add checkrt to indicate when a site has new or open tickets

add checkescalation to infer the penalty applied to a site based on the state
    of it's site and slices
add extra RT configuration fields to auth.py

diff --git a/nagios/auth.py b/nagios/auth.py
index 101299b..4755a8b 100644 (file)
--- a/nagios/auth.py
+++ b/nagios/auth.py
@@ -4,4 +4,11 @@
 #auth = {'Username' : '', 'AuthMethod' : 'password', 'AuthString' : ''}
 auth = {'AuthMethod' : "anonymous"}
 plc = "https://boot.planet-lab.org/PLCAPI/"
+www = "https://www.planet-lab.org"
+
+RTSERVER="https://rt.planet-lab.org/"
+RTUSER="monitor"
+RTPASSWD=""
+RTDEBUG="0"
+
 

diff --git a/nagios/plugins/checkescalation.py b/nagios/plugins/checkescalation.py
new file mode 100755 (executable)
index 0000000..1ab1156
--- /dev/null
+++ b/nagios/plugins/checkescalation.py
@@ -0,0 +1,59 @@
+#!/usr/bin/python
+
+import time
+import sys
+import plc
+
+def argv_to_dict(argv):
+    """
+        NOTE: very bare-bones, no error checking, will fail easily.
+    """
+    d = {}
+    prev=None
+    for a in argv:
+        if "--" == a[0:2]:
+            prev = a[2:]
+        elif "-" == a[0:1]:
+            prev = a[1:]
+        else:
+            d[prev] = a
+    return d
+
+def main():
+    d = argv_to_dict(sys.argv[1:])
+
+    site = None
+    if 'site' in d:
+        site = d['site'].replace('site-cluster-for-','')
+    else:
+        print "No site specified"
+        sys.exit(1)
+
+    # define escalation by observed properties about the site.
+    # i.e. EXEMPT, level 0 (site enabled and slices ok), level 1 (site disabled), level 2 (slices)
+    if plc.isSiteExempt(site):
+        tags = plc.api.GetSiteTags({'login_base' : site, 'tagname' : 'exempt_site_until'})
+        print "Exempt: %s is exempt until %s" % (site, tags[0]['value'])
+        sys.exit(0)
+
+    extra_str = ""
+    
+    # are slices disabled?
+    slices_enabled = plc.areSlicesEnabled(site)
+    if isinstance(slices_enabled, bool) and not slices_enabled:
+        print "Level >= 2: slices are disabled at %s" % (site)
+        sys.exit(0)
+    elif isinstance(slices_enabled, type(None)):
+        extra_str = "And, no slices."
+
+    # Site is not exempt, so is it disabled?
+    if not plc.isSiteEnabled(site):
+        print "Level >= 1: site is disabled at %s. %s" % (site, extra_str)
+        sys.exit(0)
+
+    print "Level 0: no policy applied to site %s" % (site)
+    sys.exit(0)
+
+        
+if __name__ == '__main__':
+    main()

diff --git a/nagios/plugins/checkrt.py b/nagios/plugins/checkrt.py
new file mode 100755 (executable)
index 0000000..befb1e3
--- /dev/null
+++ b/nagios/plugins/checkrt.py
@@ -0,0 +1,72 @@
+#!/usr/bin/python
+
+import os
+import time
+import sys
+import auth
+
+def argv_to_dict(argv):
+    """
+        NOTE: very bare-bones, no error checking, will fail easily.
+    """
+    d = {}
+    prev=None
+    for a in argv:
+        if "--" == a[0:2]:
+            prev = a[2:]
+        elif "-" == a[0:1]:
+            prev = a[1:]
+        else:
+            d[prev] = a
+    return d
+
+def main():
+    d = argv_to_dict(sys.argv[1:])
+
+    if 'pattern' in d or 'p' in d:
+        try:
+            pattern = d['pattern']
+        except:
+            pattern = d['p']
+    else:
+        print "UNKNOWN: Argument error"
+        sys.exit(3)
+
+
+    # TODO: check that RT is configured correctly
+    os.environ["RTSERVER"] = auth.RTSERVER
+    os.environ["RTUSER"] = auth.RTUSER
+    os.environ["RTPASSWD"] = auth.RTPASSWD
+    os.environ["RTDEBUG"] = auth.RTDEBUG
+
+    # TODO: may need to add a timeout
+    # NOTE: RT3.8
+    query = "Subject like '%%%s%%' and Queue='Monitor' and ( Status='new' or Status='open' )" % pattern
+    cmd = """rt ls -s -t ticket "%s" 2>&1 """ % query
+    cmd = cmd + """| grep -vi "no match" | wc -l """
+
+    out = os.popen(cmd, 'r')
+    open_tickets = out.read()
+
+    try:
+        open_tickets_i = int(open_tickets)
+    except:
+        print "UNKNOWN: failed to convert %s to open ticket count" % open_tickets
+        sys.exit(3)
+
+    if open_tickets_i == 0:
+        print "OK: no open tickets for site"
+        sys.exit(0)
+    elif open_tickets_i != 0:
+        print "WARNING: %s open tickets" % open_tickets_i
+        sys.exit(1)
+    else:
+        print "FAKE-CRITICAL: RT check failed"
+        sys.exit(2)
+
+
+if __name__ == '__main__':
+    f = open("/tmp/checkpcu", 'a')
+    f.write("checkpcu %s %s\n" % (time.time(), " ".join(sys.argv[1:])))
+    f.close()
+    main()