Skip to content
Closed
Show file tree
Hide file tree
Changes from all commits
Commits
File filter

Filter by extension

Filter by extension

Conversations
Failed to load comments.
Loading
Jump to
Jump to file
Failed to load files.
Loading
Diff view
Diff view
18 changes: 0 additions & 18 deletions mgmt/Main.cc
Original file line number Diff line number Diff line change
Expand Up @@ -757,24 +757,6 @@ main(int argc, char **argv)
Debug("lm", "Created Web Agent thread (%" PRId64 ")", (int64_t)webThrId);
lmgmt->listenForProxy();

/* Check the permissions on vip_config */
if (lmgmt->virt_map->enabled) {
char absolute_vipconf_binary[1024];
struct stat buf;

snprintf(absolute_vipconf_binary, sizeof(absolute_vipconf_binary), "%s/vip_config", lmgmt->bin_path);
if (stat(absolute_vipconf_binary, &buf) < 0) {
mgmt_elog(stderr, "[main] Unable to stat vip_config for proper permissions\n");
} else if (!((buf.st_mode & S_ISUID) &&
(buf.st_mode & S_IRWXU) &&
(buf.st_mode & S_IRGRP) &&
(buf.st_mode & S_IXGRP) && (buf.st_mode & S_IROTH) && (buf.st_mode & S_IXOTH))) {
lmgmt->alarm_keeper->signalAlarm(MGMT_ALARM_PROXY_SYSTEM_ERROR,
"Virtual IP Addressing enabled, but improper permissions on '/inktomi/bin/vip_config'"
"[requires: setuid root and at least a+rx]\n");
}
}

ticker = time(NULL);
mgmt_log("[TrafficManager] Setup complete\n");

Expand Down
12 changes: 6 additions & 6 deletions mgmt/cluster/ClusterCom.cc
Original file line number Diff line number Diff line change
Expand Up @@ -63,16 +63,16 @@ drainIncomingChannel(void *arg)
struct sockaddr_in cli_addr;

// Fix for INKqa07688: There was a problem at Genuity where if you
// pulled out the cable on the cluser interface (or just ifconfig'd
// down/up the cluster interface), the fd assocated with that
// inteface would somehow get into a bad state... and the multicast
// pulled out the cable on the cluster interface (or just ifconfig'd
// down/up the cluster interface), the fd associated with that
// interface would somehow get into a bad state... and the multicast
// packets from other nodes wouldn't be received anymore.
//
// The fix for the problem was to close() and re-open the multicast
// socket if we detected that no activity has occured for 30
// socket if we detected that no activity has occurred for 30
// seconds. 30 seconds was based on the default peer_timeout
// (proxy.config.cluster.peer_timeout) value. davey showed that
// this value worked out well experiementally (though more testing
// this value worked out well experimentally (though more testing
// and experimentation would be beneficial).
//
// traffic_manager running w/ no cop: In this case, our select()
Expand Down Expand Up @@ -1236,7 +1236,7 @@ ClusterCom::handleMultiCastAlarmPacket(char *last, char *ip)

/*
* handleMultiCastVMapPacket(...)
* Handles incoming reports from peers about which virtua interfaces
* Handles incoming reports from peers about which virtual interfaces
* they are servicing. This then updates the VMap class to indicate who
* is holding what.
*/
Expand Down
Loading