From Joel at smartz.com  Thu Dec  9 15:55:26 2004
From: Joel at smartz.com (Joel Gray)
Date: Thu Dec  9 15:55:36 2004
Subject: Navigation links disappeared
Message-ID: <FCFB8D38040DC54BBED7F8A4A8F94EBC115090@mission.smartz.com>

Hello all,
 
I recently (this morning) upgraded from version 3.2 to 3.3.  The upgrade
went ok as far as I can tell, however the links on the pages are no
longer there.  When I go to the Top page, no links are available on the
left.  I do have all of my monitor groups and I can drill down into
them.  When doing that, the only link I have on the left is the "Top"
link.
 
Looking at the rendered HTML, I noticed that all of the missing links as
well as some other missing titles for columns being out put as <a
href="...."><L10N Display Config></a>.  What was the <L10N> tag for?  I
removed those tags from the necessary .pm files in /usr/local/lib/argus
(backing up the originals of course) but I am not sure what their
purpose was.
 
Can anyone tell me if removing those tags will cause me problems?
 
Thanks in advance
-Joel
 
From jaw+arguslist at tcp4me.com  Thu Dec  9 18:46:55 2004
From: jaw+arguslist at tcp4me.com (Jeff Weisberg)
Date: Thu Dec  9 18:47:01 2004
Subject: Navigation links disappeared
Message-ID: <200412092346.iB9Nkt214667@penelope.tcp4me.com>


| I recently (this morning) upgraded from version 3.2 to 3.3.  The upgrade
| went ok as far as I can tell, however the links on the pages are no
| longer there.  When I go to the Top page, no links are available on the
| left.  I do have all of my monitor groups and I can drill down into
| them.  When doing that, the only link I have on the left is the "Top"
| link.
| 
| Looking at the rendered HTML, I noticed that all of the missing links as
| well as some other missing titles for columns being out put as <a
| href="...."><L10N Display Config></a>.  What was the <L10N> tag for?  I

they are used to provide i18n/l10n support, and are processed
by arguscgi.


| removed those tags from the necessary .pm files in /usr/local/lib/argus
| (backing up the originals of course) but I am not sure what their
| purpose was.
| 
| Can anyone tell me if removing those tags will cause me problems?

probably.


you are still running the old version of arguscgi. verify
that your web server is finding the correct cgi. then, if
you are using mod_perl, stop and restart the web server.



	--jeff

From Joel at smartz.com  Thu Dec  9 20:33:45 2004
From: Joel at smartz.com (Joel Gray)
Date: Thu Dec  9 20:33:52 2004
Subject: Navigation links disappeared
Message-ID: <FCFB8D38040DC54BBED7F8A4A8F94EBC1150AA@mission.smartz.com>

Jeff,

Thank you!  That was exactly the problem.  I had not moved the arguscgi
into my web directory, now all is well.

-Joel 

-----Original Message-----
From: arguslist-bounces@tcp4me.com [mailto:arguslist-bounces@tcp4me.com]
On Behalf Of Jeff Weisberg
Sent: Thursday, December 09, 2004 3:47 PM
To: arguslist@tcp4me.com
Subject: Re: Navigation links disappeared


| I recently (this morning) upgraded from version 3.2 to 3.3.  The 
| upgrade went ok as far as I can tell, however the links on the pages 
| are no longer there.  When I go to the Top page, no links are 
| available on the left.  I do have all of my monitor groups and I can 
| drill down into them.  When doing that, the only link I have on the
left is the "Top"
| link.
| 
| Looking at the rendered HTML, I noticed that all of the missing links 
| as well as some other missing titles for columns being out put as <a 
| href="...."><L10N Display Config></a>.  What was the <L10N> tag for?  
| I

they are used to provide i18n/l10n support, and are processed by
arguscgi.


| removed those tags from the necessary .pm files in 
| /usr/local/lib/argus (backing up the originals of course) but I am not

| sure what their purpose was.
| 
| Can anyone tell me if removing those tags will cause me problems?

probably.


you are still running the old version of arguscgi. verify that your web
server is finding the correct cgi. then, if you are using mod_perl, stop
and restart the web server.



	--jeff

_______________________________________________
http://argus.tcp4me.com/
Arguslist@tcp4me.com
http://www.tcp4me.com/mailman/listinfo/arguslist

--
This message has been scanned for viruses and dangerous content, and is
believed to be clean.

From Sebastian.Kayser at bifab.de  Fri Dec 10 08:06:46 2004
From: Sebastian.Kayser at bifab.de (Sebastian.Kayser@bifab.de)
Date: Fri Dec 10 08:06:57 2004
Subject: object uname in notification subject?
Message-ID: <OF58450DF9.DA56B03E-ONC1256F66.004645CD-C1256F66.00480822@notes.bifab.de>

Hi all,

i have been playing around with customized notifications methods a while 
ago. Currently my overriden mail notifcation method looks like this:

Method "mail" {
   command:    /usr/sbin/sendmail -t
   send:       To: %R\nFrom: %F\nSubject: [Argus] - %O %S\n\n%M\n\n\
Overview: https://argus.ma.bifab.de/cgi-bin/arguscgi?func=ntfylist\n\
Notify: https://argus.ma.bifab.de/cgi-bin/arguscgi?func=ntfyack;idno=%I
}

Is it possible to display the uname of an object in the subject of the 
notification mails? Currently the subjects get rather messed up by those 
long object names. 

I tried the %O{param} sequence (%O{'uname'}, %O{uname}) to no avail. 
Aren't the items listed on the debugging page of an object its parameters?

Regards,

Sebastian Kayser
From Sebastian.Kayser at bifab.de  Wed Dec 15 12:54:16 2004
From: Sebastian.Kayser at bifab.de (Sebastian Kayser)
Date: Wed Dec 15 12:50:24 2004
Subject: object uname in notification subject?
In-Reply-To: <OF58450DF9.DA56B03E-ONC1256F66.004645CD-C1256F66.00480822@notes.bifab.de>
Message-ID: <OFC90FA05A.FB087A93-ONC1256F6B.0060FB2C-C1256F6B.00621DB9@notes.bifab.de>

> i have been playing around with customized notifications methods a while 

> ago. Currently my overriden mail notifcation method looks like this:
> 
> Method "mail" {
>    command:    /usr/sbin/sendmail -t
>    send:       To: %R\nFrom: %F\nSubject: [Argus] - %O %S\n\n%M\n\n\
> Overview: https://argus.ma.bifab.de/cgi-bin/arguscgi?func=ntfylist\n\
> Notify: https://argus.ma.bifab.de/cgi-bin/arguscgi?func=ntfyack;idno=%I
> }
> 
> Is it possible to display the uname of an object in the subject of the 
> notification mails? Currently the subjects get rather messed up by those 

> long object names. 
> 
> I tried the %O{param} sequence (%O{'uname'}, %O{uname}) to no avail. 
> Aren't the items listed on the debugging page of an object its 
parameters?

I just dug into the source code - nice comments there ;) - and it seems as 
if the problem is hidden in NotMe.pm. The message text is normally 
referred to as $txt and when it comes to expansion for %O parameters as 
$msg.

The following "patch" fixes this for both the current dev version 
(20041008) and the stable 3.3.

--- src/NotMe.pm        2004-12-15 18:36:32.000000000 +0100
+++ src/NotMe.pm.patched        2004-12-15 18:36:53.000000000 +0100
@@ -189,7 +189,7 @@
     if( $txt =~ /%O/ ){
        my( $param, $value );

-       if( $msg =~ /%O{([^\}]+)}/ ){
+       if( $txt =~ /%O{([^\}]+)}/ ){
            $param = $1;
            $param =~ s/^\s+//;
            $param =~ s/\s+$//;

Regards,

Sebastian Kayser
From jaw+arguslist at tcp4me.com  Thu Dec 23 15:07:05 2004
From: jaw+arguslist at tcp4me.com (Jeff Weisberg)
Date: Thu Dec 23 15:07:10 2004
Subject: object uname in notification subject?
Message-ID: <200412232007.iBNK75x17183@penelope.tcp4me.com>



| I just dug into the source code - nice comments there ;) - and it seems as 
| if the problem is hidden in NotMe.pm. The message text is normally 
| referred to as $txt and when it comes to expansion for %O parameters as 
| $msg.


yes, that does appear to be a bug. fixed. thanks.


	--jeff
From jaw+arguslist at tcp4me.com  Fri Dec 24 17:10:13 2004
From: jaw+arguslist at tcp4me.com (Jeff Weisberg)
Date: Fri Dec 24 17:10:20 2004
Subject: MySQL replication testing
Message-ID: <200412242210.iBOMADE06254@penelope.tcp4me.com>



Resurrecting an old thread:

| > I'm trying to monitor the replication status on my MySQL
| > server and I want to ensure that Slave_IO_Running and Slave_SQL_Running
| > are both Yes.  

[...]

| this could, possibly, yield false positives, though.  To be realy _really_
| sure, run some perl:

[...]

yes, this would work, but running an external script
to do something only slightly different than what argus
can already do, feels wrong.

it's just that no has filed the right feature request...

I don't have any MySQL servers to test with, but I just
added a feature that should get you what you want easier.


	Service DB {
		dsn:	DBI:mysql:host=hostname
		user:	username
		pass:	password
		sql:	show slave status
		rowfmt:	%{Slave_IO_Running} %{Slave_SQL_Running}
		expect:	Yes Yes
	}

notice new feature 'rowfmt'. comments welcome.


	--jeff

From jaw+arguslist at tcp4me.com  Fri Dec 24 17:15:41 2004
From: jaw+arguslist at tcp4me.com (Jeff Weisberg)
Date: Fri Dec 24 17:15:46 2004
Subject: new features
Message-ID: <200412242215.iBOMFfr06410@penelope.tcp4me.com>



| notice new feature 'rowfmt'. comments welcome.

today's dev code also adds tests for
TCP/SIP, UDP/SIP, and UDP/IAX2
for anyone running VOIP


	--jeff
From friz at godshell.com  Fri Dec 24 19:13:23 2004
From: friz at godshell.com (Jason 'XenoPhage' Frisvold)
Date: Fri Dec 24 19:13:50 2004
Subject: MySQL replication testing
In-Reply-To: <200412242210.iBOMADE06254@penelope.tcp4me.com>
References: <200412242210.iBOMADE06254@penelope.tcp4me.com>
Message-ID: <41CCB0A3.4030605@godshell.com>

Jeff Weisberg wrote:

>yes, this would work, but running an external script
>to do something only slightly different than what argus
>can already do, feels wrong.
>  
>
Agreed :)

>it's just that no has filed the right feature request...
>  
>
I'll have to keep that in mind..  :)

Speaking of features...  It would be really nice to see the current 
value of a test on the web page for that object.  (ping response time, 
snmp response, etc)  It would be even better if it could be viewed on 
both the web page, *AND* be included in the notify..  I haven't seen a 
way to do this thus far...  So...  Feature request!

>notice new feature 'rowfmt'. comments welcome.
>  
>
Awesome!  I'll give this a shot!

This should work with any DBD?

>	--jeff
>  
>


-- 
---------------------------
Jason 'XenoPhage' Frisvold
Engine / Technology Programmer
friz@godshell.com
RedHat Certified - RHCE # 803004140609871
MySQL Pro Certified - ID# 207171862
MySQL Core Certified - ID# 205982910
---------------------------
"Something mysterious is formed, born in the silent void. Waiting alone
and unmoving, it is at once still and yet in constant motion. It is the
source of all programs. I do not know its name, so I will call it the
Tao of Programming."

From friz at godshell.com  Fri Dec 24 19:13:54 2004
From: friz at godshell.com (Jason 'XenoPhage' Frisvold)
Date: Fri Dec 24 19:14:02 2004
Subject: new features
In-Reply-To: <200412242215.iBOMFfr06410@penelope.tcp4me.com>
References: <200412242215.iBOMFfr06410@penelope.tcp4me.com>
Message-ID: <41CCB0C2.9060601@godshell.com>

Jeff Weisberg wrote:

>| notice new feature 'rowfmt'. comments welcome.
>
>today's dev code also adds tests for
>TCP/SIP, UDP/SIP, and UDP/IAX2
>for anyone running VOIP
>  
>
Excellent!  Now I just need to get VoIP going :)

>	--jeff
>  
>

-- 
---------------------------
Jason 'XenoPhage' Frisvold
Engine / Technology Programmer
friz@godshell.com
RedHat Certified - RHCE # 803004140609871
MySQL Pro Certified - ID# 207171862
MySQL Core Certified - ID# 205982910
---------------------------
"Something mysterious is formed, born in the silent void. Waiting alone
and unmoving, it is at once still and yet in constant motion. It is the
source of all programs. I do not know its name, so I will call it the
Tao of Programming."

From darren at ieworks.net  Fri Dec 24 19:20:01 2004
From: darren at ieworks.net (Darren Ellis)
Date: Fri Dec 24 19:20:22 2004
Subject: Event Correlation
In-Reply-To: <41CCB0C2.9060601@godshell.com>
References: <200412242215.iBOMFfr06410@penelope.tcp4me.com>
	<41CCB0C2.9060601@godshell.com>
Message-ID: <41CCB231.4050401@ieworks.net>

Hi,

I use Argus to monitor our network, which has xDSL, wireless and 
dialup.  One feature that I haven't found out how to configure is event 
correlation.  Is there a way to set up a hierarchy of nodes such that we 
get alerted only when a parent node goes down, and the other nodes 
"behind" it, from Argus' view, are simply marked unreachable?

Thanks much.

Darren Ellis
From friz at godshell.com  Fri Dec 24 19:30:29 2004
From: friz at godshell.com (Jason 'XenoPhage' Frisvold)
Date: Fri Dec 24 19:30:37 2004
Subject: Event Correlation
In-Reply-To: <41CCB231.4050401@ieworks.net>
References: <200412242215.iBOMFfr06410@penelope.tcp4me.com>	<41CCB0C2.9060601@godshell.com>
	<41CCB231.4050401@ieworks.net>
Message-ID: <41CCB4A5.3070806@godshell.com>

Darren Ellis wrote:

> Hi,
>
> I use Argus to monitor our network, which has xDSL, wireless and 
> dialup.  One feature that I haven't found out how to configure is 
> event correlation.  Is there a way to set up a hierarchy of nodes such 
> that we get alerted only when a parent node goes down, and the other 
> nodes "behind" it, from Argus' view, are simply marked unreachable?

Yeah, you need to set up a dependency tree... 

I'm not sure where it is in the documentation, but you set it up like 
this : (note the depends keyword)

Group "UPS" {
   Host "UPS 1" {
      hostname: 10.1.0.11
      graph: yes
      Service Ping
      Service UDP/SNMP {
         depends: Top:UPS:UPS_1:Ping_10.1.0.11
         label: On Battery
         community: dream-on
         oid: .1.3.6.1.4.1.318.1.1.1.4.1.1.0
         eqvalue: 2
         graph: no
      }
}

Not sure if you can set up multiple or circular dependencies..  I never 
tried...  :)

> Thanks much.
>
> Darren Ellis


-- 
---------------------------
Jason 'XenoPhage' Frisvold
Engine / Technology Programmer
friz@godshell.com
RedHat Certified - RHCE # 803004140609871
MySQL Pro Certified - ID# 207171862
MySQL Core Certified - ID# 205982910
---------------------------
"Something mysterious is formed, born in the silent void. Waiting alone
and unmoving, it is at once still and yet in constant motion. It is the
source of all programs. I do not know its name, so I will call it the
Tao of Programming."

From darren at ieworks.net  Fri Dec 24 19:42:27 2004
From: darren at ieworks.net (Darren Ellis)
Date: Fri Dec 24 19:42:38 2004
Subject: Event Correlation
In-Reply-To: <41CCB4A5.3070806@godshell.com>
References: <200412242215.iBOMFfr06410@penelope.tcp4me.com>	<41CCB0C2.9060601@godshell.com>	<41CCB231.4050401@ieworks.net>
	<41CCB4A5.3070806@godshell.com>
Message-ID: <41CCB773.7080803@ieworks.net>

Jason,

Thanks for the clue.  I'll give it a whirl.

Darren

Jason 'XenoPhage' Frisvold wrote:

> Darren Ellis wrote:
>
>> Hi,
>>
>> I use Argus to monitor our network, which has xDSL, wireless and 
>> dialup.  One feature that I haven't found out how to configure is 
>> event correlation.  Is there a way to set up a hierarchy of nodes 
>> such that we get alerted only when a parent node goes down, and the 
>> other nodes "behind" it, from Argus' view, are simply marked 
>> unreachable?
>
>
> Yeah, you need to set up a dependency tree...
> I'm not sure where it is in the documentation, but you set it up like 
> this : (note the depends keyword)
>
> Group "UPS" {
>   Host "UPS 1" {
>      hostname: 10.1.0.11
>      graph: yes
>      Service Ping
>      Service UDP/SNMP {
>         depends: Top:UPS:UPS_1:Ping_10.1.0.11
>         label: On Battery
>         community: dream-on
>         oid: .1.3.6.1.4.1.318.1.1.1.4.1.1.0
>         eqvalue: 2
>         graph: no
>      }
> }
>
> Not sure if you can set up multiple or circular dependencies..  I 
> never tried...  :)
>
>> Thanks much.
>>
>> Darren Ellis
>
>
>

From friz at godshell.com  Fri Dec 24 19:48:24 2004
From: friz at godshell.com (Jason 'XenoPhage' Frisvold)
Date: Fri Dec 24 19:48:32 2004
Subject: Event Correlation
In-Reply-To: <41CCB773.7080803@ieworks.net>
References: <200412242215.iBOMFfr06410@penelope.tcp4me.com>	<41CCB0C2.9060601@godshell.com>	<41CCB231.4050401@ieworks.net>	<41CCB4A5.3070806@godshell.com>
	<41CCB773.7080803@ieworks.net>
Message-ID: <41CCB8D8.7020405@godshell.com>

Darren Ellis wrote:

> Jason,
>
> Thanks for the clue.  I'll give it a whirl.

No problem, happy holidays!

> Darren



-- 
---------------------------
Jason 'XenoPhage' Frisvold
Engine / Technology Programmer
friz@godshell.com
RedHat Certified - RHCE # 803004140609871
MySQL Pro Certified - ID# 207171862
MySQL Core Certified - ID# 205982910
---------------------------
"Something mysterious is formed, born in the silent void. Waiting alone
and unmoving, it is at once still and yet in constant motion. It is the
source of all programs. I do not know its name, so I will call it the
Tao of Programming."

From jaw+arguslist at tcp4me.com  Fri Dec 24 19:56:36 2004
From: jaw+arguslist at tcp4me.com (Jeff Weisberg)
Date: Fri Dec 24 19:56:41 2004
Subject: MySQL replication testing
Message-ID: <200412250056.iBP0ua714947@penelope.tcp4me.com>


| Speaking of features...  It would be really nice to see the current 
| value of a test on the web page for that object.  (ping response time, 
| snmp response, etc)  It would be even better if it could be viewed on 
| both the web page, *AND* be included in the notify..  I haven't seen a 
| way to do this thus far...  So...  Feature request!

you can include '%v' in a notification message
	messagedn: pos server broke again (%v)

you cannot currently include it on the web page.


| >notice new feature 'rowfmt'. comments welcome.
| >  
| >
| Awesome!  I'll give this a shot!
| 
| This should work with any DBD?


yep. although I can't think of any other places
it would be useful


	--jeff
From jaw+arguslist at tcp4me.com  Fri Dec 24 20:10:08 2004
From: jaw+arguslist at tcp4me.com (Jeff Weisberg)
Date: Fri Dec 24 20:10:14 2004
Subject: Event Correlation
Message-ID: <200412250110.iBP1A8I15942@penelope.tcp4me.com>


| I'm not sure where it is in the documentation, but you set it up like 
| this : (note the depends keyword)

it is one of *many* poorly documented features only
documented on:
    http://argus.tcp4me.com/config-details.html


[...]
| Not sure if you can set up multiple or circular dependencies..  I never 
| tried...  :)

you can.

	--jeff
From Sebastian.Kayser at bifab.de  Tue Dec 28 05:34:05 2004
From: Sebastian.Kayser at bifab.de (Sebastian Kayser)
Date: Tue Dec 28 05:34:36 2004
Subject: inheritable messagedn/messageup?
Message-ID: <OF0CA0F09F.F06DA4CB-ONC1256F78.0037C4EC-C1256F78.00399838@notes.bifab.de>

Hi,

is it possible to define some kind of global messagedn/messageup? 
Currently i am working with message_fmt:

message_fmt:   %t - State %s\nReason: %r\nValue: %v

Obviously reason is not given for a up notification and i don't care about 
the value of a positive test either, so i would like to get rid of them 
for the up message.

The message format for the up notification should solely consist of 
something like:

message_fmt:   %t - State %s

With messagedn/messageup this would be possible, but those are not 
inheritable (and i don't want to define a messagedn/up for each of my ~90 
monitored objects)? Is there another way to do this or could argus be 
enhanced to have inheritable messagedn/messageup?

Regards,

Sebastian Kayser
From jaw+arguslist at tcp4me.com  Fri Dec 31 13:22:26 2004
From: jaw+arguslist at tcp4me.com (Jeff Weisberg)
Date: Fri Dec 31 13:22:31 2004
Subject: inheritable messagedn/messageup?
Message-ID: <200412311822.iBVIMQv10029@penelope.tcp4me.com>


| is it possible to define some kind of global messagedn/messageup? 
| Currently i am working with message_fmt:
| 
| message_fmt:   %t - State %s\nReason: %r\nValue: %v
| 
| Obviously reason is not given for a up notification and i don't care about 
| the value of a positive test either, so i would like to get rid of them 
| for the up message.
| 
| The message format for the up notification should solely consist of 
| something like:
| 
| message_fmt:   %t - State %s
| 
| With messagedn/messageup this would be possible, but those are not 
| inheritable (and i don't want to define a messagedn/up for each of my ~90 
| monitored objects)? Is there another way to do this 

there is not currently any other way to do that.


| or could argus be 
| enhanced to have inheritable messagedn/messageup?

or separate message_fmt(up,dn), or something else,
I'll need to think about that.


	--jeff