afs-analyze-avoid-potential-loop-20030423

avoid potentially looping forever in e.g. afs_FetchStatus because
we exit afs_Analyze too early to increment busyCount, and despite having
areq->busyCount non-zero we have servers still not_busy which are already down
that we'll never talk to again. that combined with no conn means we never
touch most of afs_Analyze, exiting early with shouldRetry set.

incrementing busyCount whenever we exit with shouldRetry set should
ensure we can never end up looping forever.

diff --git a/src/afs/afs_analyze.c b/src/afs/afs_analyze.c
index 12e0bf85b661fe38acb0b40f9027da008ac86578..14d0068551026487eee6173cde7f1517d97fab41 100644 (file)
--- a/src/afs/afs_analyze.c
+++ b/src/afs/afs_analyze.c
@@ -469,10 +469,10 @@ int afs_Analyze(register struct conn *aconn, afs_int32 acode,
 
         for (i=0; i < MAXHOSTS; i++) {
            if (tvp->status[i] != not_busy && tvp->status[i] != offline) {
-              tvp->status[i] = not_busy; 
+               tvp->status[i] = not_busy; 
            }
-           if (tvp->status[i] == not_busy)
-                shouldRetry = 1;
+           if (tvp->status[i] == not_busy) 
+               shouldRetry = 1;
         }
         afs_PutVolume(tvp, READ_LOCK);
       } else {
@@ -488,6 +488,9 @@ int afs_Analyze(register struct conn *aconn, afs_int32 acode,
       } else {
        VSleep(afs_BusyWaitPeriod);         /* poll periodically */
       }
+      if (shouldRetry != 0)
+         areq->busyCount++;
+
       return shouldRetry; /* should retry */
     }