X-Git-Url: http://git.onelab.eu/?a=blobdiff_plain;f=README.txt;h=6863776c6b0dabf12921355082eabd2b4dff8cc8;hb=55a9635aa8015ed709b55ca19e2f9472a1eda923;hp=cf118317cbe1b76d672c80869c0ffa9bbece9444;hpb=e35960f004a9666f7f36582ef70307ceeb880d7a;p=mom.git

diff --git a/README.txt b/README.txt
index cf11831..6863776 100644
--- a/README.txt
+++ b/README.txt
@@ -1,3 +1,5 @@
+# Copyright...something.
+
 Overview
 --------
 
@@ -5,26 +7,17 @@ Overview
 much memory, and by rebooting the machine once the swap space is
 nearly full.
 
-Specifically, pl_mom takes the following actions based on the amount
+Specifically, swapmon takes the following actions based on the amount
 of consumed swap space:
 
-* at 85% full, pl_mom starts logging the available swap space to
-  syslogd
+* at 80% full, swapmon starts logging the available swap space to
+  syslogd, restart largest slice and send mail.
 
-* at 90% full, pl_mom resets the slice that is consuming the most
-  physical memory and sends an alert mail to that slice and the pl-mom
-  alias 
+  After 3 restarts, kill slice.
 
-* at 95% full, pl_mom reboots the machine and sends mail to the pl-mom
+* at 95% full, swapmon reboots the machine and sends mail to the pl-mom
   alias
 
-Pl_mom is organized as two processes.  The reboot process runs every
-second, and the slice reset process runs every 30 seconds.  The reboot
-process is very lightweight and only reads /proc/swaps.  However, the
-slice reset process may pull data from the slicestat sensor on port
-3100, which can potentially take a long time (on the order of minutes
-if there are thousands of processes on the machine).
-
 
 Building
 ---------