Xymon Mailing List Archive search

Custom Server Rules for analysis.cfg

9 messages in this thread

list Thomas Hawkins · Fri, 21 Dec 2012 09:52:01 -0330 ·
We have 50 or 60 servers configured across a number of pages in our xymon
instance.  Instead of having 60 host=hostname entries with generic entries
across all of the servers, I want to set up some groups/tags/etc... that
can be applied to a host.

So, I figured out I could do:
PAGE=DELTS/*/*
        #Defaults for all DELTS Servers
        PROC crond 1 -1 red "TEXT=crond : Task Scheduler"
        PROC ntpd 1 -1 yellow "TEXT=ntpd : Network Time Client"

To go across all of our servers on our subpages to apply these two proc
checks.

What I would like to do is get a bit more specific.  We have a mix of
physical machines, virtual machines, and linux machines, windows machines,
and production machines and development machines.

I'd like to be able to tag hosts somehow so I can do something like this:
Host 1 = Linux, Production, VM
Host 2 = Linux, Physical, Dev

Linux:
PROC - ntp, cron

Production:
PROC - backup client

VM:
PROC - vmwaretools

Physical:
nothing additional.

Dev:
nothing additional

So for Host1 it will do the following PROC tests:
ntp, cron, backup client, vmwaretools

Whereas Host2 will only do:
ntp, cron

I understand that if I had hosts configured into like groups, or subpages
then I could target the whole works that way, but I currently have my hosts
configured in a way that this is not practical.  Essentially I have all
hosts divided into 3 major groups, and then a number of subgroups, and then
the hosts are listed here based on application cluster or function.  Basing
it on groups or pages would also require me to create a bunch of additional
rules to account for each of the situations above, whereas something like a
tag that a host could have multiple of would simplify things.

Can anyone provide any insight as to how I can achieve these goals?

Thanks,
Thomas
list Michael Burger · Fri, 21 Dec 2012 09:38:19 -0500 ·
FWIW, I have a number of cases where I have hosts on multiple
(sub)pages. I don't use them for the analysis.cfg as much as for the
alerts.cfg, but in this way, I can have a page for team x's servers,
team y's servers, prod, non-prod, or all server.

 
--

Mike Burger

AIX Administrator

  <http://freedomhome.fhmc.local/intranet/main.jsp>; 

Phone (XXX) XXX-XXXX, Fax (XXX) XXX-XXXX, Cell (XXX) XXX-XXXX 

E-mail:  user-c26873f0522a@xymon.invalid

"Once word leaks out that a pirate's gone soft, people begin to disobey
you and then it's nothing but work, work, work, all the time."
--Westley/The Dread Pirate Roberts

 
From: xymon-bounces at xymon.com [mailto:xymon-bounces at xymon.com] On Behalf
Of Thomas Hawkins
Sent: Friday, December 21, 2012 8:22 AM
To: xymon at xymon.com
Subject: [Xymon] Custom Server Rules for analysis.cfg
quoted from Thomas Hawkins

 
We have 50 or 60 servers configured across a number of pages in our
xymon instance.  Instead of having 60 host=hostname entries with generic
entries across all of the servers, I want to set up some
groups/tags/etc... that can be applied to a host.

 
So, I figured out I could do:

PAGE=DELTS/*/*

        #Defaults for all DELTS Servers

        PROC crond 1 -1 red "TEXT=crond : Task Scheduler"

        PROC ntpd 1 -1 yellow "TEXT=ntpd : Network Time Client"

 
To go across all of our servers on our subpages to apply these two proc
checks.

 
What I would like to do is get a bit more specific.  We have a mix of
physical machines, virtual machines, and linux machines, windows
machines, and production machines and development machines.

 
I'd like to be able to tag hosts somehow so I can do something like
this:

Host 1 = Linux, Production, VM

Host 2 = Linux, Physical, Dev

 
Linux:

PROC - ntp, cron

 
Production:

PROC - backup client

 
VM:

PROC - vmwaretools

 
Physical:

nothing additional.

 
Dev:

nothing additional

 
So for Host1 it will do the following PROC tests:

ntp, cron, backup client, vmwaretools

 
Whereas Host2 will only do:

ntp, cron

 
I understand that if I had hosts configured into like groups, or
subpages then I could target the whole works that way, but I currently
have my hosts configured in a way that this is not practical.
Essentially I have all hosts divided into 3 major groups, and then a
number of subgroups, and then the hosts are listed here based on
application cluster or function.  Basing it on groups or pages would
also require me to create a bunch of additional rules to account for
each of the situations above, whereas something like a tag that a host
could have multiple of would simplify things.

 
Can anyone provide any insight as to how I can achieve these goals?

 
Thanks,

Thomas


CONFIDENTIALITY NOTICE: This e-mail message, including all attachments, is for
the sole use of the intended recipient(s) and may contain confidential
information. Any unauthorized review, use, disclosure, alteration or
distribution is strictly prohibited and may violate state or federal law. If you
are not the intended recipient, please contact the sender by reply e-mail,
delete this email and destroy all copies of the message.
list Thomas Hawkins · Fri, 21 Dec 2012 14:55:28 +0000 ·
Hummm, that is interesting.  You define the host multiple times in hosts.cfg?  Is that a supported setup?

How does your main homepage look?

We have three sub pages, and one sub-sub page.  I assume under this setup, you would need more subpages splintering off the main home page for this to work.  Although it would be nice at a glance to see all of the server grouped by my various 'tags' as well as function.

Thomas
quoted from Michael Burger
-----Original Message-----
From: "Michael Burger" <user-c26873f0522a@xymon.invalid>
Date: Fri, 21 Dec 2012 09:38:19 
To: Thomas Hawkins<user-5a1d8f143e48@xymon.invalid>; <xymon at xymon.com>
Subject: RE: [Xymon] Custom Server Rules for analysis.cfg

FWIW, I have a number of cases where I have hosts on multiple
(sub)pages. I don't use them for the analysis.cfg as much as for the
alerts.cfg, but in this way, I can have a page for team x's servers,
team y's servers, prod, non-prod, or all server.

 
--

Mike Burger

AIX Administrator

  <http://freedomhome.fhmc.local/intranet/main.jsp>; 

Phone (XXX) XXX-XXXX, Fax (XXX) XXX-XXXX, Cell (XXX) XXX-XXXX 

E-mail:  user-c26873f0522a@xymon.invalid

"Once word leaks out that a pirate's gone soft, people begin to disobey
you and then it's nothing but work, work, work, all the time."
--Westley/The Dread Pirate Roberts

 
From: xymon-bounces at xymon.com [mailto:xymon-bounces at xymon.com] On Behalf
Of Thomas Hawkins
Sent: Friday, December 21, 2012 8:22 AM
To: xymon at xymon.com
Subject: [Xymon] Custom Server Rules for analysis.cfg

 
We have 50 or 60 servers configured across a number of pages in our
xymon instance.  Instead of having 60 host=hostname entries with generic
entries across all of the servers, I want to set up some
groups/tags/etc... that can be applied to a host.

 
So, I figured out I could do:

PAGE=DELTS/*/*

        #Defaults for all DELTS Servers

        PROC crond 1 -1 red "TEXT=crond : Task Scheduler"

        PROC ntpd 1 -1 yellow "TEXT=ntpd : Network Time Client"

 
To go across all of our servers on our subpages to apply these two proc
checks.

 
What I would like to do is get a bit more specific.  We have a mix of
physical machines, virtual machines, and linux machines, windows
machines, and production machines and development machines.

 
I'd like to be able to tag hosts somehow so I can do something like
this:

Host 1 = Linux, Production, VM

Host 2 = Linux, Physical, Dev

 
Linux:

PROC - ntp, cron

 
Production:

PROC - backup client

 
VM:

PROC - vmwaretools

 
Physical:

nothing additional.

 
Dev:

nothing additional

 
So for Host1 it will do the following PROC tests:

ntp, cron, backup client, vmwaretools

 
Whereas Host2 will only do:

ntp, cron

 
I understand that if I had hosts configured into like groups, or
subpages then I could target the whole works that way, but I currently
have my hosts configured in a way that this is not practical.
Essentially I have all hosts divided into 3 major groups, and then a
number of subgroups, and then the hosts are listed here based on
application cluster or function.  Basing it on groups or pages would
also require me to create a bunch of additional rules to account for
each of the situations above, whereas something like a tag that a host
could have multiple of would simplify things.

 
Can anyone provide any insight as to how I can achieve these goals?

 
Thanks,

Thomas


CONFIDENTIALITY NOTICE: This e-mail message, including all attachments, is for
the sole use of the intended recipient(s) and may contain confidential
information. Any unauthorized review, use, disclosure, alteration or
distribution is strictly prohibited and may violate state or federal law. If you
are not the intended recipient, please contact the sender by reply e-mail,
delete this email and destroy all copies of the message.
list Michael Burger · Fri, 21 Dec 2012 10:17:13 -0500 ·
Yup...I define the hosts in different subpage groupings.

 
My home page lists strictly pages, each page lists the subpages, and the
subpages list the hosts contained within.

 
This would likely also allow me to have different (sub)pages list the
same hosts, but with different columns displayed. I haven't tried
monkeying around with that level of configuration, though.
quoted from Thomas Hawkins

--

Mike Burger

AIX Administrator

  <http://freedomhome.fhmc.local/intranet/main.jsp>; 

Phone (XXX) XXX-XXXX, Fax (XXX) XXX-XXXX, Cell (XXX) XXX-XXXX 

E-mail:  user-c26873f0522a@xymon.invalid

"Once word leaks out that a pirate's gone soft, people begin to disobey
you and then it's nothing but work, work, work, all the time."
--Westley/The Dread Pirate Roberts

 
From: Thomas Hawkins [mailto:user-5a1d8f143e48@xymon.invalid] 
Sent: Friday, December 21, 2012 9:55 AM
To: Michael Burger; xymon at xymon.com
Subject: Re: [Xymon] Custom Server Rules for analysis.cfg

 
Hummm, that is interesting. You define the host multiple times in
hosts.cfg? Is that a supported setup?

How does your main homepage look?

We have three sub pages, and one sub-sub page. I assume under this
setup, you would need more subpages splintering off the main home page
for this to work. Although it would be nice at a glance to see all of
the server grouped by my various 'tags' as well as function.

Thomas


From: "Michael Burger" <user-c26873f0522a@xymon.invalid> 

Date: Fri, 21 Dec 2012 09:38:19 -0500

To: Thomas Hawkins<user-5a1d8f143e48@xymon.invalid>; <xymon at xymon.com>

Subject: RE: [Xymon] Custom Server Rules for analysis.cfg

 
FWIW, I have a number of cases where I have hosts on multiple
(sub)pages. I don't use them for the analysis.cfg as much as for the
alerts.cfg, but in this way, I can have a page for team x's servers,
team y's servers, prod, non-prod, or all server.

 
--

Mike Burger

AIX Administrator

 <http://freedomhome.fhmc.local/intranet/main.jsp>; 

Phone (XXX) XXX-XXXX, Fax (XXX) XXX-XXXX, Cell (XXX) XXX-XXXX 

E-mail:  user-c26873f0522a@xymon.invalid

"Once word leaks out that a pirate's gone soft, people begin to disobey
you and then it's nothing but work, work, work, all the time."
--Westley/The Dread Pirate Roberts

 
From: xymon-bounces at xymon.com [mailto:xymon-bounces at xymon.com] On Behalf
Of Thomas Hawkins
Sent: Friday, December 21, 2012 8:22 AM
To: xymon at xymon.com
Subject: [Xymon] Custom Server Rules for analysis.cfg

 
We have 50 or 60 servers configured across a number of pages in our
xymon instance.  Instead of having 60 host=hostname entries with generic
entries across all of the servers, I want to set up some
groups/tags/etc... that can be applied to a host.

 
So, I figured out I could do:

PAGE=DELTS/*/*

        #Defaults for all DELTS Servers

        PROC crond 1 -1 red "TEXT=crond : Task Scheduler"

        PROC ntpd 1 -1 yellow "TEXT=ntpd : Network Time Client"

 
To go across all of our servers on our subpages to apply these two proc
checks.

 
What I would like to do is get a bit more specific.  We have a mix of
physical machines, virtual machines, and linux machines, windows
machines, and production machines and development machines.

 
I'd like to be able to tag hosts somehow so I can do something like
this:

Host 1 = Linux, Production, VM

Host 2 = Linux, Physical, Dev

 
Linux:

PROC - ntp, cron

 
Production:

PROC - backup client

 
VM:

PROC - vmwaretools

 
Physical:

nothing additional.

 
Dev:

nothing additional

 
So for Host1 it will do the following PROC tests:

ntp, cron, backup client, vmwaretools

 
Whereas Host2 will only do:

ntp, cron

 
I understand that if I had hosts configured into like groups, or
subpages then I could target the whole works that way, but I currently
have my hosts configured in a way that this is not practical.
Essentially I have all hosts divided into 3 major groups, and then a
number of subgroups, and then the hosts are listed here based on
application cluster or function.  Basing it on groups or pages would
also require me to create a bunch of additional rules to account for
each of the situations above, whereas something like a tag that a host
could have multiple of would simplify things.

 
Can anyone provide any insight as to how I can achieve these goals?

 
Thanks,

Thomas

 
CONFIDENTIALITY NOTICE: This e-mail message, including all attachments,
is for
the sole use of the intended recipient(s) and may contain confidential
information. Any unauthorized review, use, disclosure, alteration or
distribution is strictly prohibited and may violate state or federal
law. If you
are not the intended recipient, please contact the sender by reply
e-mail,
delete this email and destroy all copies of the message.


CONFIDENTIALITY NOTICE: This e-mail message, including all attachments, is for
the sole use of the intended recipient(s) and may contain confidential
information. Any unauthorized review, use, disclosure, alteration or
distribution is strictly prohibited and may violate state or federal law. If you
are not the intended recipient, please contact the sender by reply e-mail,
delete this email and destroy all copies of the message.
list Japheth Cleaver · Fri, 21 Dec 2012 10:19:08 -0800 (PST) ·
I generate my hosts.cfg in separate include files. One set includes the
canonical list of all tests for each host, but uses netinclude so it's not
displayed. The other set is just lots of different page-subparent
configurations for different groups (by DC, by customer, by responsible
party, etc..)

-jc
quoted from Michael Burger

Yup...I define the hosts in different subpage groupings.


My home page lists strictly pages, each page lists the subpages, and the
subpages list the hosts contained within.


This would likely also allow me to have different (sub)pages list the
same hosts, but with different columns displayed. I haven't tried
monkeying around with that level of configuration, though.

--

Mike Burger

AIX Administrator

  <http://freedomhome.fhmc.local/intranet/main.jsp>;

Phone (XXX) XXX-XXXX, Fax (XXX) XXX-XXXX, Cell (XXX) XXX-XXXX

E-mail:  user-c26873f0522a@xymon.invalid

"Once word leaks out that a pirate's gone soft, people begin to disobey
you and then it's nothing but work, work, work, all the time."
--Westley/The Dread Pirate Roberts


From: Thomas Hawkins [mailto:user-5a1d8f143e48@xymon.invalid]
Sent: Friday, December 21, 2012 9:55 AM
To: Michael Burger; xymon at xymon.com
Subject: Re: [Xymon] Custom Server Rules for analysis.cfg


Hummm, that is interesting. You define the host multiple times in
hosts.cfg? Is that a supported setup?

How does your main homepage look?

We have three sub pages, and one sub-sub page. I assume under this
setup, you would need more subpages splintering off the main home page
for this to work. Although it would be nice at a glance to see all of
the server grouped by my various 'tags' as well as function.

Thomas


From: "Michael Burger" <user-c26873f0522a@xymon.invalid>

Date: Fri, 21 Dec 2012 09:38:19 -0500

To: Thomas Hawkins<user-5a1d8f143e48@xymon.invalid>; <xymon at xymon.com>

Subject: RE: [Xymon] Custom Server Rules for analysis.cfg


FWIW, I have a number of cases where I have hosts on multiple
(sub)pages. I don't use them for the analysis.cfg as much as for the
alerts.cfg, but in this way, I can have a page for team x's servers,
team y's servers, prod, non-prod, or all server.


--

Mike Burger

AIX Administrator

 <http://freedomhome.fhmc.local/intranet/main.jsp>;

Phone (XXX) XXX-XXXX, Fax (XXX) XXX-XXXX, Cell (XXX) XXX-XXXX

E-mail:  user-c26873f0522a@xymon.invalid

"Once word leaks out that a pirate's gone soft, people begin to disobey
you and then it's nothing but work, work, work, all the time."
--Westley/The Dread Pirate Roberts


From: xymon-bounces at xymon.com [mailto:xymon-bounces at xymon.com] On Behalf
Of Thomas Hawkins
Sent: Friday, December 21, 2012 8:22 AM
To: xymon at xymon.com
Subject: [Xymon] Custom Server Rules for analysis.cfg


We have 50 or 60 servers configured across a number of pages in our
xymon instance.  Instead of having 60 host=hostname entries with generic
entries across all of the servers, I want to set up some
groups/tags/etc... that can be applied to a host.


So, I figured out I could do:

PAGE=DELTS/*/*

        #Defaults for all DELTS Servers

        PROC crond 1 -1 red "TEXT=crond : Task Scheduler"

        PROC ntpd 1 -1 yellow "TEXT=ntpd : Network Time Client"


To go across all of our servers on our subpages to apply these two proc
checks.


What I would like to do is get a bit more specific.  We have a mix of
physical machines, virtual machines, and linux machines, windows
machines, and production machines and development machines.


I'd like to be able to tag hosts somehow so I can do something like
this:

Host 1 = Linux, Production, VM

Host 2 = Linux, Physical, Dev


Linux:

PROC - ntp, cron


Production:

PROC - backup client


VM:

PROC - vmwaretools


Physical:

nothing additional.


Dev:

nothing additional


So for Host1 it will do the following PROC tests:

ntp, cron, backup client, vmwaretools


Whereas Host2 will only do:

ntp, cron


I understand that if I had hosts configured into like groups, or
subpages then I could target the whole works that way, but I currently
have my hosts configured in a way that this is not practical.
Essentially I have all hosts divided into 3 major groups, and then a
number of subgroups, and then the hosts are listed here based on
application cluster or function.  Basing it on groups or pages would
also require me to create a bunch of additional rules to account for
each of the situations above, whereas something like a tag that a host
could have multiple of would simplify things.


Can anyone provide any insight as to how I can achieve these goals?


Thanks,

Thomas


CONFIDENTIALITY NOTICE: This e-mail message, including all attachments,
is for
the sole use of the intended recipient(s) and may contain confidential
information. Any unauthorized review, use, disclosure, alteration or
distribution is strictly prohibited and may violate state or federal
law. If you
are not the intended recipient, please contact the sender by reply
e-mail,
delete this email and destroy all copies of the message.


CONFIDENTIALITY NOTICE: This e-mail message, including all attachments, is
for
the sole use of the intended recipient(s) and may contain confidential
information. Any unauthorized review, use, disclosure, alteration or
distribution is strictly prohibited and may violate state or federal law.
If you
are not the intended recipient, please contact the sender by reply e-mail,
delete this email and destroy all copies of the
message.
list Thomas Hawkins · Sat, 22 Dec 2012 00:41:35 -0330 ·
Ok, I thought I had this working, but not quite.  I'm noticing that only
one PROC is ever evaluated from my groups in addition to any custom procs
listed for the specific server.  Here is an example:
Host: tee which is defined in 4 places:
DELTS/DB
DELTS2/LINUX
DELTS2/L-VM
DELTS2/T-L-PROD

analysis.cfg
PAGE=DELTS2/LINUX
        PROC cron 1 -1 red "TEXT=crond : Task Scheduler"
        PROC ntpd 1 -1 yellow "TEXT=ntpd : Network Time Client"

PAGE=DELTS2/L-VM
        PROC vmware-guestd 1 -1 yellow "TEXT=VMWare Tools"

HOST=tee
        PROC mysqld 2 -1 red

But when I look at tee, only two procs are monitored, cron and mysqld.  If
I comment out cron, the ntpd is monitored along with mysqld.  This seems to
happen across all servers.  If there are additional procs under the
individual HOST entry, they are also monitored successfully.

I did a bit of testing with ./xymond_client --test --debug and I seem to
get the same result, but I can't figure out why it will only evaluate one
PROC test from my lists.

The only other change I made that may be significant is that I configured a
Windows server to use central mode so that I can start monitoring services
on Windows machines.  I had to make some of the Xymon message variables a
bit bigger to accomplish this.  Not sure if it is related though.

Any help would be appreciated.

Thanks,
Thomas


On Fri, Dec 21, 2012 at 11:47 AM, Michael Burger <
quoted from Japheth Cleaver
user-c26873f0522a@xymon.invalid> wrote:
Yup…I define the hosts in different subpage groupings.****

** **

My home page lists strictly pages, each page lists the subpages, and the
subpages list the hosts contained within.****

** **

This would likely also allow me to have different (sub)pages list the same
hosts, but with different columns displayed. I haven’t tried monkeying
around with that level of configuration, though.****

--****

*Mike Burger*****

*AIX Administrator*

*[image: cid:image001.gif at 01CC72C8.834C1990]*<http://freedomhome.fhmc.local/intranet/main.jsp>;
****

*Phone (XXX) XXX-XXXX**, **Fax** **(XXX) XXX-XXXX, Cell (XXX) XXX-XXXX ***

*E-mail:  **user-c26873f0522a@xymon.invalid*<user-9fdcca966073@xymon.invalid>
quoted from Japheth Cleaver
****

"Once word leaks out that a pirate's gone soft, people begin to disobey
you and then it's nothing but work, work, work, all the time."
--Westley/The Dread Pirate Roberts****

** **

*From:* Thomas Hawkins [mailto:user-5a1d8f143e48@xymon.invalid]
*Sent:* Friday, December 21, 2012 9:55 AM
*To:* Michael Burger; xymon at xymon.com
*Subject:* Re: [Xymon] Custom Server Rules for analysis.cfg****

** **

Hummm, that is interesting. You define the host multiple times in
hosts.cfg? Is that a supported setup?

How does your main homepage look?

We have three sub pages, and one sub-sub page. I assume under this setup,
you would need more subpages splintering off the main home page for this to
work. Although it would be nice at a glance to see all of the server
grouped by my various 'tags' as well as function.

Thomas****

*From: *"Michael Burger" <user-c26873f0522a@xymon.invalid> ****

*Date: *Fri, 21 Dec 2012 09:38:19 -0500****

*To: *Thomas Hawkins<user-5a1d8f143e48@xymon.invalid>; <xymon at xymon.com>****

*Subject: *RE: [Xymon] Custom Server Rules for analysis.cfg****

** **

FWIW, I have a number of cases where I have hosts on multiple (sub)pages.
I don’t use them for the analysis.cfg as much as for the alerts.cfg, but in
this way, I can have a page for team x’s servers, team y’s servers, prod,
non-prod, or all server.****

** **

--****

*Mike Burger*****

*AIX Administrator*

*[image: cid:image001.gif at 01CC72C8.834C1990]*<http://freedomhome.fhmc.local/intranet/main.jsp>;
****

*Phone (XXX) XXX-XXXX**, **Fax** **(XXX) XXX-XXXX, Cell (XXX) XXX-XXXX ***

*E-mail:  **user-c26873f0522a@xymon.invalid*<user-9fdcca966073@xymon.invalid>
quoted from Japheth Cleaver
****

"Once word leaks out that a pirate's gone soft, people begin to disobey
you and then it's nothing but work, work, work, all the time."
--Westley/The Dread Pirate Roberts****

** **

*From:* xymon-bounces at xymon.com [mailto:xymon-bounces at xymon.com<xymon-bounces at xymon.com>]
quoted from Japheth Cleaver
*On Behalf Of *Thomas Hawkins
*Sent:* Friday, December 21, 2012 8:22 AM
*To:* xymon at xymon.com
*Subject:* [Xymon] Custom Server Rules for analysis.cfg****

** **

We have 50 or 60 servers configured across a number of pages in our xymon
instance.  Instead of having 60 host=hostname entries with generic entries
across all of the servers, I want to set up some groups/tags/etc... that
can be applied to a host.****

** **

So, I figured out I could do:****

PAGE=DELTS/*/*****

        #Defaults for all DELTS Servers****

        PROC crond 1 -1 red "TEXT=crond : Task Scheduler"****

        PROC ntpd 1 -1 yellow "TEXT=ntpd : Network Time Client"****

** **

To go across all of our servers on our subpages to apply these two proc
checks.****

** **

What I would like to do is get a bit more specific.  We have a mix of
physical machines, virtual machines, and linux machines, windows machines,
and production machines and development machines.****

** **

I'd like to be able to tag hosts somehow so I can do something like this:*
***

Host 1 = Linux, Production, VM****

Host 2 = Linux, Physical, Dev****

** **

Linux:****

PROC - ntp, cron****

** **

Production:****

PROC - backup client****

** **

VM:****

PROC - vmwaretools****

** **

Physical:****

nothing additional.****

** **

Dev:****

nothing additional****

** **

So for Host1 it will do the following PROC tests:****

ntp, cron, backup client, vmwaretools****

** **

Whereas Host2 will only do:****

ntp, cron****

** **

I understand that if I had hosts configured into like groups, or subpages
then I could target the whole works that way, but I currently have my hosts
configured in a way that this is not practical.  Essentially I have all
hosts divided into 3 major groups, and then a number of subgroups, and then
the hosts are listed here based on application cluster or function.  Basing
it on groups or pages would also require me to create a bunch of additional
rules to account for each of the situations above, whereas something like a
tag that a host could have multiple of would simplify things.****

** **

Can anyone provide any insight as to how I can achieve these goals?****

** **

Thanks,****

Thomas****

** **

** **

CONFIDENTIALITY NOTICE: This e-mail message, including all attachments, is for****

the sole use of the intended recipient(s) and may contain confidential****

information. Any unauthorized review, use, disclosure, alteration or****

distribution is strictly prohibited and may violate state or federal law. If you****

are not the intended recipient, please contact the sender by reply e-mail,****

delete this email and destroy all copies of the message.****


CONFIDENTIALITY NOTICE: This e-mail message, including all attachments, is for
the sole use of the intended recipient(s) and may contain confidential
information. Any unauthorized review, use, disclosure, alteration or
distribution is strictly prohibited and may violate state or federal law. If you
are not the intended recipient, please contact the sender by reply e-mail,
delete this email and destroy all copies of the message.

list Andrey Chervonets · Sun, 23 Dec 2012 18:10:28 +0200 ·
Try to assign hosts to variables like:

$GROUP_PROD=host1,host2, host5
$GROUP_TEST=host3,host6....

$GROUP_ALL=$GROUP_PROD,$GROUP_TEST
etc.

and then define rules using such variables instead of real hostnames or 
reg.expressions

This is working for alerts, may be it is working for analysis as well. 
Not checked yet.


Best regards,

Andrey Chervonets
CoMinder Ltd.
http://www.cominder.eu/


On 22.12.2012 13:00, xymon-request at xymon.com wrote:
Re: Custom Server Rules for analysis.cfg
list Thomas Hawkins · Sun, 23 Dec 2012 21:47:27 -0330 ·
Thanks for the suggestion.  It didn't seem to work at all though.

I think I have messed with the file so much now I can't remember what sort
of worked and what didn't work at all.

I think I'm still stuck with the problem where only custom page declaration
gets picked up, and only one of the directives in the group, plus anything
in the specific host group.

So for this:
quoted from Thomas Hawkins
PAGE=DELTS2/LINUX
        PROC cron 1 -1 red "TEXT=crond : Task Scheduler"
        PROC ntpd 1 -1 yellow "TEXT=ntpd : Network Time Client"

PAGE=DELTS2/L-VM
        PROC vmware-guestd 1 -1 yellow "TEXT=VMWare Tools"

HOST=tee
        PROC mysqld 2 -1 red

tee reports only for cron and mysql.  If I comment out cron, then I get
ntpd and mysql.  If I move the vmware rule to the top of the list, I get
vmware and mysql proc checks.  So it seems that only one test for a page
specification gets reported on, and only the first one at the top of the
page.

At this point I should probably point out that I have 4.3.10 server, but
for the most part 4.2.3 (I think) clients on linux, and bbwin .12 or .13 on
windows.  However I don't think that will matter much.

Thomas

On Sun, Dec 23, 2012 at 12:40 PM, Andrey Chervonets <
quoted from Andrey Chervonets
user-e7fb5c02322c@xymon.invalid> wrote:
Try to assign hosts to variables like:

$GROUP_PROD=host1,host2, host5
$GROUP_TEST=host3,host6....

$GROUP_ALL=$GROUP_PROD,$GROUP_**TEST
etc.

and then define rules using such variables instead of real hostnames or
reg.expressions

This is working for alerts, may be it is working for analysis as well. Not
checked yet.


Best regards,

Andrey Chervonets
CoMinder Ltd.
http://www.cominder.eu/


On 22.12.2012 13:00, xymon-request at xymon.com wrote:
Re: Custom Server Rules for analysis.cfg
______________________________**

Xymon at xymon.com<
list Thomas Hawkins · Fri, 28 Dec 2012 09:06:27 -0330 ·
I decided to try something else.  I decided to create a display group for
each type of server, and then specify host checks specific to that group.
 For example...from analysis.cfg:

PAGE=DELTS2/LINUX
        PROC cron 1 -1 red
        PROC ntpd 1 -1 yellow

DISPLAYGROUP="Linux External Production Servers - Virtual"
        PROC vmware-guestd 1 -1 yellow
        PROC /opt/tivoli/tsm/client/ba/bin/dsmc 1 -1 red
        PROC /opt/tivoli/tsm/client/ba/bin/dsmcad 1 -1 red

DISPLAYGROUP="Linux Internal Production Servers - Virtual"
        PROC vmware-guestd 1 -1 yellow
        PROC /opt/tivoli/tsm/client/ba/bin/dsmc 1 -1 red
        PROC /opt/tivoli/tsm/client/ba/bin/dsmcad 1 -1 red

DISPLAYGROUP="Linux Internal Production Servers - Physical"
        PROC /opt/tivoli/tsm/client/ba/bin/dsmc 1 -1 red
        PROC /opt/tivoli/tsm/client/ba/bin/dsmcad 1 -1 red

DISPLAYGROUP="Linux External Production Servers - Physical"
        PROC /opt/tivoli/tsm/client/ba/bin/dsmc 1 -1 red
        PROC /opt/tivoli/tsm/client/ba/bin/dsmcad 1 -1 red

PAGE=DELTS2/WIN
        SVC MpsSvc status=started startup=automatic

DISPLAYGROUP="Windows External Production Servers - Virtual"
        SVC VMTools status=started startup=automatic

DISPLAYGROUP="Windows Internal Production Servers - Virtual"
        SVC VMTools status=started startup=automatic

And from hosts.cfg:
page DELTS2 DELTS Servers - By Type
subpage LINUX Linux Servers
group Linux External Production Servers - Physical

group Linux External Production Servers - Virtual

group Linux Internal Production Servers - Physical

.... and so on ...

I confirmed that each server only belongs to one group, and that the group
name is unique across xymon.

For tee, I'm now getting:
 cron (found 1, req. 1 or more)
 vmware-guestd (found 1, req. 1 or more)
 /opt/tivoli/tsm/client/ba/bin/dsmc (found 1, req. 1 or more)
 /opt/tivoli/tsm/client/ba/bin/dsmcad (found 1, req. 1 or more)
 vmware-guestd (found 1, req. 1 or more)
 /opt/tivoli/tsm/client/ba/bin/dsmc (found 1, req. 1 or more)
 /opt/tivoli/tsm/client/ba/bin/dsmcad (found 1, req. 1 or more)
 /opt/tivoli/tsm/client/ba/bin/dsmc (found 1, req. 1 or more)
 /opt/tivoli/tsm/client/ba/bin/dsmcad (found 1, req. 1 or more)
 /opt/tivoli/tsm/client/ba/bin/dsmc (found 1, req. 1 or more)
 /opt/tivoli/tsm/client/ba/bin/dsmcad (found 1, req. 1 or more)
 mysqld (found 2, req. 2 or more)

So for the page declaration, I'm still only getting the first proc
definition (ntpd is missing) and then I'm getting VMWare twice, and our
backup client 4 times, even though it should only match once.  Then I'm
getting the mysqld test, which is specified in a HOST=tee entry.

I've also noticed that these linux tests are now being applied against all
linux and windows hosts in Xymon.  So I'm clearly abusing or
misunderstanding the DISPLAYGROUP functionality.

Any advice?
quoted from Thomas Hawkins


On Sun, Dec 23, 2012 at 9:47 PM, Thomas Hawkins <user-5a1d8f143e48@xymon.invalid>wrote:
Thanks for the suggestion.  It didn't seem to work at all though.

I think I have messed with the file so much now I can't remember what sort
of worked and what didn't work at all.

I think I'm still stuck with the problem where only custom page
declaration gets picked up, and only one of the directives in the group,
plus anything in the specific host group.

So for this:
PAGE=DELTS2/LINUX
        PROC cron 1 -1 red "TEXT=crond : Task Scheduler"
        PROC ntpd 1 -1 yellow "TEXT=ntpd : Network Time Client"

PAGE=DELTS2/L-VM
        PROC vmware-guestd 1 -1 yellow "TEXT=VMWare Tools"

HOST=tee
        PROC mysqld 2 -1 red

tee reports only for cron and mysql.  If I comment out cron, then I get
ntpd and mysql.  If I move the vmware rule to the top of the list, I get
vmware and mysql proc checks.  So it seems that only one test for a page
specification gets reported on, and only the first one at the top of the
page.

At this point I should probably point out that I have 4.3.10 server, but
for the most part 4.2.3 (I think) clients on linux, and bbwin .12 or .13 on
windows.  However I don't think that will matter much.

Thomas

On Sun, Dec 23, 2012 at 12:40 PM, Andrey Chervonets <
user-e7fb5c02322c@xymon.invalid> wrote:
Try to assign hosts to variables like:

$GROUP_PROD=host1,host2, host5
$GROUP_TEST=host3,host6....

$GROUP_ALL=$GROUP_PROD,$GROUP_**TEST
etc.

and then define rules using such variables instead of real hostnames or
reg.expressions

This is working for alerts, may be it is working for analysis as well.
Not checked yet.


Best regards,

Andrey Chervonets
CoMinder Ltd.
http://www.cominder.eu/


On 22.12.2012 13:00, xymon-request at xymon.com wrote:
Re: Custom Server Rules for analysis.cfg
______________________________**
Xymon at xymon.com<