fix: Improve traffic tracking accuracy and reduce CPU usage
- Fix tracker double-counting by capturing only on external interface instead of all interfaces (-i any → -i $CAPTURE_IFACE) - Increase docker logs tail from 30 to 200 for reliable stats capture - Add stats caching to show last known values when logs are busy - Reduce dashboard refresh interval from 4s to 10s for lower CPU load - Add docker stats caching (refresh every 20s) to reduce docker overhead
This commit is contained in:
SamNet-dev
72
conduit.sh
72
conduit.sh
@@ -1293,7 +1293,7 @@ show_dashboard() {
|
||||
echo -e "\033[K"
|
||||
fi
|
||||
|
||||
echo -e "${BOLD}Refreshes every 5 seconds. Press any key to return to menu...${NC}\033[K"
|
||||
echo -e "${BOLD}Refreshes every 10 seconds. Press any key to return to menu...${NC}\033[K"
|
||||
|
||||
# Clear any leftover lines below the dashboard content (Erase to End of Display)
|
||||
# This only cleans up if the dashboard gets shorter
|
||||
@@ -1301,9 +1301,9 @@ show_dashboard() {
|
||||
printf "\033[J"
|
||||
fi
|
||||
|
||||
# Wait 4 seconds for keypress (compensating for processing time)
|
||||
# Wait 10 seconds for keypress (balances responsiveness with CPU usage)
|
||||
# Redirect from /dev/tty ensures it works when the script is piped
|
||||
if read -t 4 -n 1 -s < /dev/tty 2>/dev/null; then
|
||||
if read -t 10 -n 1 -s < /dev/tty 2>/dev/null; then
|
||||
stop_dashboard=1
|
||||
fi
|
||||
done
|
||||
@@ -1630,10 +1630,15 @@ touch "$STATS_FILE" "$IPS_FILE"
|
||||
TCPDUMP_BIN=$(command -v tcpdump 2>/dev/null || echo "tcpdump")
|
||||
AWK_BIN=$(command -v gawk 2>/dev/null || command -v awk 2>/dev/null || echo "awk")
|
||||
|
||||
# Detect local IP
|
||||
# Detect local IP and primary external interface
|
||||
LOCAL_IP=$(ip route get 1.1.1.1 2>/dev/null | awk '{for(i=1;i<=NF;i++) if($i=="src") print $(i+1)}')
|
||||
[ -z "$LOCAL_IP" ] && LOCAL_IP=$(hostname -I 2>/dev/null | awk '{print $1}')
|
||||
|
||||
# Detect primary external interface (to avoid double-counting on docker bridges)
|
||||
CAPTURE_IFACE=$(ip route get 1.1.1.1 2>/dev/null | awk '{for(i=1;i<=NF;i++) if($i=="dev") print $(i+1)}')
|
||||
[ -z "$CAPTURE_IFACE" ] && CAPTURE_IFACE=$(ip route show default 2>/dev/null | awk '{print $5; exit}')
|
||||
[ -z "$CAPTURE_IFACE" ] && CAPTURE_IFACE="any"
|
||||
|
||||
# Batch process: resolve GeoIP + merge into cumulative files in bulk
|
||||
process_batch() {
|
||||
local batch="$1"
|
||||
@@ -1861,7 +1866,7 @@ while true; do
|
||||
break
|
||||
fi
|
||||
fi
|
||||
done < <($TCPDUMP_BIN -tt -l -ni any -n -q -s 96 "(tcp or udp) and not port 22" 2>/dev/null | $AWK_BIN -v local_ip="$LOCAL_IP" '
|
||||
done < <($TCPDUMP_BIN -tt -l -ni "$CAPTURE_IFACE" -n -q -s 96 "(tcp or udp) and not port 22" 2>/dev/null | $AWK_BIN -v local_ip="$LOCAL_IP" '
|
||||
BEGIN { last_sync = 0; OFMT = "%.0f"; CONVFMT = "%.0f" }
|
||||
{
|
||||
# Parse timestamp
|
||||
@@ -2035,7 +2040,7 @@ show_advanced_stats() {
|
||||
local cname=$(get_container_name $ci)
|
||||
if echo "$docker_ps_cache" | grep -q "^${cname}$"; then
|
||||
adv_running_names+=" $cname"
|
||||
( docker logs --tail 30 "$cname" 2>&1 | grep "\[STATS\]" | tail -1 > "$_adv_tmpdir/logs_${ci}" ) &
|
||||
( docker logs --tail 200 "$cname" 2>&1 | grep "\[STATS\]" | tail -1 > "$_adv_tmpdir/logs_${ci}" ) &
|
||||
fi
|
||||
done
|
||||
local adv_all_stats=""
|
||||
@@ -2319,7 +2324,7 @@ show_peers() {
|
||||
for ci in $(seq 1 $CONTAINER_COUNT); do
|
||||
local cname=$(get_container_name $ci)
|
||||
if echo "$docker_ps_cache" | grep -q "^${cname}$"; then
|
||||
( docker logs --tail 30 "$cname" 2>&1 | grep "\[STATS\]" | tail -1 > "$_peer_tmpdir/logs_${ci}" ) &
|
||||
( docker logs --tail 200 "$cname" 2>&1 | grep "\[STATS\]" | tail -1 > "$_peer_tmpdir/logs_${ci}" ) &
|
||||
fi
|
||||
done
|
||||
wait
|
||||
@@ -2466,6 +2471,13 @@ get_net_speed() {
|
||||
fi
|
||||
}
|
||||
|
||||
# Global cache for container stats (persists between show_status calls)
|
||||
declare -A _STATS_CACHE_UP _STATS_CACHE_DOWN _STATS_CACHE_CONN _STATS_CACHE_CING
|
||||
|
||||
# Global cache for docker stats (CPU/RAM) - refreshes every 2 cycles (20s)
|
||||
_DOCKER_STATS_CACHE=""
|
||||
_DOCKER_STATS_CYCLE=0
|
||||
|
||||
show_status() {
|
||||
local mode="${1:-normal}" # 'live' mode adds line clearing
|
||||
local EL=""
|
||||
@@ -2500,7 +2512,7 @@ show_status() {
|
||||
if echo "$docker_ps_cache" | grep -q "[[:space:]]${cname}$"; then
|
||||
_c_running[$i]=true
|
||||
running_count=$((running_count + 1))
|
||||
( docker logs --tail 30 "$cname" 2>&1 | grep "\[STATS\]" | tail -1 > "$_st_tmpdir/logs_${i}" ) &
|
||||
( docker logs --tail 200 "$cname" 2>&1 | grep "\[STATS\]" | tail -1 > "$_st_tmpdir/logs_${i}" ) &
|
||||
fi
|
||||
done
|
||||
wait
|
||||
@@ -2524,11 +2536,26 @@ show_status() {
|
||||
_c_cing[$i]="${c_connecting:-0}"
|
||||
_c_up[$i]="${c_up_val}"
|
||||
_c_down[$i]="${c_down_val}"
|
||||
# Update global cache with fresh data
|
||||
_STATS_CACHE_UP[$i]="${c_up_val}"
|
||||
_STATS_CACHE_DOWN[$i]="${c_down_val}"
|
||||
_STATS_CACHE_CONN[$i]="${c_connected:-0}"
|
||||
_STATS_CACHE_CING[$i]="${c_connecting:-0}"
|
||||
total_connecting=$((total_connecting + ${c_connecting:-0}))
|
||||
total_connected=$((total_connected + ${c_connected:-0}))
|
||||
if [ -z "$uptime" ]; then
|
||||
uptime="${c_uptime_val}"
|
||||
fi
|
||||
else
|
||||
# No stats in logs - use cached values if available
|
||||
if [ -n "${_STATS_CACHE_UP[$i]}" ]; then
|
||||
_c_up[$i]="${_STATS_CACHE_UP[$i]}"
|
||||
_c_down[$i]="${_STATS_CACHE_DOWN[$i]}"
|
||||
_c_conn[$i]="${_STATS_CACHE_CONN[$i]:-0}"
|
||||
_c_cing[$i]="${_STATS_CACHE_CING[$i]:-0}"
|
||||
total_connecting=$((total_connecting + ${_c_cing[$i]:-0}))
|
||||
total_connected=$((total_connected + ${_c_conn[$i]:-0}))
|
||||
fi
|
||||
fi
|
||||
fi
|
||||
done
|
||||
@@ -2588,15 +2615,24 @@ show_status() {
|
||||
|
||||
if [ "$running_count" -gt 0 ]; then
|
||||
|
||||
# Run all 3 resource stat calls in parallel
|
||||
# Run resource stat calls (docker stats cached every 2 cycles for CPU savings)
|
||||
local _rs_tmpdir=$(mktemp -d /tmp/.conduit_rs.XXXXXX)
|
||||
# mktemp already created the directory
|
||||
( get_container_stats > "$_rs_tmpdir/cstats" ) &
|
||||
_DOCKER_STATS_CYCLE=$(( (_DOCKER_STATS_CYCLE + 1) % 2 ))
|
||||
if [ "$_DOCKER_STATS_CYCLE" -eq 1 ] || [ -z "$_DOCKER_STATS_CACHE" ]; then
|
||||
# Fresh fetch cycle - get new docker stats
|
||||
( get_container_stats > "$_rs_tmpdir/cstats" ) &
|
||||
fi
|
||||
( get_system_stats > "$_rs_tmpdir/sys" ) &
|
||||
( get_net_speed > "$_rs_tmpdir/net" ) &
|
||||
wait
|
||||
|
||||
local stats=$(cat "$_rs_tmpdir/cstats" 2>/dev/null)
|
||||
local stats
|
||||
if [ -f "$_rs_tmpdir/cstats" ]; then
|
||||
stats=$(cat "$_rs_tmpdir/cstats" 2>/dev/null)
|
||||
_DOCKER_STATS_CACHE="$stats"
|
||||
else
|
||||
stats="$_DOCKER_STATS_CACHE"
|
||||
fi
|
||||
local sys_stats=$(cat "$_rs_tmpdir/sys" 2>/dev/null)
|
||||
local net_speed=$(cat "$_rs_tmpdir/net" 2>/dev/null)
|
||||
rm -rf "$_rs_tmpdir"
|
||||
@@ -3495,7 +3531,7 @@ manage_containers() {
|
||||
if echo "$docker_ps_cache" | grep -q "^${cname}$"; then
|
||||
running_names+=" $cname"
|
||||
# Fetch logs in parallel background jobs
|
||||
( docker logs --tail 30 "$cname" 2>&1 | grep "\[STATS\]" | tail -1 > "$_mc_tmpdir/logs_${ci}" ) &
|
||||
( docker logs --tail 200 "$cname" 2>&1 | grep "\[STATS\]" | tail -1 > "$_mc_tmpdir/logs_${ci}" ) &
|
||||
fi
|
||||
done
|
||||
# Fetch stats in parallel with logs
|
||||
@@ -3538,6 +3574,16 @@ manage_containers() {
|
||||
c_down="${mc_down:-"-"}"
|
||||
[ -z "$c_up" ] && c_up="-"
|
||||
[ -z "$c_down" ] && c_down="-"
|
||||
# Update global cache
|
||||
_STATS_CACHE_UP[$ci]="${mc_up}"
|
||||
_STATS_CACHE_DOWN[$ci]="${mc_down}"
|
||||
_STATS_CACHE_CONN[$ci]="${conn:-0}"
|
||||
_STATS_CACHE_CING[$ci]="${cing:-0}"
|
||||
elif [ -n "${_STATS_CACHE_UP[$ci]}" ]; then
|
||||
# Use cached values as fallback
|
||||
c_clients="${_STATS_CACHE_CONN[$ci]:-0}/${_STATS_CACHE_CING[$ci]:-0}"
|
||||
c_up="${_STATS_CACHE_UP[$ci]:-"-"}"
|
||||
c_down="${_STATS_CACHE_DOWN[$ci]:-"-"}"
|
||||
fi
|
||||
local dstats_line=$(echo "$all_dstats" | grep "^${cname} " 2>/dev/null)
|
||||
if [ -n "$dstats_line" ]; then
|
||||
|
||||
Reference in New Issue
Block a user