circular parent/child chain

Lucy Pelzer lucy at pelzeronline.de
Mon Nov 19 19:57:51 CET 2012


-----BEGIN PGP SIGNED MESSAGE-----
Hash: SHA1

Hi List,

thanks for your fast replies, but I think you missed my point. I know
that I could configure one parent and alert about spanning tree or do a
lot of other stuff, but this would not help me at all in the following
situations.

Lets say that I have a lot of important infrastructure behind switch-4
and switch-5 and that the "normal" stp way is to route through switch-2
for switch-5 and for switch-4 to switch-3. Lets also suggest that at
switch-2 there is not that much important infrastructure. If now
switch-2 is failing at the same time a really important infrastructure
behind switch-5 fails. I will only get informed that switch-2 failed,
because switch-5 is a child of switch-2, but I will not be informed
about the more important stuff at switch-5. Now I will spend a lot of
time to exchange switch-2 before I even find out that there is something
really important at switch-5 going on only because I could not give it a
second parent (switch-4). Maybe the other problem would have been fixed
in minutes if I would have known about it. I think it is a bad idea to
have such parent/child chains. And please keep in mind that this is just
a simple example and that a network can be much bigger than just 5 switches.

The other point is that I will never see the real network infrastructure
in the statusmap. As far as I can read the documentation I should also
be able to build up my whole infrastructure like it is in the statusmap.
Or for what is it at all?

I would be really pleased if somebody could answer my question.

- - Why is it in nagios not possible to be parent/child of a other host at
the same time?
- - Why is it a deadlock and where is the problem?


Greetings
Lucy

On 17.11.2012 21:44, Giles Coochey wrote:
> On 17/11/2012 17:08, Lucy Pelzer wrote:
> Hi List,
> 
> I am in charge for a big network infrastructure and at the moment we
> evaluate solutions for our nagios monitoring with configuration via
> monarch. I really would like to create the new solution with nagios and
> puppet. There is only one big problem. In our infrastructure there are a
> lot of switches connected together like this:
> 
>                ------------
>                | switch-1 |
>                ------------
>                  /       \
>                 /         \
>                /           \
>        ------------     ------------
>        | switch-2 |-----| switch-3 |
>        ------------     ------------
>             |                 |
>             |                 |
>        ------------     ------------
>        | switch-5 |-----| switch-4 |
>        ------------     ------------
>> Hi Lucy,
> 
>> Although you have redundant physical interconnects for your switches,
>> Spanning Tree, or Rapid Spanning Tree or even Multiple Spanning Tree
>> protocols are taking effect and (probably) setting switch-1 as the Root
>> with the other switches set to a higher spanning tree priority value -
>> in any case, at layer two, during normal operation you don't have
>> multiple active parents in the layer-2 infrastructure.
> 
>> While you have redundancy, physically for your infrastructure, you are
>> only using those links (for each VLAN at least), in a non-redundant way.
> 
>> Setup your Nagios child / parent relationships as per spanning tree has
>> worked out how it should be when everything is up, add some monitoring
>> for spanning-tree within Nagios, that way everything will work, and
>> you'll always be informed when something is wrong.
> 
>> Thanks
> 
>> Giles
> 
> 
> All the time I try to configure this switch connections in nagios with
> more than one parent, where the hosts are at the same time parent and
> child, I get the following error:
> 
> ERROR: The host 'switch-2' is part of a circular parent/child chain!
> ...
> 
> I know by now that this is because nagios has to be hierarchically, but
> I do not understand why. Maybe somebody could tell me exactly why I
> can’t configure this in nagios or better could tell me how I can do it.
> I only found the explanation that this could conduct a deadlock. Maybe I
> am wrong, but is this not a likely network infrastructure? At last it is
> no problem to build something like this with STP running in your network.
> 
> 
> Great Thanks in advanced
> 
> Lucy
>>
>> ------------------------------------------------------------------------------
>>
>> Monitor your physical, virtual and cloud infrastructure from a single
>> web console. Get in-depth insight into apps, servers, databases, vmware,
>> SAP, cloud infrastructure, etc. Download 30-day Free Trial.
>> Pricing starts from $795 for 25 servers or applications!
>> http://p.sf.net/sfu/zoho_dev2dev_nov
>> _______________________________________________
>> Nagios-users mailing list
>> Nagios-users at lists.sourceforge.net
>> https://lists.sourceforge.net/lists/listinfo/nagios-users
>> ::: Please include Nagios version, plugin version (-v) and OS when
>> reporting any issue.
>> ::: Messages without supporting info will risk being sent to /dev/null
> 
> 
> 
> 
> ------------------------------------------------------------------------------
> Monitor your physical, virtual and cloud infrastructure from a single
> web console. Get in-depth insight into apps, servers, databases, vmware,
> SAP, cloud infrastructure, etc. Download 30-day Free Trial.
> Pricing starts from $795 for 25 servers or applications!
> http://p.sf.net/sfu/zoho_dev2dev_nov
> 
> 
> 
> _______________________________________________
> Nagios-users mailing list
> Nagios-users at lists.sourceforge.net
> https://lists.sourceforge.net/lists/listinfo/nagios-users
> ::: Please include Nagios version, plugin version (-v) and OS when reporting any issue. 
> ::: Messages without supporting info will risk being sent to /dev/null
> 

-----BEGIN PGP SIGNATURE-----
Version: GnuPG v2.0.19 (GNU/Linux)
Comment: Using GnuPG with undefined - http://www.enigmail.net/

iQEcBAEBAgAGBQJQqoEvAAoJEBSbfrVvipeepxAIALhMBXkAgTKKtfLJaknRQ7hx
QH6zuV2CbU0pUHbNrdtGhdT+GvoLV8zp0+aSinFxIhNDGbc13xEJQ0zzLA4t4Xjj
9SH8HvK7KHYYUvLAuW3Vxaq69sruEZW6J22GRB7tzJk02a8IqZjEUGa1ygwSC1NS
KuWiji7JRS/p1D7M8QI+bBJs0SYpAWHAmfCGno9JJtJg80p/0SMwvioy+trolg2E
QUQbWN+XWolgRR9YD75dJHugjD2RJysdbs/6hOzYnMKkpMi11IS9OvLGmWzdHpCh
ENhcsN+2O290A39Bvt8fbjavUt6Jl2o88EXhzTtNaTFgViSNXvGTh64bCkHSy58=
=Ka95
-----END PGP SIGNATURE-----

------------------------------------------------------------------------------
Monitor your physical, virtual and cloud infrastructure from a single
web console. Get in-depth insight into apps, servers, databases, vmware,
SAP, cloud infrastructure, etc. Download 30-day Free Trial.
Pricing starts from $795 for 25 servers or applications!
http://p.sf.net/sfu/zoho_dev2dev_nov
_______________________________________________
Nagios-users mailing list
Nagios-users at lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/nagios-users
::: Please include Nagios version, plugin version (-v) and OS when reporting any issue. 
::: Messages without supporting info will risk being sent to /dev/null





More information about the Users mailing list