will crash. These can be repeated easily. gdb show the position: Program received signal SIGSEGV, Segmentation fault. 0xff376a6c in fr_packet_cmp (a=0x3d4070, b=0x0) at packet.c:139 139 if (a->sockfd < b->sockfd) return -1; (gdb) (gdb) (gdb) bt #0 0xff376a6c in fr_packet_cmp (a=0x3d4070, b=0x0) at packet.c:139 #1 0xff3680ec in list_find (ht=0x156fc8, head=0x156fe4, reversed=2911412273, data=0xffbff808) at hash.c:191 #2 0xff3685ec in fr_hash_table_finddata (ht=0x2469d0, data=0x156fe4) at hash.c:491 #3 0xff3771d0 in fr_packet_list_find (pl=0x156aa8, request=0x3d4070) at packet.c:557 #4 0x00032b88 in received_request (listener=0x15b848, packet=0x3d4070, prequest=0xffbff938, client=0x1101d8) at event.c:2746 #5 0x0001d5bc in acct_socket_recv (listener=0x15b848, pfun=0xffbff93c, prequest=0xffbff938) at listen.c:908 #6 0x000337d0 in event_socket_handler (xel=0x158450, fd=10, ctx=0x15b848) at event.c:3316 #7 0xff377f84 in fr_event_loop (el=0x158450) at event.c:400 #8 0x000284b4 in main (argc=2, argv=0xffbffb3c) at radiusd.c:398 The problem seems to be the same with 35# bug. (https://bugs.freeradius.org/bugzilla/show_bug.cgi?id=35) Environmnet: MEM: 24G phys mem OS: Solaris 5.10, sparc GCC: 3.3.2 GMAKE: 3.80 GDB: 6.8 CLIENT: LoadRunner (sending packages: auth & acct start & acct stop) FreeRadius server: 2-1-7 Is there any information else I should post here? *********************************************** Liu Linyong 2010-04-07

2 1

should discard packet when received AUTH header length < 20 ?
by WANG Tinggong 01 Apr '10

01 Apr '10

hi all, in src/main/listen.c , auth_socket_recv function, rad_recv_header uses MSG_PEEK flag. in case 0 <= rcode < 20, malfromed request still in the receive queue, should it be cleaned before line 757 ? 751 rcode = rad_recv_header(listener->fd, &src_ipaddr, &src_port, &code ); 752 if (rcode < 0) return 0; 753 754 RAD_STATS_TYPE_INC(listener, total_requests); 755 756 if (rcode < 20) { /* AUTH_HDR_LEN */ 757 RAD_STATS_TYPE_INC(listener, total_malformed_requests); 758 return 0; 759 } here is my patch against branch v2.1.x , does it make sense? thanks Tinggong --- src/main/listen.c | 1 + 1 files changed, 1 insertions(+), 0 deletions(-) diff --git a/src/main/listen.c b/src/main/listen.c index 59ff314..855ee9c 100644 --- a/src/main/listen.c +++ b/src/main/listen.c @@ -754,6 +754,7 @@ static int auth_socket_recv(rad_listen_t *listener, RAD_STATS_TYPE_INC(listener, total_requests); if (rcode < 20) { /* AUTH_HDR_LEN */ + rad_recv_discard(listener->fd); RAD_STATS_TYPE_INC(listener, total_malformed_requests); return 0; }

2 1

GIT Log for 2010-03-31 23:33 GMT
by aland＠deployingradius.com 31 Mar '10

31 Mar '10

commit 3a8838362791f2e88bc1dd5b30b83bb7cc721079 Author: Alan T. DeKok <aland(a)freeradius.org> Date: Wed Mar 31 16:19:28 2010 +0200 Catch corner case of update disconnect in post-auth If we proxy packets, we don't want to send disconnect, too Files changed: src/main/event.c | 22 +++++++++++++++++++--- 1 files changed, 19 insertions(+), 3 deletions(-) ====================================================================== commit a273da9e8673ba8a8527551435b89b0c53c4f2a8 Author: Alan T. DeKok <aland(a)freeradius.org> Date: Wed Mar 31 15:33:19 2010 +0200 Fixed typo Files changed: src/modules/rlm_eap/types/Makefile | 2 +- 1 files changed, 1 insertions(+), 1 deletions(-) ====================================================================== commit 5db4baf9eaf747de083835839ff6e53af830bc34 Author: Alan T. DeKok <aland(a)freeradius.org> Date: Wed Mar 31 15:06:52 2010 +0200 Correct realm regex examples, and creation of REALM for them The old examples were wrong. The old code created the Realm attribute with value of the regex... rather than the actual realm name. Closes bug #40 Files changed: raddb/proxy.conf | 18 ++++++++---------- src/modules/rlm_realm/rlm_realm.c | 10 ++++++++-- 2 files changed, 16 insertions(+), 12 deletions(-) ====================================================================== commit 151377eadebf9d7cae8055a87d96a50323d4c1df Author: Alan T. DeKok <aland(a)freeradius.org> Date: Wed Mar 31 11:32:38 2010 +0200 Allow the "server" field to be blank People run into "" != NULL Files changed: src/modules/rlm_sql/rlm_sql.c | 2 +- 1 files changed, 1 insertions(+), 1 deletions(-) ====================================================================== commit 34dfa5058917816110a510927dbceca1619358fa Author: Alan T. DeKok <aland(a)freeradius.org> Date: Wed Mar 31 11:25:08 2010 +0200 Made 'reconfig' target use PHONY subdirectories, too And added reconfig targets all over the place. This means that we can do "make reconfig" from (almost) anywhere, and have it work. Files changed: src/include/Makefile | 1 + 1 files changed, 1 insertions(+), 0 deletions(-) ====================================================================== commit ddc27ce6718618c7ffc6f2dab053cef6b72432c7 Author: Alan T. DeKok <aland(a)freeradius.org> Date: Wed Mar 31 11:25:08 2010 +0200 Made 'reconfig' target use PHONY subdirectories, too And added reconfig targets all over the place. This means that we can do "make reconfig" from (almost) anywhere, and have it work. Files changed: Makefile | 5 ++- src/Makefile | 6 +++- src/include/Makefile | 2 + src/lib/Makefile | 3 ++ src/main/Makefile.in | 3 ++ src/modules/Makefile | 51 +++++++++++++++++++------------ src/modules/rlm_eap/Makefile.in | 4 +- src/modules/rlm_eap/types/Makefile | 4 +- src/modules/rlm_sql/Makefile.in | 4 +- src/modules/rlm_sql/drivers/Makefile.in | 4 +- 10 files changed, 55 insertions(+), 31 deletions(-) ====================================================================== commit 9958c3f6e053f64eda6902390c654ff5ee457314 Author: Alan T. DeKok <aland(a)freeradius.org> Date: Wed Mar 31 10:55:39 2010 +0200 Move to using phony targets for subdirectories Files changed: src/modules/rlm_sql/Makefile.in | 12 +++++++----- src/modules/rlm_sql/drivers/Makefile.in | 15 +++++++++------ 2 files changed, 16 insertions(+), 11 deletions(-) ======================================================================

1 0

GIT Log for 2010-03-30 23:33 GMT
by aland＠deployingradius.com 30 Mar '10

30 Mar '10

commit 2b596fc715c0948e3cbe6642f58d9f3678f0b945 Author: Alan T. DeKok <aland(a)freeradius.org> Date: Tue Mar 30 16:13:03 2010 +0200 Minor change to error output for clarity Files changed: src/main/modules.c | 2 +- 1 files changed, 1 insertions(+), 1 deletions(-) ====================================================================== commit 3077b3e3a91a096fd13663780151ea349fbf2bcf Author: Alan T. DeKok <aland(a)freeradius.org> Date: Tue Mar 30 16:12:06 2010 +0200 Re-open log file after HUP. This should have been committed in 6563dac620f, but was missed for some reason. Files changed: src/main/log.c | 143 +++++++++---------------------------------------------- 1 files changed, 24 insertions(+), 119 deletions(-) ====================================================================== commit 6c6e719cc247ba90a111768ef01f6b8eafa1b343 Author: Alan T. DeKok <aland(a)freeradius.org> Date: Tue Mar 30 16:06:15 2010 +0200 Revert "= NULL" checks. They should be "IS NULL" Files changed: raddb/sql/mysql/ippool.conf | 12 ++++++------ 1 files changed, 6 insertions(+), 6 deletions(-) ====================================================================== commit fc12c44773806dcb0640ca864f13cefe2295d391 Author: Alan T. DeKok <aland(a)freeradius.org> Date: Tue Mar 30 16:04:23 2010 +0200 Move to using phony targets for subdirectories. This removes the explicit shell loop over subdirs, which means that the main "make" program now tracks all dependencies. I've also updated some dependencies to be correct. Now "make -j 4" will work. Files changed: Makefile | 12 +++++++----- src/Makefile | 12 +++++++----- src/modules/Makefile | 17 ++++++++++------- src/modules/rlm_eap/Makefile.in | 19 +++++++++++-------- src/modules/rlm_eap/types/Makefile | 23 ++++++++++++++++------- 5 files changed, 51 insertions(+), 32 deletions(-) ======================================================================

1 0

Is that a bug of event?
by zhongwei feng 30 Mar '10

30 Mar '10

hi, all: Oracle database , 20000 users, and Simultaneous-Use := 1 ; but simul_verify_query is not enough fast ... (I have tow BAS, load-balance, some users fast switch between both...) [radius@RADSERVER1 radius]$ gdb radiusd -f core.27318 . . . Reading symbols from /usr/local/lib/rlm_detail-2.1.8.so...done. Loaded symbols for /usr/local/lib/rlm_detail-2.1.8.so #0 request_pre_handler (request=0x2a98f39310) at event.c:1763 /home/radius/fzw/freeradius-server-2.1.8/src/main/event.c:1763:44023:beg:0x420e66 (gdb)

2 1

freeradius-client "radius_timeout" and freeradius-client vs freeradius-server
by Stefan Karss 24 Mar '10

24 Mar '10

Hello list, during development of a radius test client using freeradius-client library I stumbled across line 317 in sendserver.c from CVS HEAD: result = poll(&pfd, 1, timeout * 1000000); This line came with revision 1.26 where select() was rewritten with poll(). The timeout value for poll however is in milliseconds not in microseconds so that timeout will be 1000 times higher than configured in the radiusclient.conf. As the bug seems to be in the code since November 2008 I wonder if freeradius-client is the right way to go for a radius client or if one should use the freeradius-server libraries to build a client? Or is the right way to go the 1.1.6 release downloadable on the homepage - unfortunately this release does not have the patches from Alex Massover regarding seqfile and thread safety. Rgds, Stefan

3 4

authserver round robin and multiple NAS-IP-Address avpairs
by Stefan Karss 24 Mar '10

24 Mar '10

Hello, when configuring round-robin authservers in sendserver.c function "rc_send_server" the NAS-IP-Address gets added for each fail-overed server resulting in the NAS-IP-Address avpair being sent multiple times. I fixed it by checking for presence of NAS-IP-Address avpair (see patch). Rgds, Stefan

2 1