August 2018 - dev - lists.frrouting.org

A route-map scaling improvement that has just gone in.
by Donald Sharp 22 Aug '18

22 Aug '18

All - This morning, PR #2818 (https://github.com/FRRouting/frr/pulls/2818) was merged into the FRR master code base. This fixes Issue #2493(https://github.com/FRRouting/frr/issues/2493), where zebra was causing significant processing for route-map changes being made to the system in both BGP and Zebra. Here are some timings: Pre-route-map #2818 improvements for ZEBRA: ZEBRA: Total thread statistics ------------------------- CPU (user+system): Real (wall-clock): Active Runtime(ms) Invoked Avg uSec Max uSecs Avg uSec Max uSecs Type Thread 12 22568.749 121924 185 160855 162 414190 RWTEX TOTAL BGP: Total thread statistics ------------------------- CPU (user+system): Real (wall-clock): Active Runtime(ms) Invoked Avg uSec Max uSecs Avg uSec Max uSecs Type Thread 10 31159.647 121679 256 117558 220 117713 RWTEX TOTAL Create a Unused route-map: robot# conf t robot(config)# route-map BLAR permit 10 robot(config-route-map)# end ZEBRA: Total thread statistics ------------------------- CPU (user+system): Real (wall-clock): Active Runtime(ms) Invoked Avg uSec Max uSecs Avg uSec Max uSecs Type Thread 12 40671.587 136545 297 214461 226 414190 RWTEX TOTAL BGP: Total thread statistics ------------------------- CPU (user+system): Real (wall-clock): Active Runtime(ms) Invoked Avg uSec Max uSecs Avg uSec Max uSecs Type Thread 10 32043.610 122392 261 117558 227 117713 RWTEX TOTAL This is aproximately 19 cpu seconds split between BGP and zebra to handle a full BGP feed for a route-map change that no-process is currently using. After-route-map (#2818) improvements for ZEBRA: ZEBRA: Total thread statistics ------------------------- CPU (user+system): Real (wall-clock): Active Runtime(ms) Invoked Avg uSec Max uSecs Avg uSec Max uSecs Type Thread 11 11592.894 1216075 9 152939 5 120621 RWTEX TOTAL BGP: Total thread statistics ------------------------- CPU (user+system): Real (wall-clock): Active Runtime(ms) Invoked Avg uSec Max uSecs Avg uSec Max uSecs Type Thread 9 8306.451 40216 206 116589 193 116801 RWTEX TOTAL Create a Unused route-map: robot# conf t robot(config)# route-map BLAR permit 10 robot(config-route-map)# end Total thread statistics ------------------------- CPU (user+system): Real (wall-clock): Active Runtime(ms) Invoked Avg uSec Max uSecs Avg uSec Max uSecs Type Thread 11 11596.799 1216183 9 152939 5 120621 RWTEX TOTAL Total thread statistics ------------------------- CPU (user+system): Real (wall-clock): Active Runtime(ms) Invoked Avg uSec Max uSecs Avg uSec Max uSecs Type Thread 9 8320.976 40365 206 116589 193 116801 RWTEX TOTAL This adds up to negligible amounts of cpu time. This is a wonderful improvement in processing of route-map changes for zebra( and consequentially for bgp ). I'd like to thank Soman for doing this work. thanks! donald

1 0

New Defects reported by Coverity Scan for freerangerouting/frr
by scan-admin＠coverity.com 22 Aug '18

22 Aug '18

Hi, Please find the latest report on new defect(s) introduced to freerangerouting/frr found with Coverity Scan. 2 new defect(s) introduced to freerangerouting/frr found with Coverity Scan. New defect(s) Reported-by: Coverity Scan Showing 2 of 2 defect(s) ** CID 1472966: Null pointer dereferences (REVERSE_INULL) /bgpd/bgp_pbr.c: 1887 in bgp_pbr_policyroute_add_to_zebra_unit() ________________________________________________________________________________________________________ *** CID 1472966: Null pointer dereferences (REVERSE_INULL) /bgpd/bgp_pbr.c: 1887 in bgp_pbr_policyroute_add_to_zebra_unit() 1881 } 1882 1883 /* ipset create */ 1884 if (bpm && !bpm->installed) 1885 bgp_send_pbr_ipset_match(bpm, true); 1886 /* ipset add */ >>> CID 1472966: Null pointer dereferences (REVERSE_INULL) >>> Null-checking "bpme" suggests that it may be null, but it has already been dereferenced on all paths leading to the check. 1887 if (bpme && !bpme->installed) 1888 bgp_send_pbr_ipset_entry_match(bpme, true); 1889 1890 /* iptables */ 1891 if (bpm && !bpm->installed_in_iptable) 1892 bgp_send_pbr_iptable(bpa, bpm, true); ** CID 1472965: Null pointer dereferences (REVERSE_INULL) /bgpd/bgp_pbr.c: 1884 in bgp_pbr_policyroute_add_to_zebra_unit() ________________________________________________________________________________________________________ *** CID 1472965: Null pointer dereferences (REVERSE_INULL) /bgpd/bgp_pbr.c: 1884 in bgp_pbr_policyroute_add_to_zebra_unit() 1878 bgp_send_pbr_rule_action(bpa, true); 1879 bgp_zebra_announce_default(bgp, nh, 1880 AFI_IP, bpa->table_id, true); 1881 } 1882 1883 /* ipset create */ >>> CID 1472965: Null pointer dereferences (REVERSE_INULL) >>> Null-checking "bpm" suggests that it may be null, but it has already been dereferenced on all paths leading to the check. 1884 if (bpm && !bpm->installed) 1885 bgp_send_pbr_ipset_match(bpm, true); 1886 /* ipset add */ 1887 if (bpme && !bpme->installed) 1888 bgp_send_pbr_ipset_entry_match(bpme, true); 1889 ________________________________________________________________________________________________________ To view the defects in Coverity Scan visit, https://u2389337.ct.sendgrid.net/wf/click?upn=08onrYu34A-2BWcWUl-2F-2BfV0V0…

1 0

ospfclient and ospf_api_server
by Donald Sharp 21 Aug '18

21 Aug '18

All - Is anyone using this code? It was put into place in the 2003 timeframe and from a quick look at the git logs for the files since then I see 2 actual bug fixes, one in 2005 and one in 2013( to fix a CVE ). I cannot imagine that with the plethora of changes to ospf since then and some of the other concerns outlined to me (namely 1500 byte limit on opaque_lsa ) that this functionality provides anything useful. Is anyone planning on using this or see a need to keep this code? donald

3 2

vrf-lite support in linux
by Himanshu S 17 Aug '18

17 Aug '18

Hi, I am trying out the vrf-lite ( https://www.kernel.org/doc/Documentation/networking/vrf.txt) on ubuntu 16.04.1 (kernel version 4.15). I am able to try it out on virtual box vm. When I try to do the same on Ubuntu 16.04 image on AWS marketplace, adding vrf-lite through ip link command doesn't work. Seems like its not supported. Anyone tried it? TIA Himanshu

2 1

Invitation for slack channel
by Himanshu S 17 Aug '18

17 Aug '18

Could you please get an invite?

2 1

FRR support for FPM
by Mayank Tiwari 15 Aug '18

15 Aug '18

Hi, We are using quagga in our solution for FPM connection and Reverse FPM connection with ONOS. In order to support BFD feature I need to upgrade from quagga to FRR. For that I was patching the changes present in quagga to FRR. During that patching process the following patch is causing compilation issue: diff --git a/zebra/zebra_vty.c b/zebra/zebra_vty.c index 43d76a6..03c4a6b 100644 --- a/zebra/zebra_vty.c +++ b/zebra/zebra_vty.c @@ -20,6 +20,7 @@ #include <zebra.h> +#include "zebra_fpm.h" #include "memory.h" #include "zebra_memory.h" #include "if.h" + +/* function to write the fpm config info */ +static int config_write_fpm (struct vty *vty) +{ + return fpm_remote_srv_write (vty); +} + +/* Zebra node */ +static struct cmd_node zebra_node = +{ + ZEBRA_NODE, + "", + 1 +}; /* Route VTY. */ void zebra_vty_init(void) { /* Install configuration write function. */ install_node(&table_node, config_write_table); install_node(&forwarding_node, config_write_forwarding); + install_node (&zebra_node, config_write_fpm); install_element(VIEW_NODE, &show_ip_forwarding_cmd); install_element(CONFIG_NODE, &ip_forwarding_cmd); diff --git a/zebra/zebra_fpm.h b/zebra/zebra_fpm.h new file mode 100644 index 0000000..9291ffa --- /dev/null +++ b/zebra/zebra_fpm.h @@ -0,0 +1,35 @@ +/* + * Router ID for zebra daemon. + * + * Copyright (C) 2004 James R. Leu + * + * This file is part of Quagga routing suite. + * + * Quagga is free software; you can redistribute it and/or modify it + * under the terms of the GNU General Public License as published by the + * Free Software Foundation; either version 2, or (at your option) any + * later version. + * + * Quagga is distributed in the hope that it will be useful, but + * WITHOUT ANY WARRANTY; without even the implied warranty of + * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the GNU + * General Public License for more details. + * + * You should have received a copy of the GNU General Public License along + * with this program; see the file COPYING; if not, write to the Free Software + * Foundation, Inc., 51 Franklin St, Fifth Floor, Boston, MA 02110-1301 USA + */ + +#ifndef _ZEBRA_VTY_H_ +#define _ZEBRA_VTY_H_ + +#include <zebra.h> +#include "memory.h" +#include "prefix.h" +#include "zclient.h" +#include "if.h" + + +static int fpm_remote_srv_write (struct vty *vty ); + +#endif The function 'fpm_remote_srv_write' is already defined in zebra_fpm.c file which I need to use zebra_vty.c file. I have created a header with name zebra_fpm.h and declared that function there and later included this header in zebra_vty.c. The compilation of zebra_vty.c fails as it is not able to find 'fpm_remote_srv_write' function This is the compilation error: CCLD zebra/zebra_irdp.la CC zebra/zebra_fpm.lo CC zebra/zebra_fpm_netlink.lo ….. ….. CC zebra/zebra_vty.o In file included from zebra/zebra_vty.c:23:0: zebra/zebra_fpm.h:33:12: warning: 'fpm_remote_srv_write' used but never defined static int fpm_remote_srv_write (struct vty *vty ); ^ CC zebra/zebra_vxlan.o CC zebra/zserv.o CC zebra/zebra_netns_id.o CC zebra/zebra_netns_notify.o CC zebra/table_manager.o CC zebra/zapi_msg.o CCLD zebra/zebra zebra/zebra_vty.o: In function `config_write_fpm': /frr/zebra/zebra_vty.c:3697: undefined reference to `fpm_remote_srv_write' collect2: error: ld returned 1 exit status make[2]: *** [zebra/zebra] Error 1 Makefile:3246: recipe for target 'zebra/zebra' failed make[2]: Leaving directory '/frr' Makefile:4421: recipe for target 'all-recursive' failed make[1]: Leaving directory '/frr' make[1]: *** [all-recursive] Error 1 make: *** [all] Error 2 I have observed that zebra_fpm.c is compiled as zebra_fpm.lo instead of zebra_fpm.o. I am not sure if that can cause this issue. Please suggest what I am missing here and what approach I need to take to fix this issue. Thanks and Regards, Mayank

2 4

Slack Channel Invitation
by Алексей 15 Aug '18

15 Aug '18

Hello, May I please request an invitation to the public FRR Slack Channel? -- С уважением, Алексей Черник ДП "ЦИТ Харьков Онлайн" Системный администратор +38 (050) 364-51-05 +38 (057) 707-67-05

2 1

RFC compliance
by Mike Mezeul 15 Aug '18

15 Aug '18

Hello; I am new to Free Range Routing and I am trying to find a compliance statement so that I can compare the FRR stack with the stack that we are currently using but I can't seem to find anything on this topic. Can anyone out there point me to such a list? Thanks, Mike Mezeul Michael J. Mezeul Senior Director R&D ADVA Optical Networking Inc., NA 2301 North Greenville Avenue, Suite 300 Richardson, Texas 75082 T: +1 972.759.1216 F: +1 972.759.1201 M: +1 214.448.9239 mmezeul(a)advaoptical.com<mailto:mmezeul@advaoptical.com> www.advaoptical.com<http://www.advaoptical.com/>

5 12

New Defects reported by Coverity Scan for freerangerouting/frr
by scan-admin＠coverity.com 15 Aug '18

15 Aug '18

Hi, Please find the latest report on new defect(s) introduced to freerangerouting/frr found with Coverity Scan. 15 new defect(s) introduced to freerangerouting/frr found with Coverity Scan. 3 defect(s), reported by Coverity Scan earlier, were marked fixed in the recent build analyzed by Coverity Scan. New defect(s) Reported-by: Coverity Scan Showing 15 of 15 defect(s) ** CID 1472631: Null pointer dereferences (FORWARD_NULL) ________________________________________________________________________________________________________ *** CID 1472631: Null pointer dereferences (FORWARD_NULL) /bfdd/bfdd_vty_clippy.c: 522 in bfd_show_peer() 516 } 517 #if 1 /* anything that can fail? */ 518 if (_failcnt) 519 return CMD_WARNING; 520 #endif 521 #endif >>> CID 1472631: Null pointer dereferences (FORWARD_NULL) >>> Passing null pointer "peer_str" to "bfd_show_peer_magic", which dereferences it. 522 return bfd_show_peer_magic(self, vty, argc, argv, label, peer, peer_str, local, local_str, ifname, vrfname); ** CID 1472630: Uninitialized variables (UNINIT) ________________________________________________________________________________________________________ *** CID 1472630: Uninitialized variables (UNINIT) /bfdd/bfd_packet.c: 990 in bfd_recv_cb() 984 peer.family = AF_INET; 985 strcpy(peer_addr, inet_ntoa(sin.sin_addr)); 986 #endif 987 988 /* Implement RFC 5880 6.8.6 */ 989 if (mlen < BFD_PKT_LEN) { >>> CID 1472630: Uninitialized variables (UNINIT) >>> Using uninitialized element of array "port" when calling "cp_debug". 990 cp_debug(is_mhop, &peer, &local, port, vrfname, 991 "too small (%ld bytes)", mlen); 992 return 0; 993 } 994 995 /* ** CID 1472629: Memory - corruptions (OVERRUN) ________________________________________________________________________________________________________ *** CID 1472629: Memory - corruptions (OVERRUN) /bfdd/bfd_packet.c: 302 in ptm_bfd_echo_pkt_create() 296 297 /* Construct Echo packet information */ 298 ep.data.ver = BFD_ECHO_VERSION; 299 ep.data.len = BFD_ECHO_PKT_LEN; 300 ep.data.my_discr = htonl(bfd->discrs.my_discr); 301 #ifdef BFD_LINUX >>> CID 1472629: Memory - corruptions (OVERRUN) >>> Overrunning struct type udphdr of 8 bytes by passing it to a function which accesses it at byte offset 31 using argument "32". 302 ep.udp.check = 303 #endif /* BFD_LINUX */ 304 #ifdef BFD_BSD 305 ep.udp.uh_sum = 306 #endif /* BFD_BSD */ 307 udp4_checksum(&ep.ip, (uint8_t *)&ep.udp, ** CID 1472628: Security best practices violations (DC.WEAK_CRYPTO) /bfdd/bfd.c: 113 in ptm_bfd_start_xmt_timer() ________________________________________________________________________________________________________ *** CID 1472628: Security best practices violations (DC.WEAK_CRYPTO) /bfdd/bfd.c: 113 in ptm_bfd_start_xmt_timer() 107 * between 108 * 75% and 100% of nominal value, unless detect_mult is 1, then should 109 * be 110 * between 75% and 90%. 111 */ 112 maxpercent = (bfd->detect_mult == 1) ? 16 : 26; >>> CID 1472628: Security best practices violations (DC.WEAK_CRYPTO) >>> "random" should not be used for security related applications, as linear congruential algorithms are too easy to break. 113 jitter = (xmt_TO * (75 + (random() % maxpercent))) / 100; 114 /* XXX remove that division above */ 115 116 if (is_echo) 117 bfd_echo_xmttimer_update(bfd, jitter); 118 else ** CID 1472627: Memory - illegal accesses (USE_AFTER_FREE) /bfdd/control.c: 322 in control_queue_dequeue() ________________________________________________________________________________________________________ *** CID 1472627: Memory - illegal accesses (USE_AFTER_FREE) /bfdd/control.c: 322 in control_queue_dequeue() 316 control_queue_free(bcs, bcq); 317 318 /* Get the next buffer to send. */ 319 if (TAILQ_EMPTY(&bcs->bcs_bcqueue)) 320 goto empty_list; 321 >>> CID 1472627: Memory - illegal accesses (USE_AFTER_FREE) >>> Using freed pointer "bcs->bcs_bcqueue.tqh_first". 322 bcq = TAILQ_FIRST(&bcs->bcs_bcqueue); 323 bcs->bcs_bout = &bcq->bcq_bcb; 324 325 bcs->bcs_outev = NULL; 326 thread_add_write(master, control_write, bcs, bcs->bcs_sd, 327 &bcs->bcs_outev); ** CID 1472626: Memory - corruptions (OVERRUN) /bfdd/ptm_adapter.c: 387 in _ptm_msg_read() ________________________________________________________________________________________________________ *** CID 1472626: Memory - corruptions (OVERRUN) /bfdd/ptm_adapter.c: 387 in _ptm_msg_read() 381 return -1; 382 } 383 384 bpc->bpc_has_localif = ifnamelen > 0; 385 if (bpc->bpc_has_localif) { 386 STREAM_GET(bpc->bpc_localif, msg, ifnamelen); >>> CID 1472626: Memory - corruptions (OVERRUN) >>> Overrunning array "bpc->bpc_localif" of 33 bytes at byte offset 33 using index "ifnamelen" (which evaluates to 33). 387 bpc->bpc_localif[ifnamelen] = 0; 388 } 389 } 390 391 /* Sanity check: peer and local address must match IP types. */ 392 if (bpc->bpc_local.sa_sin.sin_family != 0 ** CID 1472625: Null pointer dereferences (FORWARD_NULL) ________________________________________________________________________________________________________ *** CID 1472625: Null pointer dereferences (FORWARD_NULL) /bfdd/bfdd_vty_clippy.c: 399 in bfd_no_peer() 393 } 394 #if 1 /* anything that can fail? */ 395 if (_failcnt) 396 return CMD_WARNING; 397 #endif 398 #endif >>> CID 1472625: Null pointer dereferences (FORWARD_NULL) >>> Passing null pointer "peer_str" to "bfd_no_peer_magic", which dereferences it. 399 return bfd_no_peer_magic(self, vty, argc, argv, peer, peer_str, local, local_str, ifname, vrfname); 400 } 401 402 /* bfd_show_peers => "show bfd peers [json]" */ 403 DEFUN_CMD_FUNC_DECL(bfd_show_peers) 404 #define funcdecl_bfd_show_peers static int bfd_show_peers_magic(\ ** CID 1472624: Control flow issues (DEADCODE) /bfdd/bfdd_vty.c: 728 in bfd_configure_peer() ________________________________________________________________________________________________________ *** CID 1472624: Control flow issues (DEADCODE) /bfdd/bfdd_vty.c: 728 in bfd_configure_peer() 722 bpc->bpc_local = *local; 723 724 if (peer) { 725 bpc->bpc_peer = *peer; 726 } else { 727 /* Peer configuration is mandatory. */ >>> CID 1472624: Control flow issues (DEADCODE) >>> Execution cannot reach this statement: "snprintf(ebuf, ebuflen, "no...". 728 snprintf(ebuf, ebuflen, "no peer configured"); 729 return -1; 730 } 731 732 bpc->bpc_mhop = mhop; 733 ** CID 1472623: Uninitialized variables (UNINIT) ________________________________________________________________________________________________________ *** CID 1472623: Uninitialized variables (UNINIT) /bfdd/bfd_packet.c: 990 in bfd_recv_cb() 984 peer.family = AF_INET; 985 strcpy(peer_addr, inet_ntoa(sin.sin_addr)); 986 #endif 987 988 /* Implement RFC 5880 6.8.6 */ 989 if (mlen < BFD_PKT_LEN) { >>> CID 1472623: Uninitialized variables (UNINIT) >>> Using uninitialized element of array "vrfname" when calling "cp_debug". 990 cp_debug(is_mhop, &peer, &local, port, vrfname, 991 "too small (%ld bytes)", mlen); 992 return 0; 993 } 994 995 /* ** CID 1472622: (UNINIT) /bfdd/control.c: 517 in control_read() ________________________________________________________________________________________________________ *** CID 1472622: (UNINIT) /bfdd/control.c: 517 in control_read() 511 bcb->bcb_pos += bread; 512 bcb->bcb_left -= bread; 513 /* We need more data, return to wait more. */ 514 if (bcb->bcb_left > 0) 515 goto schedule_next_read; 516 >>> CID 1472622: (UNINIT) >>> Using uninitialized value "bcm.bcm_type". 517 switch (bcm.bcm_type) { 518 case BMT_REQUEST_ADD: 519 control_handle_request_add(bcs, bcb->bcb_bcm); 520 break; 521 case BMT_REQUEST_DEL: 522 control_handle_request_del(bcs, bcb->bcb_bcm); /bfdd/control.c: 537 in control_read() 531 control_handle_notify_del(bcs, bcb->bcb_bcm); 532 break; 533 534 default: 535 log_debug("%s: unhandled message type: %d", __func__, 536 bcm.bcm_type); >>> CID 1472622: (UNINIT) >>> Using uninitialized value "bcm.bcm_id" when calling "control_response". 537 control_response(bcs, bcm.bcm_id, BCM_RESPONSE_ERROR, 538 "invalid message type"); 539 break; 540 } 541 542 bcs->bcs_version = 0; ** CID 1472621: Possible Control flow issues (DEADCODE) /bfdd/bfd_packet.c: 340 in ptm_bfd_echo_snd() ________________________________________________________________________________________________________ *** CID 1472621: Possible Control flow issues (DEADCODE) /bfdd/bfd_packet.c: 340 in ptm_bfd_echo_snd() 334 ep->ip.ip_sum = 0; 335 ep->ip.ip_sum = checksum((uint16_t *)&ep->ip, IP_HDR_LEN); 336 #endif /* BFD_BSD */ 337 } 338 339 if (use_layer2) { >>> CID 1472621: Possible Control flow issues (DEADCODE) >>> Execution cannot reach this statement: "pkt = bfd->echo_pkt;". 340 pkt = bfd->echo_pkt; 341 pktlen = BFD_ECHO_PKT_TOT_LEN; 342 } else { 343 pkt = &bfd->echo_pkt[ETH_HDR_LEN + IP_HDR_LEN + UDP_HDR_LEN]; 344 pktlen = BFD_ECHO_PKT_TOT_LEN 345 - (ETH_HDR_LEN + IP_HDR_LEN + UDP_HDR_LEN); ** CID 1472620: Possible Control flow issues (DEADCODE) /bfdd/bfd_packet.c: 1035 in bfd_recv_cb() ________________________________________________________________________________________________________ *** CID 1472620: Possible Control flow issues (DEADCODE) /bfdd/bfd_packet.c: 1035 in bfd_recv_cb() 1029 cp_debug(is_mhop, &peer, &local, port, vrfname, 1030 "no session found"); 1031 return 0; 1032 } 1033 1034 /* Handle VxLAN cases. */ >>> CID 1472620: Possible Control flow issues (DEADCODE) >>> Execution cannot reach the expression "ptm_bfd_validate_vxlan_pkt(bfd, &vxlan_info)" inside this statement: "if (is_vxlan && !ptm_bfd_va...". 1035 if (is_vxlan && !ptm_bfd_validate_vxlan_pkt(bfd, &vxlan_info)) 1036 return 0; 1037 1038 bfd->stats.rx_ctrl_pkt++; 1039 1040 /* ** CID 1472619: Null pointer dereferences (FORWARD_NULL) ________________________________________________________________________________________________________ *** CID 1472619: Null pointer dereferences (FORWARD_NULL) /bfdd/bfdd_vty_clippy.c: 319 in bfd_peer_label() 313 } 314 #if 0 /* anything that can fail? */ 315 if (_failcnt) 316 return CMD_WARNING; 317 #endif 318 #endif >>> CID 1472619: Null pointer dereferences (FORWARD_NULL) >>> Passing null pointer "label" to "bfd_peer_label_magic", which dereferences it. 319 return bfd_peer_label_magic(self, vty, argc, argv, label); 320 } 321 322 /* bfd_no_peer => "no peer <A.B.C.D|X:X::X:X>$peer [{multihop|local-address <A.B.C.D|X:X::X:X>$local|interface IFNAME$ifname|vrf NAME$vrfname}]" */ 323 DEFUN_CMD_FUNC_DECL(bfd_no_peer) 324 #define funcdecl_bfd_no_peer static int bfd_no_peer_magic(\ ** CID 1399484: Null pointer dereferences (NULL_RETURNS) /tools/permutations.c: 39 in main() ________________________________________________________________________________________________________ *** CID 1399484: Null pointer dereferences (NULL_RETURNS) /tools/permutations.c: 39 in main() 33 { 34 if (argc < 2) { 35 fprintf(stdout, USAGE "\n"); 36 exit(EXIT_SUCCESS); 37 } 38 struct cmd_element *cmd = calloc(1, sizeof(struct cmd_element)); >>> CID 1399484: Null pointer dereferences (NULL_RETURNS) >>> Dereferencing a null pointer "cmd". 39 cmd->string = strdup(argv[1]); 40 41 struct graph *graph = graph_new(); 42 struct cmd_token *token = 43 cmd_token_new(START_TKN, cmd->attr, NULL, NULL); 44 graph_new_node(graph, token, NULL); ** CID 1399196: Error handling issues (CHECKED_RETURN) /tools/start-stop-daemon.c: 1028 in main() ________________________________________________________________________________________________________ *** CID 1399196: Error handling issues (CHECKED_RETURN) /tools/start-stop-daemon.c: 1028 in main() 1022 /* now close all extra fds */ 1023 for (i = getdtablesize() - 1; i >= 0; --i) 1024 close(i); 1025 /* change tty */ 1026 fd = open("/dev/tty", O_RDWR); 1027 if (fd >= 0) { >>> CID 1399196: Error handling issues (CHECKED_RETURN) >>> Calling "ioctl" without checking return value (as is done elsewhere 8 out of 10 times). 1028 ioctl(fd, TIOCNOTTY, 0); 1029 close(fd); 1030 } 1031 chdir("/"); 1032 umask(022); /* set a default for dumb programs */ 1033 setpgid(0, 0); /* set the process group */ ________________________________________________________________________________________________________ To view the defects in Coverity Scan visit, https://u2389337.ct.sendgrid.net/wf/click?upn=08onrYu34A-2BWcWUl-2F-2BfV0V0…

1 0

BFD daemon is now part of master
by Rafael Zalamena 14 Aug '18

14 Aug '18

Hi all, Today the BFD daemon (`bfdd`) got merged into master. It implements the Bidirectional Forwarding Detection protocol and its main purpose is to detect link failures to notify services. The daemon is automatically built on standard distribution on all platforms. Currently the following FRR daemons support BFD: * bgpd; * ospfd; * ospf6d; * pimd; --- Background: The BFD daemon was implemented re-using some Cumulus implementation, but it was extracted from another service ( https://github.com/CumulusNetworks/ptm/blob/master/src/ptm_bfd.c ). The code went through the following changes: * it became a stand alone daemon; * added new configuration interfaces: vtysh and control socket (using JSON); * replaced uthash.h with FRR's hash implementation; * added compatibility code to run on *BSDs; * changed event manager with FRR's; * minor code/protocol fixes; Related RFCs: * RFC 5880 (protocol description); * RFC 5881 (single hop IPv4/IPv6 implementation); * RFC 5883 (multi hop implementation); Thanks to everyone that helped reviewing the code! Rafael

5 4