add alarm for TCP/AttemptsFail spikes

[netdata.git] / conf.d / health.d / tcp_resets.conf

diff --git a/conf.d/health.d/tcp_resets.conf b/conf.d/health.d/tcp_resets.conf
index a7cb325f136c878b019fc3d6877a9ff66e9c2b46..4655e68c421be712044f963d06903847d1d2c317 100644 (file)
--- a/conf.d/health.d/tcp_resets.conf
+++ b/conf.d/health.d/tcp_resets.conf
@@ -12,21 +12,41 @@
       to: sysadmin
 
 # -----------------------------------------------------------------------------
+# tcp resets this host sends
 
-   alarm: 1m_ipv4_tcp_resets
+   alarm: 1m_ipv4_tcp_resets_sent
       on: ipv4.tcphandshake
   lookup: average -1m at -10s unaligned absolute of OutRsts
    units: tcp resets/s
    every: 10s
     info: average TCP RESETS this host is sending, over the last minute
 
-   alarm: 10s_ipv4_tcp_resets
+   alarm: 10s_ipv4_tcp_resets_sent
       on: ipv4.tcphandshake
   lookup: average -10s unaligned absolute of OutRsts
    units: tcp resets/s
    every: 10s
-    warn: $this > ((($1m_ipv4_tcp_resets < 5)?(5):($1m_ipv4_tcp_resets)) * (($status >= $WARNING)  ? (1) : (4)))
+    warn: $this > ((($1m_ipv4_tcp_resets_sent < 5)?(5):($1m_ipv4_tcp_resets_sent)) * (($status >= $WARNING)  ? (1) : (4)))
    delay: up 0 down 60m multiplier 1.2 max 2h
     info: average TCP RESETS this host is sending, over the last 10 seconds (this can be an indication that a port scan is made, or that a service running on this host has crashed)
       to: silent
 
+# -----------------------------------------------------------------------------
+# tcp resets this host receives
+
+   alarm: 1m_ipv4_tcp_resets_received
+      on: ipv4.tcphandshake
+  lookup: average -1m at -10s unaligned absolute of AttemptFails
+   units: tcp resets/s
+   every: 10s
+    info: average TCP RESETS this host is sending, over the last minute
+
+   alarm: 10s_ipv4_tcp_resets_received
+      on: ipv4.tcphandshake
+  lookup: average -10s unaligned absolute of AttemptFails
+   units: tcp resets/s
+   every: 10s
+    warn: $this > ((($1m_ipv4_tcp_resets_received < 5)?(5):($1m_ipv4_tcp_resets_received)) * (($status >= $WARNING)  ? (1) : (4)))
+   delay: up 0 down 60m multiplier 1.2 max 2h
+    info: average TCP RESETS this host is sending, over the last 10 seconds (this can be an indication that a port scan is made, or that a service running on this host has crashed)
+      to: silent