NDT-DEV email list created

Revision: 319
Author: rcarlson501
Date: Mon Mar 22 20:56:00 2010
Log: Updating files to handle case where a write() or read() can return due
to an interrupt.  In this case no date is written/read and the server may
not move to the next test.  This would cause the server to timeout the  
client
and the client would report a failed test.

Previous changes also include a reduction in the firewall test time.  The
original version had a max time of 30 sec.  This may cause an alarm() signal
to go off terminating the server process.  The max time was reduced to 3  
sec.

RAC 3/22/10


http://code.google.com/p/ndt/source/detail?r=319

Modified:
 /trunk/src/testoptions.c
 /trunk/src/web100-pcap.c
 /trunk/src/web100-util.c
 /trunk/src/web100srv.c

=======================================
--- /trunk/src/testoptions.c    Sun Mar 21 11:04:42 2010
+++ /trunk/src/testoptions.c    Mon Mar 22 20:56:00 2010
@@ -412,7 +412,7 @@
     close(midfd);
     close(options->midsockfd);
     send_msg(ctlsockfd, TEST_FINALIZE, "", 0);
-    log_println(1, " <-------------------->");
+    log_println(1, " <--------- %d ----------->", options->child0);
     setCurrentTest(TEST_NONE);
   }
   /* I2AddrFree(midsrv_addr); */
@@ -701,6 +701,8 @@
       }
       if (ret > 0) {
         n = read(recvsfd, buff, sizeof(buff));
+       if ((n == -1) && (errno == EINTR))
+         continue;
         if (n == 0)
           break;
         bytes += n;
@@ -786,12 +788,18 @@
     send_msg(ctlsockfd, TEST_FINALIZE, "", 0);

     if (getuid() == 0) {
-      write(mon_pipe1[1], "c", 1);
+      for (i=0; i<5; i++) {
+        ret = write(mon_pipe1[1], "c", 1);
+       if (ret == 1)
+         break;
+       if ((ret == -1) && (errno == EINTR))
+         continue;
+      }
       close(mon_pipe1[0]);
       close(mon_pipe1[1]);
     }

-    log_println(1, " <------------------------->");
+    log_println(1, " <----------- %d -------------->",  
testOptions->child0);
     setCurrentTest(TEST_NONE);
   }
   /* I2AddrFree(c2ssrv_addr); */
@@ -1133,6 +1141,8 @@
         }

         n = write(xmitsfd, buff, RECLTH);
+       if ((n == -1) && (errno == EINTR))
+         continue;
         if (n < 0)
           break;
         bytes += n;
@@ -1264,12 +1274,18 @@
 	log_println(6, "S2C test - failed to send finalize message to pid=%d",  
mon_pid2);

     if (getuid() == 0) {
-      write(mon_pipe2[1], "c", 1);
+      for (i=0; i<5; i++) {
+        ret = write(mon_pipe2[1], "c", 1);
+       if (ret == 1)
+         break;
+       if ((ret == -1) && (errno == EINTR))
+         continue;
+      }
       close(mon_pipe2[0]);
       close(mon_pipe2[1]);
     }

-    log_println(1, " <------------------------->");
+    log_println(1, " <------------ %d ------------->",  
testOptions->child0);
     setCurrentTest(TEST_NONE);
   }
   /* I2AddrFree(s2csrv_addr); */
=======================================
--- /trunk/src/web100-pcap.c    Tue Feb  9 17:30:07 2010
+++ /trunk/src/web100-pcap.c    Mon Mar 22 20:56:00 2010
@@ -207,6 +207,7 @@
        char buff[256];
        int tzoffset = 6;
        FILE *fp;
+       int j;

   assert(cur);

@@ -343,10 +344,16 @@
 		cur->links[5], cur->links[6], cur->links[7], cur->links[8],  
cur->links[9],
 		cur->links[10], cur->links[11], cur->totalspd2, cur->inc_cnt,  
cur->dec_cnt,
                cur->same_cnt, cur->timeout, cur->dupack, ifspeed);
-       i = write(monitor_pipe[1], buff, strlen(buff));
-  log_println(6, "wrote %d bytes: link counters are '%s'", i, buff);
-  log_println(6, "#$#$#$#$ pcap routine says window increases = %d,  
decreases = %d, no change = %d",
-      cur->inc_cnt, cur->dec_cnt, cur->same_cnt);
+       for (j=0; j<5; j++) {
+         i = write(monitor_pipe[1], buff, strlen(buff));
+         if (i == strlen(buff))
+           break;
+         if ((i == -1) && (errno == EINTR))
+           continue;
+       }
+        log_println(6, "wrote %d bytes: link counters are '%s'", i, buff);
+        log_println(6, "#$#$#$#$ pcap routine says window increases = %d,  
decreases = %d, no change = %d",
+                                cur->inc_cnt, cur->dec_cnt, cur->same_cnt);
 }

 void calculate_spd(struct spdpair *cur, struct spdpair *cur2, int portA,  
int portB)
@@ -679,6 +686,7 @@
   pcap_if_t *alldevs, *dp;
   pcap_addr_t *curAddr;
   DIR *dip;
+  int rc;

   cnt = -1;  /* read forever, or until end of file */
   sig1 = 0;
@@ -865,10 +873,23 @@
   }

   printer = (pcap_handler) print_speed;
-  if (dumptrace == 0)
-     write(monitor_pipe[1], "Ready", 6);
-  else
-     write(monitor_pipe[1], dir, strlen(dir));
+  if (dumptrace == 0) {
+     for (i=0; i<5; i++) {
+       rc = write(monitor_pipe[1], "Ready", 6);
+       if (rc == 6)
+        break;
+       if ((rc == -1) && (errno == EINTR))
+        continue;
+     }
+  } else {
+     for (i=0; i<5; i++) {
+       rc = write(monitor_pipe[1], dir, strlen(dir));
+       if (rc == strlen(dir))
+        break;
+       if ((rc == -1) && (errno == EINTR))
+        continue;
+     }
+  }

   /* kill process off if parent doesn't send a signal. */
   alarm(60);
=======================================
--- /trunk/src/web100-util.c    Sun Mar 21 11:04:42 2010
+++ /trunk/src/web100-util.c    Mon Mar 22 20:56:00 2010
@@ -163,6 +163,8 @@
     }

     k = write(sock, sndbuff, octets);
+    if ((k == -1) && (errno == EINTR))
+      continue;
     if (k < 0 )
       break;
   }
=======================================
--- /trunk/src/web100srv.c      Sun Mar 21 12:11:15 2010
+++ /trunk/src/web100srv.c      Mon Mar 22 20:56:00 2010
@@ -1704,7 +1704,7 @@
     }

     if ((waiting < 0) || (mclients < 0)) {
-	log_println(6, "Fault: Negtive number of clents waiting=$d, mclients=%d,  
nuke them", waiting, mclients);
+	log_println(6, "Fault: Negative number of clents waiting=%d, mclients=%d,  
nuke them", waiting, mclients);
        while (head_ptr != NULL) {
             tpid = head_ptr->pid;
            child_sig(-1);
@@ -1745,9 +1745,7 @@
        log_println(5, "Empty slot in test queue, find new client to 
dispatch");
        tmp_ptr = head_ptr;
        mchild = head_ptr;
-	log_println(2, "starting queue look for non-running client current=%d,  
running=%d, next=0x%x",
-                       tmp_ptr->pid, tmp_ptr->running, tmp_ptr->next);
-       while (tmp_ptr->next != NULL) {
+       while (tmp_ptr != NULL) {
 	   log_println(2, "walking queue look for non-running client current=%d,  
running=%d, next=0x%x",
                        tmp_ptr->pid, tmp_ptr->running, tmp_ptr->next);
            if (tmp_ptr->running == 0) {
@@ -1758,8 +1756,32 @@
        }
        if ((tmp_ptr->next == NULL) && (tmp_ptr->running == 0))
          mchild = tmp_ptr;
-       if (mchild != head_ptr)
-         goto multi_client;
+       if (mchild != head_ptr) {
+         tmp_ptr = mchild;
+         /* update queued clients, send message to client when it moves
+          * up in the queue enough to get closer to running a test.  This 
happens
+          * when the client falls into the next lower maxquee bin
+          * RAC 3/21/10
+          */
+         if (waiting > (2*max_clients)) {
+           for (i=max_clients; i<=waiting; i++) {
+             if (tmp_ptr == NULL)
+               break;
+             if (i == (2*max_clients)) {
+	        log_println(6, "Updating client list position client %d moved now  
45 sec away",
+                       tmp_ptr->pid);
+                send_msg(tmp_ptr->ctlsockfd, SRV_QUEUE, "1", 1);
+             }
+             if (i == (3*max_clients)) {
+	      log_println(6, "Updating client list position client %d moved now  
90 sec away",
+                       tmp_ptr->pid);
+              send_msg(tmp_ptr->ctlsockfd, SRV_QUEUE, "2", 1);
+             }
+             tmp_ptr = tmp_ptr->next;
+           }
+           goto multi_client;
+         }
+       }
     }

     if ((multiple == 1) && (mclients > waiting)) {
@@ -2127,7 +2149,7 @@
          mchild->stime = time(0);
          mchild->running = 1;
          mclients++;
-          sprintf(tmpstr, "go %d %s", t_opts, test_suite);
+          sprintf(tmpstr, "go %d %s", t_opts, mchild->tests);
           send_msg(mchild->ctlsockfd, SRV_QUEUE, "0", 1);
          for (i=0; i<5; i++) {
             rc = write(mchild->pipe, tmpstr, strlen(tmpstr));
@@ -2182,7 +2204,9 @@
  * should be checked for and the read() restarted if necessary
  * RAC 3/18/10
  */
-          read(chld_pipe[0], buff, 32);
+          rc = read(chld_pipe[0], buff, 32);
+         if ((rc == -1) && (errno == EINTR))
+           continue;
           if (strncmp(buff, "go", 2) == 0) {
             log_println(6, "Got 'go' signal from parent, ready to start  
testing");
             break;