aboutsummaryrefslogtreecommitdiff
path: root/net-mgmt/ehnt
diff options
context:
space:
mode:
authorEdwin Groothuis <edwin@FreeBSD.org>2002-11-16 14:36:35 +0000
committerEdwin Groothuis <edwin@FreeBSD.org>2002-11-16 14:36:35 +0000
commit7288714b1cabf5304ebf87e878c44a514d991d0d (patch)
tree004adeb65054e6df994971bd3f9e8caa47e83d46 /net-mgmt/ehnt
parentd3674bb370dcfce2f4723e92fac4f958df0e5a48 (diff)
downloadports-7288714b1cabf5304ebf87e878c44a514d991d0d.tar.gz
ports-7288714b1cabf5304ebf87e878c44a514d991d0d.zip
Maintainer update to version 0.3_2
PR: ports/45296 Submitted by: Dmitry Morozovsky <marck@rinet.ru>
Notes
Notes: svn path=/head/; revision=70249
Diffstat (limited to 'net-mgmt/ehnt')
-rw-r--r--net-mgmt/ehnt/Makefile2
-rw-r--r--net-mgmt/ehnt/files/patch-config-h21
-rw-r--r--net-mgmt/ehnt/files/patch-ehnt__client.c14
-rw-r--r--net-mgmt/ehnt/files/patch-ehnt__processflow.c14
-rw-r--r--net-mgmt/ehnt/files/patch-ehntserv.c176
5 files changed, 219 insertions, 8 deletions
diff --git a/net-mgmt/ehnt/Makefile b/net-mgmt/ehnt/Makefile
index e02f1594367e..e05bc812b5a4 100644
--- a/net-mgmt/ehnt/Makefile
+++ b/net-mgmt/ehnt/Makefile
@@ -7,7 +7,7 @@
PORTNAME= ehnt
PORTVERSION= 0.3
-PORTREVISION= 1
+PORTREVISION= 2
CATEGORIES= net
MASTER_SITES= ${MASTER_SITE_SOURCEFORGE}
MASTER_SITE_SUBDIR= ${PORTNAME}
diff --git a/net-mgmt/ehnt/files/patch-config-h b/net-mgmt/ehnt/files/patch-config-h
index 28d975604eb3..0c7607a95b4f 100644
--- a/net-mgmt/ehnt/files/patch-config-h
+++ b/net-mgmt/ehnt/files/patch-config-h
@@ -1,11 +1,18 @@
---- config.h.orig Wed Jul 18 00:47:37 2001
-+++ config.h Fri Aug 3 19:57:51 2001
-@@ -27,3 +27,8 @@
- #define DEFAULT_UDP_PORT 4444
+
+$FreeBSD$
+
+--- config.h.orig Thu Oct 4 22:18:29 2001
++++ config.h Thu Nov 14 22:20:47 2002
+@@ -32,4 +32,12 @@
/* The default mode of the client. Can be either EM_REPORT or EM_DUMP.*/
- #define EM_DEFAULT EM_REPORT
-+
+ #define EM_DEFAULT EM_TOP
+
+#ifndef ASNCDIR
+#define ASNCDIR "."
+#endif
-+
+
++/* Default size of socket buffer. FreeBSD default is Limited to 256k,
++ + * but misteriously enough maximum space available for buf is 233016 */
++#define DEFAULT_SOCKBUF (192*1024)
++/* Reasonable minimum socket buffer size. I assume it's about 16k */
++#define MIN_SOCKBUF (16*1024)
diff --git a/net-mgmt/ehnt/files/patch-ehnt__client.c b/net-mgmt/ehnt/files/patch-ehnt__client.c
new file mode 100644
index 000000000000..a518e52f5eca
--- /dev/null
+++ b/net-mgmt/ehnt/files/patch-ehnt__client.c
@@ -0,0 +1,14 @@
+
+$FreeBSD$
+
+--- ehnt_client.c.orig Thu Oct 4 22:18:29 2001
++++ ehnt_client.c Thu Nov 14 22:12:07 2002
+@@ -89,7 +89,7 @@
+ if (FD_ISSET(sockfd,&read_fds)) {
+ ehnt_client_processmessage(sockfd,routeraddr,my_ehnt_struct);
+ }
+- if (FD_ISSET(STDIN_FILENO,&read_fds)) {
++ if (e_cfg->mode != EM_COLONDUMP && FD_ISSET(STDIN_FILENO,&read_fds)) {
+ char key;
+ key=getkey();
+ switch (key) {
diff --git a/net-mgmt/ehnt/files/patch-ehnt__processflow.c b/net-mgmt/ehnt/files/patch-ehnt__processflow.c
new file mode 100644
index 000000000000..7441034b9ceb
--- /dev/null
+++ b/net-mgmt/ehnt/files/patch-ehnt__processflow.c
@@ -0,0 +1,14 @@
+
+$FreeBSD$
+
+--- ehnt_processflow.c.orig Thu Oct 4 22:18:29 2001
++++ ehnt_processflow.c Thu Nov 14 22:17:18 2002
+@@ -218,7 +218,7 @@
+ dstip.s_addr=rec->dstaddr;
+ nexthop.s_addr=rec->nexthop;
+
+- printf("%.0f:%s",local_flowcount,routeraddr_string);
++ printf("%.0f:%s:",local_flowcount,routeraddr_string);
+ printf("%d:%s:%d:",
+ rec->input_index,
+ inet_ntoa(srcip),
diff --git a/net-mgmt/ehnt/files/patch-ehntserv.c b/net-mgmt/ehnt/files/patch-ehntserv.c
new file mode 100644
index 000000000000..71703c65150b
--- /dev/null
+++ b/net-mgmt/ehnt/files/patch-ehntserv.c
@@ -0,0 +1,176 @@
+
+$FreeBSD$
+
+--- ehntserv.c.orig Thu Oct 4 22:18:29 2001
++++ ehntserv.c Thu Nov 14 22:18:40 2002
+@@ -33,8 +33,8 @@
+ #define MAXCLIENTS 256
+ #define MAXPACKET 8192
+
+-int SetUDPOpts(int);
+-int SetTCPOpts(int);
++int SetUDPOpts(int, int);
++int SetTCPOpts(int, int);
+
+ int debug;
+ int use_syslog;
+@@ -54,6 +54,8 @@
+ int packetcounter=0;
+ extern char *optarg;
+ int c_udpport, c_tcpport;
++ int one = -1;
++ int bufsize = DEFAULT_SOCKBUF;
+
+ debug=0;
+ use_syslog=0;
+@@ -61,7 +63,7 @@
+ c_udpport=DEFAULT_UDP_PORT;
+ c_tcpport=DEFAULT_TCP_PORT;
+
+- while( (c=getopt(argc,argv,"u:t:d")) != -1) {
++ while( (c=getopt(argc,argv,"u:t:s:d")) != -1) {
+
+ switch(c) {
+ case 'u':
+@@ -70,6 +72,13 @@
+ case 't':
+ sscanf(optarg,"%d",&c_tcpport);
+ break;
++ case 's':
++ sscanf(optarg,"%d",&bufsize);
++ if (bufsize < MIN_SOCKBUF) {
++ perror("too low buffer size or bad format: %s", optarg);
++ exit(1);
++ }
++ break;
+ case 'd':
+ debug=1;
+ break;
+@@ -77,6 +86,7 @@
+ printf("Usage: %s [ -u <udpport> ] [ -t <tcpport> ]\n",argv[0]);
+ printf(" -u <udpport> Listen on UDP port <udpport> for netflow packets. Default %d\n",DEFAULT_UDP_PORT);
+ printf(" -t <tcpport> Listen on TCP port <tcpport> for client connections. Default %d\n",DEFAULT_TCP_PORT);
++ printf(" -s <bufsize> Set tcp/udp buffer size. Default %d\n",DEFAULT_SOCKBUF);
+ printf(" -d Debug. Don't become a daemon, displays all messages on stdout,");
+ printf(" displays lots of extra information\n");
+ exit(0);
+@@ -87,22 +97,22 @@
+
+ /* set up the netflow udp socket */
+ if ( (nf_fd = socket(AF_INET, SOCK_DGRAM, 0)) < 0)
+- { log_error("socket (udp) error"); exit(0); }
++ { log_error("socket (udp) error"); exit(1); }
+ memset(&nf_servaddr,0,sizeof(nf_servaddr));
+ nf_servaddr.sin_family = AF_INET;
+ nf_servaddr.sin_addr.s_addr = htonl(INADDR_ANY);
+ nf_servaddr.sin_port = htons(c_udpport);
+ if ( bind(nf_fd, (struct sockaddr *) &nf_servaddr, sizeof(nf_servaddr)) < 0)
+- { log_error("bind UDP error"); exit(0); }
++ { log_error("bind UDP error"); exit(1); }
+
+- SetUDPOpts(nf_fd);
++ SetUDPOpts(nf_fd, bufsize);
+
+ nf_clilen = (socklen_t) sizeof(nf_cliaddr);
+ l_clilen = (socklen_t) sizeof(l_cliaddr);
+
+ /* set up the listening tcp socket*/
+ if ( (l_fd = socket(AF_INET, SOCK_STREAM, 0)) < 0)
+- { log_error("socket (tcp) error"); exit(0); }
++ { log_error("socket (tcp) error"); exit(1); }
+ memset(&l_servaddr,0,sizeof(l_servaddr));
+ l_servaddr.sin_family = AF_INET;
+ l_servaddr.sin_addr.s_addr = htonl(INADDR_ANY);
+@@ -112,9 +122,9 @@
+ setsockopt(l_fd, SOL_SOCKET, SO_REUSEADDR, (void*)&one, sizeof(one));
+ }
+ if ( bind(l_fd,(struct sockaddr *) &l_servaddr, sizeof(l_servaddr)) < 0)
+- { log_error ("bind TCP error"); exit(0); }
++ { log_error ("bind TCP error"); exit(1); }
+ if ( listen(l_fd, 32) < 0)
+- { log_error("listen error"); exit(0); }
++ { log_error("listen error"); exit(1); }
+ printf("Listening on UDP port %d and TCP port %d (with a backlog of 32)\n",c_udpport,c_tcpport);
+ memset(l_clients,0,sizeof(l_clients));
+
+@@ -157,7 +167,7 @@
+ }
+
+ if ( (c=select(maxfd+1,&read_fds,NULL,NULL,NULL)) < 0)
+- { log_error("select (read) error"); exit(0); }
++ { log_error("select (read) error"); exit(1); }
+
+ if (FD_ISSET(nf_fd,&read_fds)) { /*a netflow message is waiting*/
+
+@@ -168,7 +178,7 @@
+ /* new packet */
+ if ( (nf_rcount=recvfrom(nf_fd,(char *)(nf_packet + sizeof(nf_cliaddr.sin_addr.s_addr)),
+ MAXPACKET,0,(struct sockaddr *)&nf_cliaddr,&nf_clilen)) < 0)
+- { log_error("recvfrom error"); exit(0); }
++ { log_error("recvfrom error"); exit(1); }
+
+ /* Here we use a non-blocking select to find out which tcp clients we
+ can write to. This ensures the write won't block, which would of course
+@@ -189,7 +199,7 @@
+ timeout.tv_sec=0;
+ timeout.tv_usec=0;
+ if ( (c=select(maxfd+1,NULL,&write_fds,NULL,&timeout)) < 0)
+- { log_error("select (write) error"); exit(0); }
++ { log_error("select (write) error"); exit(1); }
+ for (i=0; i<MAXCLIENTS; i++) {
+ if (l_clients[i] > 0) {
+ if (FD_ISSET(l_clients[i],&write_fds)) {
+@@ -220,9 +230,9 @@
+ /* printf("incoming connection on tcp 4444\n");*/
+ l_addrlen=sizeof(l_cliaddr);
+ if ( (l_connfd = accept(l_fd,(struct sockaddr *) l_cliaddr, &l_addrlen)) < 0)
+- { log_error("accept error"); exit(0); }
++ { log_error("accept error"); exit(1); }
+
+- SetTCPOpts(l_connfd);
++ SetTCPOpts(l_connfd, bufsize);
+
+ for (i=0; i<MAXCLIENTS; i++) {
+ if (l_clients[i]==0) {
+@@ -268,25 +278,26 @@
+ }
+
+ /* Increase our receive buffer for nf_fd: */
+-int SetUDPOpts(int fd){
+- int r,p,n;
++int SetUDPOpts(int fd, int rcvsize){
++ int r,p;
+ socklen_t optlen;
+
+- n=512*1024;
+ optlen = sizeof(p);
+
+ if (debug) {
+ r=getsockopt(fd,SOL_SOCKET,SO_RCVBUF,&p,&optlen);
+ printf("Before setsockopt, SO_RCVBUF is %d\n",p);
+ }
+- r=setsockopt(fd,SOL_SOCKET,SO_RCVBUF,&n,sizeof(n));
++ if (setsockopt(fd,SOL_SOCKET,SO_RCVBUF,&rcvsize,sizeof(rcvsize)) < 0) {
++ perror("setsockopt SO_RCVBUF:");
++ }
+ if (debug) {
+ r=getsockopt(fd,SOL_SOCKET,SO_RCVBUF,&p,&optlen);
+ printf("After setsockopt, SO_RCVBUF is %d\n",p);
+ }
+ }
+
+-int SetTCPOpts(int fd) {
++int SetTCPOpts(int fd, int bufsize) {
+ int r,p,n;
+ socklen_t optlen;
+
+@@ -296,8 +307,7 @@
+ r=getsockopt(fd,SOL_SOCKET,SO_SNDBUF,&p,&optlen);
+ printf("Before setsockopt, SO_SNDBUF is %d\n",p);
+ }
+- n=512*1024;
+- r=setsockopt(fd,SOL_SOCKET,SO_SNDBUF,&n,sizeof(n));
++ r=setsockopt(fd,SOL_SOCKET,SO_SNDBUF,&bufsize,sizeof(bufsize));
+ if (r<0) { log_error("setsockopt SO_SNDBUF"); }
+ if (debug) {
+ r=getsockopt(fd,SOL_SOCKET,SO_SNDBUF,&p,&optlen);