[checkmk-commits] Check_MK Git: check_mk: Add new Check for monitoring HSRP redundancy groups on Cisco routers

Module: check_mk Branch: master Commit: f2c5a46d4d466f79094dffa89c615453a1950d33 URL: http://git.mathias-kettner.de/git/?p=check_mk.git;a=commit;h=f2c5a46d4d466f… Author: Florian Heigl &lt;fh(a)mathias-kettner.de&gt; Date: Thu Aug 16 19:16:28 2012 +0200 Add new Check for monitoring HSRP redundancy groups on Cisco routers --- ChangeLog | 1 + checkman/cisco_hsrp | 37 ++++++++++++++++ checks/cisco_hsrp | 117 +++++++++++++++++++++++++++++++++++++++++++++++++++ 3 files changed, 155 insertions(+), 0 deletions(-) diff --git a/ChangeLog b/ChangeLog index 0f12101..e34dad1 100644 --- a/ChangeLog +++ b/ChangeLog @@ -45,6 +45,7 @@ * carel_uniflair_cooling: new check for monitoring datacenter air conditioning by "CAREL" * Added Agent for OpenBSD * Added Checks for MasterGuard UPS devices + * cisco_hsrp: New Check for monitoring HSRP groups on Cisco Routers. (SMIv2 version) WATO: * Added permission to control the "clone host" feature in WATO diff --git a/checkman/cisco_hsrp b/checkman/cisco_hsrp new file mode 100644 index 0000000..77457f5 --- /dev/null +++ b/checkman/cisco_hsrp @@ -0,0 +1,37 @@ +title: Cisco HSRP group status +agents: snmp +author: Florian Heigl &lt;fh(a)mathias-kettner.de&gt; +license: GPLv3 +distribution: check_mk +description: + Cisco routers support virtual redundant interfaces by using HSRP + (Hot Standby Router Protocol). + HSRP is configured by assigning IP interfaces to a group that has a + "floating" virtual IP. + If, for example, a router has an interface failure on either of the routed + interfaces, the protocol lowers the routers priority, and if that priority + drops too low, it will fail over the virtual IP. + + + The check returns {OK} if the HSRP status is a good one ("active", + "standby") and the same as during inventory. + If the states flip from "active" to "standby" or vice-versa, the check goes + to {WARN}, assuming that HSRP is doing it's job. + + Should the status ever be a different one that for example points to HSRP + still initializing, the check will return {CRIT} as HSRP is assumet to be + inoperable in these states. + +perfdata: + none + + +item: + The Virtual IP of the failover group. + +inventory: + The check reads the HSRP MIB and creates one service per virtual IP. It also + stores the HSRP state of + the IP as seen by the monitored device, normally either standby or active. + It also stores the ID of + the failover group. (For future use) diff --git a/checks/cisco_hsrp b/checks/cisco_hsrp new file mode 100644 index 0000000..5593cb0 --- /dev/null +++ b/checks/cisco_hsrp @@ -0,0 +1,117 @@ +#!/usr/bin/python + + + +# Good docs: +# http://www.cisco.com/en/US/tech/tk648/tk362/technologies_tech_note09186a008… +#.1.3.6.1.4.1.9.9.106.1.1.1.0 5 +# cHsrpGrpTable +########################### +#.1.3.6.1.4.1.9.9.106.1.2.1.1.2.1.192 "HSRP Secret key here" +#.1.3.6.1.4.1.9.9.106.1.2.1.1.2.7.193 "HSRP Secret key here" +#.1.3.6.1.4.1.9.9.106.1.2.1.1.3.1.192 100 +#.1.3.6.1.4.1.9.9.106.1.2.1.1.3.7.193 100 +#.1.3.6.1.4.1.9.9.106.1.2.1.1.4.1.192 1 +#.1.3.6.1.4.1.9.9.106.1.2.1.1.4.7.193 1 +#.1.3.6.1.4.1.9.9.106.1.2.1.1.5.1.192 300 +#.1.3.6.1.4.1.9.9.106.1.2.1.1.5.7.193 300 +#.1.3.6.1.4.1.9.9.106.1.2.1.1.6.1.192 2 +#.1.3.6.1.4.1.9.9.106.1.2.1.1.6.7.193 2 +#.1.3.6.1.4.1.9.9.106.1.2.1.1.7.1.192 0 +#.1.3.6.1.4.1.9.9.106.1.2.1.1.7.7.193 0 +#.1.3.6.1.4.1.9.9.106.1.2.1.1.8.1.192 0 +#.1.3.6.1.4.1.9.9.106.1.2.1.1.8.7.193 0 +#.1.3.6.1.4.1.9.9.106.1.2.1.1.9.1.192 3000 +#.1.3.6.1.4.1.9.9.106.1.2.1.1.9.7.193 3000 +#.1.3.6.1.4.1.9.9.106.1.2.1.1.10.1.192 10000 +#.1.3.6.1.4.1.9.9.106.1.2.1.1.10.7.193 10000 +#.1.3.6.1.4.1.9.9.106.1.2.1.1.11.1.192 192.168.10.4 +#.1.3.6.1.4.1.9.9.106.1.2.1.1.11.7.193 172.20.10.20 <- hsrp ip +#.1.3.6.1.4.1.9.9.106.1.2.1.1.12.1.192 1 +#.1.3.6.1.4.1.9.9.106.1.2.1.1.12.7.193 1 +# HSRP Monitored IP interfaces. If any of those go down, the priority of +# the router will be lowered. +#.1.3.6.1.4.1.9.9.106.1.2.1.1.13.1.192 192.168.10.5 <- ip Router 1 int 1 +#.1.3.6.1.4.1.9.9.106.1.2.1.1.13.7.193 172.20.10.21 <- ip Router 2 int 7 +#.1.3.6.1.4.1.9.9.106.1.2.1.1.14.1.192 192.168.10.6 <- ip Router 1 int 1 +#.1.3.6.1.4.1.9.9.106.1.2.1.1.14.7.193 172.20.10.22 <- ip Router 2 int 7 +#.1.3.6.1.4.1.9.9.106.1.2.1.1.15.1.192 6 <- group #1 "standby" state +#.1.3.6.1.4.1.9.9.106.1.2.1.1.15.7.193 6 <- group #2 "standby" state +#.1.3.6.1.4.1.9.9.106.1.2.1.1.16.1.192 "00 00 0C 07 AC C0 " +#.1.3.6.1.4.1.9.9.106.1.2.1.1.16.7.193 "00 00 0C 07 AC C1 " +#.1.3.6.1.4.1.9.9.106.1.2.1.1.17.1.192 1 +#.1.3.6.1.4.1.9.9.106.1.2.1.1.17.7.193 1 + + +# we'll be alerting if the state is not either 5 or 6. +# We could also not inventorize if the state isn't 5/6 but +# since you have to configure a group, to even show up in the +# MIB it's supposedly ok to alert if something isn't right there. +# otherwise modify the inventory. +hsrp_states = { 1: "initial", 2: "learn", 3: "listen", 4: "speak", 5: "standby", 6: "active" } + + +def inventory_cisco_hsrp(info): + + inventory = [] + for line in info: + hsrp_grp_entry, vip, actrouter, sbrouter, hsrp_state, vmac = line + interface_index, hsrp_grp = hsrp_grp_entry.split(".") + # inventorize HSRP group name+IP and the standby state as seen from "this" box. + inventory.append( (vip, (hsrp_grp, int(hsrp_state))) ) + + return inventory + + +def check_cisco_hsrp(item, params, info): + + hsrp_grp_wanted, hsrp_state_wanted = params + + for line in info: + hsrp_grp_entry, vip, actrouter, sbrouter, hsrp_state, vmac = line + interface_index, hsrp_grp = hsrp_grp_entry.split(".") + hsrp_state = int(hsrp_state) + + if vip == item: + # FIXME: This should be shorter. + # Validate that we the inventorized state is a "good one" + # if it's also the one we have now, then we're fine. + + if hsrp_state_wanted == 5 and hsrp_state == hsrp_state_wanted: + state = 0 + msgtxt = "Redundancy Group %s is OK" % hsrp_grp + elif hsrp_state_wanted == 6 and hsrp_state == hsrp_state_wanted: + state = 0 + msgtxt = "Redundancy Group %s is OK" % hsrp_grp + # otherwise if it's a good one, but flipped, then we are in a failover + elif hsrp_state == 5 or hsrp_state == 6: + state = 1 + msgtxt = "Redundancy Group %s has failed over" % hsrp_grp + # anything else must be a non-operative state already + else: + state = 2 + msgtxt = "Redundancy Group %s is %s" % ( hsrp_grp, hsrp_states[hsrp_state]) + + return (state, nagios_state_names[state] + " - " + msgtxt) + + return (3, "UNKNOWN - HSRP Group %s not found in Agent output" % hsrp_grp_wanted ) + + + +# FIXME: Outdated format. Fix after discussing WATO options. +check_info["cisco_hsrp"] = (check_cisco_hsrp, "HSRP Group %s", 0, inventory_cisco_hsrp) + +snmp_info["cisco_hsrp"] = \ + ( ".1.3.6.1.4.1.9.9.106.1.2.1.1", [ + OID_END, + "11", # cHsrpGrpVirtualIpAddr + "13", # cHsrpGrpActiveRouter + "14", # cHsrpGrpStandbyRouter + "15", # cHsrpGrpStandbyState + "16", # cHsrpGrpVirtualMacAddr + ]) + +# We can't scan for the HSRP table since the entries are indexed +# based on information we dont have without fetching all of it +# Instead we use the HSRP timeout +snmp_scan_functions["cisco_hsrp"] = lambda oid: "cisco" in oid(".1.3.6.1.2.1.1.1.0").lower() and oid(".1.3.6.1.4.1.9.9.106.1.1.1.0")