Monitoring Port-Channels on Cisco Switches/Routers with Check_mk

Monitoring Cisco Switches with OMD (check_mk included) is easier than ever. After the first inventory WATO offered me hundreds of checks – interfaces, modules, CPU utilization and temperatures! However, it wouldn´t show me Port-Channels. Port-Channels are aggregated physical interfaces which are usually used for inter-switch connectivity. As an example, you can combine 4 x 1 Gbit/s interfaces in order to connect two switches with 4 Gbit/s. Besides the higher bandwidth you also have a higher security since individual links can go down without having a full disconnect (it´s only affecting the bandwidth).

And here is what you have to-do in order to see Port-Channels in WATO as services (tried with check_mk 1.2.0p2 and a cisco 6500 Switch):

Go to global settings in the Check_mk webinterface –> click Network interface port types to inventorize in the first passage –> check propVirtual(53) –>  click save

Further reboots are not needed. After the next inventory you should see Port-Channels!

It took me forever to find out which of the 228 options enables Port-Channes, and I still don´t know why it’s called propVirtual – probably it’s short for proprietary Virtual Channel?!

04/29/13: After updating to check_mk 1.2.2 it wouldn´t recognize new port-channels. The solution seems to enable ieee8023adLag(161) in the way described above! Dunno what that one is about…

Advertisements

About sitweak
Monitoring, Network, Firewall, Mobile Security. I´m totally into that stuff!

6 Responses to Monitoring Port-Channels on Cisco Switches/Routers with Check_mk

  1. marco says:

    TraduttoreDisattiva traduzione istantanea
    ItalianoIngleseFranceseRileva linguaIngleseItalianoSpagnoloTraduci

    Hi sitweak

    I’m testing check_mk for a couple of weeks and it seems very nice and easy to use, at least compared to other nagios solutions tested in the past
    And today I founded your post about port-channells that I have immediately applied in my test environmente. I really thank you for this nice tip!
    Now…I have a couple of questions about how to properly monitor switch/router interfaces

    1) How to monitor switch/router ports in general?
    I know that every environment has its own peculiarity and needs but, in my opinion, a “general” and good way is to monitor operational status ONLY for those interfaces that also are administratively up. In this way I have all performance graphs and I get a CRITICAL whenever a port is operational down (and this shouldn’t happen).
    What do you think about this? Do you agree?

    2) Check_mk – monitor ONLY interface administratively up
    To be honest, I haven’t understood if this is possibile or not.
    As per the documentation:

    (https://mathias-kettner.de/checkmk_check_if64.html)
    Depending on the check paramters this check can go WARN or CRIT when the port status changes (i.e. is down), when the link speed changes (e.g. a port expected to be set to 1GBit/s operates only at 100MBit/s), when the absolute or procentual traffic of a port exceeds certain levels or if the rate of errors or discards exceeds configurable limits.

    Using WATO, I tried to play with rules in this way:

    Host & Service Parameters, Parameteres for discovered services, Network interfaces and switch ports, Operational state, Allowed States:

    up
    down

    admin down

    but it seems that is not possible to apply my logic within check_mk.

    Do you have any suggestion please?

    bye
    Marco
    (Italy)

    Hi sitweak

    Sto testando check_mk per un paio di settimane e sembra molto bello e facile da usare, almeno rispetto ad altre soluzioni Nagios testate in passato
    E oggi ho fondato tuo post su Port-channells che ho subito applicate nel mio contesto ambientale di test. Ho davvero grazie per questa bella punta!
    Ora … Ho un paio di domande su come monitorare correttamente interruttore / interfacce del router

    1) Come monitorare interruttori / porte del router in generale?
    So che ogni ambiente ha la sua peculiarità e le esigenze, ma, a mio parere, un modo “generale” e buona è quello di monitorare lo stato operativo solo per quelle interfacce che sono anche amministrativamente in su. In questo modo ho tutti i grafici delle prestazioni e ottenere un CRITICO ogni volta che una porta è operativo verso il basso (e questo non dovrebbe accadere).
    Cosa ne pensi di questo? Sei d’accordo?

    2) Check_mk – monitorare solo l’interfaccia amministrativamente su
    Per essere onesti, non ho capito se questo è possibile o no.
    Come per la documentazione:

    (https://mathias-kettner.de/checkmk_check_if64.html)
    A seconda delle paramters controllo questo controllo può andare WARN o CRIT quando cambia stato della porta (cioè verso il basso), al variare della velocità di collegamento (ad esempio una porta dovrebbe essere impostato a 1GBit / s funziona solo a 100 Mbit / s), quando il traffico assoluta o procentual di un porto supera determinati livelli o se il tasso di errori o scarti supera i limiti configurabili.

    Utilizzando WATO, ho provato a giocare con le regole in questo modo:

    Host & Parametri di servizio, parameteres per i servizi scoperti, interfacce di rete e porte switch, Stato operativo, Stati ammessi:

    su
    giù

    Admin giù

    ma sembra che non sia possibile applicare la mia logica all’interno check_mk.

    Avete qualche suggerimento per favore?

    addio
    Marco
    (Italia)
    Suggerisci una modifica
    Google Traduttore per il Business:Translator ToolkitTraduttore di siti webStrumento a supporto dell’export
    Informazioni su Google TraduttoreCommunityPer cellulariTutto su GooglePrivacy e TerminiGuidaInvia commenti

    • sitweak says:

      Hello Marco,
      Thanks for your input.

      Regarding 1) Yes, I usually only monitored ports that are administratively up. This is especially important for ports which are in a port-channel/LAG as this link will stay in state ”up” as long as one port is still functional. Personally, I would monitor the port-channel itself for bandwidth and the underlying ports for state. This will help decrease redundancy in information, bandwidth and complexity (vs. a scenario where you monitor bandwidth/state on each port separately AND on the port-channel).
      Regarding 2) I haven’t worked with check_mk in the last year. I might find some time in the next days to test this out and then I will get back to you. Per my memory: I always added the ports via inventory when they were up. As a result I would receive and CRIT if the port status changed (see “The default is to remember the state found during inventory and enforce this. I” @ https://mathias-kettner.de/checkmk_check_if64.html). Are you trying to add ports which are currently down?

      Cheers, Heiko

      • mlist@libero.it says:

        Hi Heiko,

        before all thank you for your kind reply and I apologize for the “dirty text” (Traduttore…), I did a bad copy and paste from my google translator 🙂

        About point number 1, I’m happy to hear that; this is excatly what I always thought about it. Instead, about point number 2, you are also right, check_mk per default remember the state found during first inventory and a CRIT is generated if the port status changes. But, as you said, this is not the best way to monitor switches port status (at least not always); the problem is that, in my understanding, it seems not possibile to generate alerts ONLY for those ports that are administratively up. I have already posted the same question in the mailing-list but no one replied to me about this topic (instead I have recevied excellent replies to other questions). I just found a question “similar” to mine at the following link but it is not clear to me: http://lists.mathias-kettner.de/pipermail/checkmk-en/2014-July/012717.html

        About your question, the answer is yes, I just added a switch that have some ports administratively up and I know that if I will change the configuration (no shut), an alert will be created but…this is not what I need. On the following link, an image of my switch: http://imgur.com/yBD0gAE

        Apart that, can I ask you why did you stop using check_mk? Did you find something better? I would like to have your opinion about check_mk because now I’m using the raw edition in a test environment and, to be honest, it seems really great compared to other Nagios solutions that I tried in the past. I’m going to propose the enterprise version (that has support) to my boss. Do you think am I in the right direction or am I going to suicide? 🙂

      • sitweak says:

        Marco, I relocated to a different country. Check_mk is not well known here and I am now working in an environment where monitoring is done with Cacti and PRTG. Personally speaking, I still prefer checkl_mk over any other nagios based solution and I came from the same situation: I tested check_mk in a test environment and later purchased the enterprise license. I was able to talk to Matthias a couple of times and he is so knowledgeable and you can tell he knows every single bit in this software. That being said, you can expect an outstanding support from them once you purchase the license. Regarding your question: Unfortunately, I can’t give you any better advice. Actually, my advice would have been to ask in the mailing list 😉 I am surprised noone answered your question there.

      • marco says:

        Heiko

        I really thank you because apart the “problem” with the switches port monitoring, what I really needed, was an opinion about check_mk in general (product and support) and you this is what you did. I’m sure that once I will have purchased the licence, they will help me with this topic. Yes, really strange that noone answered to me also because as I already said, I had excellent replies about other questions…

        Best
        Marco

  2. Mohamed says:

    Hello, I am mohamed currently on internship. I set up a network monitoring tool under check_mk. My monitoring server works fine, but I wanted to set up a visual system from nagvis to have a global view of all my equipment. (Image, ip address, name, state) and the list of all supervised services. I ask for help from all. Thank you

Leave a Reply

Fill in your details below or click an icon to log in:

WordPress.com Logo

You are commenting using your WordPress.com account. Log Out / Change )

Twitter picture

You are commenting using your Twitter account. Log Out / Change )

Facebook photo

You are commenting using your Facebook account. Log Out / Change )

Google+ photo

You are commenting using your Google+ account. Log Out / Change )

Connecting to %s

%d bloggers like this: