I have multiple slaves on a RS485 bus. I have been using pymodbus so far but I'm not quite happy with it's performance and other issues. So I wanted to to test libmodus and use that instead.
I wrote a minimal program that reads the model number of my slaves
#include <stdio.h>
#include <stdlib.h>
#include <string.h>
#include <stdint.h>
#include <modbus.h>
#include <errno.h>
char *bigendian_vec_to_str(uint16_t *vec, size_t vec_size, char *buff, size_t buff_len)
{
memset(buff, 0, sizeof *buff * buff_len);
int i;
for(i = 0; i < vec_size; ++i)
{
uint16_t fl = vec[i] >> 8;
uint16_t sl = vec[i] & 0xff;
if(2*i >= buff_len - 1)
return buff;
if(fl == 0)
return buff;
buff[2 * i] = fl;
if(2*i + 1 >= buff_len - 1)
return buff;
if(sl == 0)
return buff;
buff[2 * i + 1] = sl;
}
return buff;
}
char *get_model_name_of(modbus_t *modbus, int slave, char *buff, size_t buff_len)
{
modbus_flush(modbus);
modbus_set_slave(modbus, slave);
int rc;
uint16_t reg[9];
memset(reg, 0, sizeof reg);
rc = modbus_read_registers(modbus, 0xe, 8, reg);
if (rc == -1) {
fprintf(stderr, "Error %d while reading: %s\n", errno, modbus_strerror(errno));
return NULL;
}
return bigendian_vec_to_str(reg, 8, buff, buff_len);
}
int main(void)
{
modbus_t *modbus = modbus_new_rtu("/dev/ttyUSB0", 9600, 'N', 8, 1);
modbus_rtu_set_serial_mode(modbus, MODBUS_RTU_RS485);
if (modbus_connect(modbus) == -1) {
fprintf(stderr, "Connexion failed: %s\n", modbus_strerror(errno));
modbus_free(modbus);
return -1;
}
char buff[1024];
int i;
for(i = 2; i < 5; ++i)
{
printf("Model of slave %d: %s\n", i, get_model_name_of(modbus, i, buff, sizeof buff));
}
modbus_free(modbus);
return 0;
}
When I ran this code I got
Model of slave 2: LEFS25B-600
Error 110 while reading: Connection timed out
Model of slave 3: (null)
Model of slave 4: LEHF10K2-16
and it seemed strange that the 2nd module was not responding. So I looped get_model_name_of through 2,3,4,2,3,4,2,3,4.... and every second read attempt ended with Error 110 while reading: Connection timed out. After the iine modbus_set_slave(modbus, slave); I added
usleep(0.005 * 1000000);
and then I didn't get timeouts anymore. I read the man pages twice and I didn't find anything warning me about this. I also searched google but none of the "similar" threads I found were of any help.
What is the best way to deal with multiple slaves? Why does adding a sleep of half of milisecond help here? I mean the code on libmodus does
static int _modbus_set_slave(modbus_t *ctx, int slave)
{
/* Broadcast address is 0 (MODBUS_BROADCAST_ADDRESS) */
if (slave >= 0 && slave <= 247) {
ctx->slave = slave;
} else {
errno = EINVAL;
return -1;
}
return 0;
}
is setting an internal value in the context. Are there any time constraints between the change of an internal value in the context and reading/writing to the bus? If so, how long should I wait after a set_slave? Why does libmodbus set the slave id globally instead of having it as a parameter in the read/write method as other libraries (like pymodbus) do?
Or am I using this API just incorrectly?
Thanks
I may be wrong.. but.. as I understand it. The modbus master sends out a request, targeted at a specfic slave number. The intention is to recieve a reply from the targeted slave and then send a request to the next slave and await a reply from second slave. If the requests are sent out without waiting for reply from the first slave.. then there is a possibility to miss the reply from the second slave(or third or whatever number slave) , while the first slave reply is being sent and recieved by the master.
I am not good in C programming.. but I recommend you check this..as I think that may be why you adding a delay seems to help... ( Also.. part of Modbus protocol does require a pause in signal transmission to define start and end of transmission.)
If I am correct , then the use of a delay will only work well if you know the size of data being sent and the time to calculate a response..For other situations a handshake of some kind would be safe.. Such as read a coil.. that indicates whether data is refreshed and ready to be read from the slave as a possible traffic light . to control timing of requests going to other slave and to avoid collision of responses.
Again.. I am not good in C and if I have misinterpreted the program.. please ignore what I have said.. If it helps.. I would be happy hear.
Peter
Related
I'm currently trying to send raw binary data in the format of decimal to an external device over serial. I currently have the data in a buffer array but would like it in a structure like this:
struct packetData{
uint8_t sync1;
uint8_t sync2;
uint16_t messageId;
uint16_t dataWordCount;
uint16_t flags;
uint16_t checksum;
};
I'm also using 9600 baud, and have all the termios settings set using cfmakeraw and I'm currently writing using:
#include <stdio.h>
#include <stdint.h>
#include <unistd.h>
#include <fcntl.h>
#include <termios.h>
#include <string.h>
#include <errno.h>
#include <stdlib.h>
int flags = O_RDWR | O_NOCTTY | O_NDELAY;
fd = open(device, flags);
uint16_t buf_tx[BUFFER_SIZE] = {255,129,191,0,2057,0};
if(fd == -1){
printf("\n Failed to open port! ");
return -1;
}
tcgetattr(fd, &tty); //Get the current attributes of the Serial port
cfmakeraw(&tty);
cfsetispeed(&tty, B9600); //Set read speed as 9600 baud
cfsetospeed(&tty, B9600); //Set write speed as 9600 baud
if((tcsetattr(fd, TCSANOW, &tty)) != 0){
printf("Error! Can't set attributes.\n");
return -1;
}
else{
printf("Connection successful! \n");
}
while(x < 1000){
memset(buf_tx, 0, sizeof(buf_tx));
tcflush(fd, TCOFLUSH);
if(y < 5){
if(write(fd, buf_tx, 5) == -1){
printf("\n");
printf("Error>>: %s\n",strerror(errno));
y++;
}
}
tcflush(fd, TCIOFLUSH);
usleep(1000);
x++;
}
This code isnt the full code, just the setup/write parts so no need to worry about its syntax. if possible it would be nice not to have that buffer array and just use the struct directly, but I'll take what I can get.
It seems you have the serial port opening more or less in hand. I prefer to set the termios member components explicitly myself, but cfmakeraw() is perfectly fine too.
What you should consider, is having a separate function to send one or more of those structures at a time. For example,
int write_all(const int fd, const void *buf, const size_t len)
{
const char *data = buf;
size_t written = 0;
ssize_t n;
while (written < len) {
n = write(fd, data + written, len - written);
if (n > 0) {
written += n;
} else
if (n != -1) {
/* C library bug, should never occur */
errno = EIO;
return -1;
} else {
/* Error; n == -1, so errno is already set. */
return -1;
}
}
/* Success. */
return 0;
}
The function will return 0 if all data was successfully written, and -1 with errno set if an error occurs.
To send a struct packetData pkt; just use write_all(fd, &pkt, sizeof pkt).
To send a full array struct packetData pkts[5]; use write_all(fd, pkts, sizeof pkts).
To send n packets starting at pkts[i], use write_all(fd, pkts + i, n * sizeof pkts[0]).
However, you do not want to use tcflush(). It does not do what you think it does; it actually just discards data.
Instead, to ensure that the data you have written has been transmitted, you need to use tcdrain(fd).
I recommend against adding tcdrain(fd) at the end of write_all() function, because it blocks, pauses the program, until the data has been transmitted. This means that you should only use tcdrain() before you do something that requires the other end has received the transmission; for example before trying to read the response.
However, if this is a query-response interface, and you do intend to also read from the serial device, you should set tty.c_cc[VMIN] and tty.c_cc[VTIME] to reflect how you intend to use the interface. I prefer asynchronous full-duplex operation, but that requires select()/poll() handling. For half-duplex, with these exact structures only, you can use tty.c_cc[VMIN] = sizeof (struct packetData) with say tty.c_cc[VTIME] = 30, which causes read() to try and wait until a full structure is available, but at most 30 deciseconds (3.0 seconds). Something like tty.c_cc[VMIN] = 1; tty.c_cc[VTIME] = 1; is more common; that causes read() to return a short count (even 0!) if there is no additional data received within a decisecond (0.1 seconds). Then, the receive function could be along the following lines:
int read_all(const int fd, void *buf, const size_t len)
{
char *const ptr = buf;
size_t have = 0;
ssize_t n;
/* This function is to be used with half-duplex query-response protocol,
so make sure we have transmitted everything before trying to
receive a response. Also assumes c_cc[VTIME] is properly set for
both the first byte of the response, and interbyte response interval
in deciseconds. */
tcdrain(fd);
while (have < len) {
n = read(fd, ptr + have, len - have);
if (n > 0) {
have += n;
} else
if (n == 0) {
/* Timeout or disconnect */
errno = ETIMEDOUT;
return -1;
} else
if (n != -1) {
/* C library bug, should never occur */
errno = EIO;
return -1;
} else {
/* Read error; errno set by read(). */
return -1;
}
}
/* Success; no errors. */
return 0;
}
If this returns -1 with errno == ETIMEDOUT, the other side took too long to answer. There may be remainder of the late response in the buffer, which you can discard with tcflush(TCIFLUSH) (or with tcflush(TCIOFLUSH), which also discards any written data not yet transmitted). Synchronization in this case is a bit difficult, because the above read_all() function doesn't return how many bytes it received (and therefore how many bytes to discard of a partial structure).
Sometimes the interface used always returns the number of bytes, but also sets errno (to 0 if no error occurred, and a nonzero error constant otherwise). That would be better for a query-response interface read and write functions, but many programmers find this use case "odd", even though it is perfectly okay by POSIX.1 standard (which is the relevant standard here).
I have weird problem. I try to communicate with ifm AY1020 via modbusTCP using libmodbus from PC.
My code looks as follow:
#include <stdio.h>
#include <unistd.h>
#include <string.h>
#include <stdlib.h>
#include <errno.h>
#include <modbus/modbus.h>
int main()
{
modbus_t *ctx;
uint16_t *tab_reg;
int rc;
int i;
ctx = modbus_new_tcp("192.168.1.250", 502);
modbus_set_debug(ctx, TRUE);
tab_reg = (uint16_t *) malloc(5 * sizeof(uint16_t));
memset(tab_reg, 0, 5 * sizeof(uint16_t));
if (modbus_connect(ctx) == -1)
{
fprintf(stderr, "Connection failed: %s\n",modbus_strerror(errno));
modbus_free(ctx);
return -1;
}
rc = modbus_read_registers(ctx, 3002, 2, tab_reg);
if (rc == -1)
{
fprintf(stderr, "%s\n", modbus_strerror(errno));
return -1;
}
for (i=0; i < rc; i++) {
printf("reg[%d]=%d (0x%X)\n", i, tab_reg[i], tab_reg[i]);
}
modbus_close(ctx);
modbus_free(ctx);
}
Thanks to debug I was able to get the frame that is generated in modbus_read_registers function:
[00][01][00][00][00][06][FF][03][0B][BA][00][02]
And I get this
ERROR Gateway path unavailable
Gateway path unavailable
After analysis you can find that device id in that frame is FF, but according to this error PLC expects 1.
Going further if during debugging I force change this value from FF to 01 everything works fine. It looks like it assign wrong ID.
I would be grateful for any help, advice, solution.
Best,
Paweł
Looking at the Man
You should call modbus_set_slave to set a specific destination device.
TCP
The slave number is only required in TCP if the message must reach a device on a serial network. The special value MODBUS_TCP_SLAVE (0xFF) can be used in TCP mode to restore the default value.
Emphasis mine
Your code should be
modbus_set_slave(ctx, 1);
rc = modbus_read_registers(ctx, 3002, 2, tab_reg);
I am a new C programmer and so you will have to excuse my lack of knowledge. I am trying to use sockets in C on a windows machine to send data back and forth between a client and server. I am using the tools of cygwin with the codeblocks IDE. Simple send and receives were not working and so after some searching I was under the impression my problem was I needed a send_all and recv_all function. I have written the following two functions but receive seems to always get stuck in an infinite loop. I am not really sure why.
void send_all(int socket, void *buffer, int length) {
size_t i = 0;
for (i = 0; i < length; i += send(socket, buffer, length - i, 0)){
printf("Completed: %d bytes \r", i);
}
printf("Send Completed: %d bytes \n", length);
}
void recv_all(int sockfd, void *buffer, int length){
size_t i = 0;
for (i = 0; i < length; i+= recv(sockfd, buffer + i, length - i, 0)){
printf("Completed: %d bytes \r", i);
}
printf("Receive Completed: %d bytes \n", length);
}
I am wondering if it is because the receive doesn't know how many bytes the send is sending it. All advice is appreciated but please keep it constructive. Thanks.
recv() actually returns a signed value (int in Winsock, ssize_t in POSIX). Its return value can be a negative number if a read error occurred, OR if the socket is in non-blocking mode and no data is available. Its return value is zero if the socket was closed gracefully (this would cause an infinite loop in your code).
You will need to check the return value before you add it to your byte counter, to detect both of these conditions.
If your socket is in blocking mode (the default), your code will block indefinitely until the required amount of data has been received, or an error occurs (once you add code to check for that). Given the name of your function this seems to be the behavior you want. If so, your general approach is sound.
ssize_t bytesRead = 0;
while (bytesRead < length)
{
ssize_t rv = recv(/* ... */);
if (rv == 0)
{
printf("Socket closed gracefully before enough data received\n");
break;
}
else if (rv < 0)
{
// if your socket is non-blocking, check for EAGAIN which
// would mean no data is currently available; in this case you
// could do something like call select() on the socket to
// go to sleep until more data comes in
printf("Read error occurred before enough data received\n");
break;
}
bytesRead += rv;
}
I currently have a client app that works but it is single threaded.
my packets look like this: < len_of_data>|< data>"
"|" is used as a separator for my data.
< len_of_data> is always 4 digits long followed.
< data> looks like: |< transaction id>|< command>|< buflen>|< buf>|< checksum>|
my code to create the packets is:
_snprintf_s(data_buffer, WS_MAX_DATA_PACKET_SIZE,
WS_MAX_DATA_PACKET_SIZE - 1,
"%s%d%s%d%s%d%s%s%s%d%s",
WS_PACKET_SEP, pkt->transaction_id,
WS_PACKET_SEP, pkt->command,
WS_PACKET_SEP, pkt->bufsize,
WS_PACKET_SEP, pkt->buf,
WS_PACKET_SEP, pkt->checksum, WS_PACKET_SEP);
buf_len = strlen(data_buffer);
_snprintf_s(send_buffer, WS_MAX_DATA_PACKET_SIZE,
WS_MAX_DATA_PACKET_SIZE - 1, "%04d%s%s",
buf_len, WS_PACKET_SEP, data_buffer);
buf_len = strlen(send_buffer);
// Send buffer
bytes_sent = send(ConnectSocket, send_buffer, buf_len, 0);
The client thread sends a command to the server, then calls a GetIncomingPackets() function. In GetIncomingPackets(), I call recv() to get 5 bytes, this should be the len of the rest of packet, I parse these 5 bytes and verify that they match my expected format. Then I convert the first 4 bytes to an integer, x. Then I call recv() again to get x bytes more and then parse those out into my packet structure.
The problem happens when I add another thread to do the same thing (send and receive commands).
I start my app and fire 2 threads and send them to send different commands and wait for responses. When the threads call GetIncomingPackets(), the data I am getting back is invalid. The first 5 bytes I am expecting are missing sometimes, and I just get the following 5 bytes, therefore I am unable to get my < len_of_data > packet.
I even added a critical section block between the 2 recv() calls in my GetIncomingPackets() so the treads dont interrupt each other while getting a full packet.
Without some extra code for error checking, this how the function looks like
#define WS_SIZE_OF_LEN_PACKET 5
bool GetIncomingPackets(SOCKET sd, dev_sim_packet_t *pkt )
{
char len_str_buf[WS_SIZE_OF_LEN_PACKET + 1] = {0}; // + 1 for NULL char
char data_buf[WS_MAX_DATA_PACKET_SIZE + 1] = {0};
int ret = 0;
int data_len = 0;
EnterCriticalSection( &recv_critical_section );
nReadBytes = WS_RecvAll(sd, len_str_buf, WS_SIZE_OF_LEN_PACKET );
ret = WS_VerifyLenPacket(len_str_buf);
// Convert data packet lenght string received to int
data_len = WS_ConvertNumberFromString(len_str_buf, WS_SIZE_OF_LEN_PACKET );
// Get data from packet
nReadBytes = WS_RecvAll(sd, data_buf, data_len);
LeaveCriticalSection( &recv_critical_section );
ret = ParseMessager(data_buf, data_len, pkt);
}
My question is, what could be causing this problem, and how could I fix it? Or is there better ways to do what i am trying to do. The reason that I'm trying to make it multi-threaded is because my app will communicate with 2 other sources, and I want to have a thread to handle each request that comes in from either source.
thanks in advance and feel free to ask any questions if I didn't explain something well.
Here's the code for WS_RecvAll(). The buffer is a static buffer declared in GetIncomingPackets() like this:
char data_buf[WS_MAX_DATA_PACKET_SIZE + 1] = {0}; // + 1 for NULL char
int WS_RecvAll(SOCKET socket_handle, char* buffer, int size)
{
int ret = 0;
int read = 0;
int i = 0;
char err_buf[100] = {0};
while(size)
{
ret = recv(socket_handle, &buffer[read], size, 0);
if (ret == SOCKET_ERROR)
{
printf("***ERROR***: recv failed, error = %d\n", WSAGetLastError());
return WS_ERROR_RECV_FAILED;
}
if (ret == 0) {
break;
}
read += ret;
size -= ret;
}
return read;
}
It's very difficult to debug MT problems, particularly at one remove, but if you are using astatic buffer, should not:
LeaveCriticalSection( &recv_critical_section );
ret = ParseMessager(data_buf, data_len, pkt);
be:
ret = ParseMessager(data_buf, data_len, pkt);
LeaveCriticalSection( &recv_critical_section );
And why use a static buffer in any case?
Im curious to know whether you have used the same socked descriptor in both the threads to connect to the server.
I'm looking for a way to get hold of network stats in C on Linux and MacOSX. Specifically, I need to monitor the number of bytes uploaded and downloaded from each network adapter on the system - I don't need to do packet inspection, or differentiate between protocols, just a 'total bytes' counter which I can poll at intervals would be fine. In Windows I can do this using the iphlpapi.dll library via GetIfTable (to list the network adapters) and GetIfEntry (to read the stats), but I can't find the Linux/OSX equivalents. My knowledge of C is fairly basic so I would appreciate a solution that isn't too involved. Any help would be much appreciated!
The Darwin netstat source code uses sysctl.
Here's some code that prints the number of bytes in and out on OSX:
#import <Foundation/Foundation.h>
#include <sys/sysctl.h>
#include <netinet/in.h>
#include <net/if.h>
#include <net/route.h>
int main (int argc, const char * argv[]) {
NSAutoreleasePool * pool = [[NSAutoreleasePool alloc] init];
int mib[] = {
CTL_NET,
PF_ROUTE,
0,
0,
NET_RT_IFLIST2,
0
};
size_t len;
if (sysctl(mib, 6, NULL, &len, NULL, 0) < 0) {
fprintf(stderr, "sysctl: %s\n", strerror(errno));
exit(1);
}
char *buf = (char *)malloc(len);
if (sysctl(mib, 6, buf, &len, NULL, 0) < 0) {
fprintf(stderr, "sysctl: %s\n", strerror(errno));
exit(1);
}
char *lim = buf + len;
char *next = NULL;
u_int64_t totalibytes = 0;
u_int64_t totalobytes = 0;
for (next = buf; next < lim; ) {
struct if_msghdr *ifm = (struct if_msghdr *)next;
next += ifm->ifm_msglen;
if (ifm->ifm_type == RTM_IFINFO2) {
struct if_msghdr2 *if2m = (struct if_msghdr2 *)ifm;
totalibytes += if2m->ifm_data.ifi_ibytes;
totalobytes += if2m->ifm_data.ifi_obytes;
}
}
printf("total ibytes %qu\tobytes %qu\n", totalibytes, totalobytes);
[pool drain];
return 0;
}
I can't speak to OSX but on linux take a look at /proc/net/dev.
If you do 'cat /proc/net/dev' you should see statistics including 'bytes' - the total number of bytes of data transmitted or received by the interface. You can read the file within your own program.
EDIT:
I didn't read your whole question. This article should help you get started with /proc and has a section on /proc/net/dev.
Also, to list the interfaces you can call ioctl with the SIOCGIFCONF option. You can Google for a decent code example on how to loop through the returned data. Or you can simply pull it out of the /proc.net/dev data mentioned above, which should be easier.
on Linux:
low level: check /sys/class/net/eth0/statistics/
slightly higher level: ip -s link show eth0
graphical: iftop
interactive: iptraf