Hi,

for the failover part, what about using a real cluster like Linux HA cluster with DRBD? I've setup such a system a few weeks ago for us.

hh

Ralph Mitchell schrieb:

Take a look at the Xymon alert man page:

http://www.hswn.dk/hobbit/help/manpages/man5/hobbit-alerts.cfg.5.html

You can specify a script to be executed for any given alert. The ALERT SCRIPTS section in the man page gives details of the environment variables passed in with the the alert information. So, you could write a script to send snmp traps.

I don't know about alert failover.

Ralph Mitchell

On Wed, Mar 25, 2009 at 9:07 PM, J Sloan <joe at tmsusa.com> wrote:

After a bit of a hiatus from bb/hobbit/xymon issues, I'm looking at it again.

I've grabbed a copy of xymon-4.4.0 svn and set it up on a test server here - very nice.

As hope springs eternal, I'm again hoping to replace our aging big brother servers with xymon, since it's the closest thing to a drop-in replacement. Hopefully xymon can be made to do some specific things that bb does, which we depend on here, namely:

1. snmp traps - this is just a basic requirement to send out an snmp trap on an alert. We have no need to receive snmp traps, only to send them, as big brother does. We depend on that behavior to generate tickets by sending traps to netcool whenever there is a bb alert condition.

2. alerting failover - we have redundant bb servers are set up, in pairs. each pair of bb servers monitors certain network segments, while only one of the pair is designated as the bbpager, i.e. only one of the pair sends out alerts. But, should the passive bbpager be unable to reach the normally active bbpager, failover kicks in, and the normally passive server begins sending out all alerts. Worst case, we have a split brain condition and get 2 copies of an alert, but in practice, that's only happened a handful of times since 2005.

Can xymon 4.4.0 do these things, or is there a way to implement this behavior with some appropriate add-ons?

Joe

-- Harald Husemann Netzwerk- und Systemadministrator Operation Management Center (OMC) MATERNA GmbH Information & Communications

Westfalendamm 98 44141 Dortmund

Geschäftsführer: Dr. Winfried Materna, Helmut an de Meulen, Ralph Hartwig Amtsgericht Dortmund HRB 5839

Tel: +49 231 9505 222 Fax: +49 231 9505 100 www.annyway.com <http://www.annyway.com/> www.materna.com <http://www.materna.com/>

While I don't want to minimize TJ's work, I don't believe people realize that Xymonton, the Xymon repository for custom tests, is actually built on top of Dokuwiki. This is exactly the type of information the tutorials page was meant to capture. It might be helpful if we had a one stop shop for all things Xymon instead of going one place for scripts and another for docs.

Just my 2 cents (not that I have any vested interest in this ;-) )

=G=

-----Original Message----- From: Martin Flemming [mailto:martin.flemming at desy.de] Sent: Thursday, March 26, 2009 6:20 AM To: hobbit at hswn.dk Subject: Re: [hobbit] xymon ready to replace bb?

Hi, Harald !

Xymon with Linux HA cluster with DRBD

did you make some doku's or notes for it ?

.. there is an old thread from Februar but unfortunatley no notes till this moment ...

http://www.hswn.dk/hobbiton/2009/02/msg00146.html

But there is already some place for it :-)

http://en.wikibooks.org/wiki/System_Monitoring_with_Xymon/Administration_Gui...

HA-LAN using LinuxHA and DRBD

[edit]

If you have some notes, please contact T.J. Yang <tj_yang at hotmail.com> to submit it to the wiki

cheers,

    Martin

Martin Flemming DESY / IT office : Building 2b / 008a Notkestr. 85 phone : 040 - 8998 - 4667 22603 Hamburg mail : martin.flemming at desy.de

On Thu, 26 Mar 2009, Husemann, Harald wrote:

Hi,

for the failover part, what about using a real cluster like Linux HA cluster with DRBD? I've setup such a system a few weeks ago for us.

hh

Ralph Mitchell schrieb:

...

Take a look at the Xymon alert man page:

http://www.hswn.dk/hobbit/help/manpages/man5/hobbit-alerts.cfg.5.html

You can specify a script to be executed for any given alert. The ALERT SCRIPTS section in the man page gives details of the environment variables passed in with the the alert information. So, you could write a script to send snmp traps.

I don't know about alert failover.

Ralph Mitchell

On Wed, Mar 25, 2009 at 9:07 PM, J Sloan <joe at tmsusa.com> wrote:

After a bit of a hiatus from bb/hobbit/xymon issues, I'm looking at it again.

I've grabbed a copy of xymon-4.4.0 svn and set it up on a test server here - very nice.

As hope springs eternal, I'm again hoping to replace our aging big brother servers with xymon, since it's the closest thing to a drop-in replacement. Hopefully xymon can be made to do some specific things that bb does, which we depend on here, namely:

1. snmp traps - this is just a basic requirement to send out an snmp trap on an alert. We have no need to receive snmp traps, only to send them, as big brother does. We depend on that behavior to generate tickets by sending traps to netcool whenever there is a bb alert condition.

2. alerting failover - we have redundant bb servers are set up, in pairs. each pair of bb servers monitors certain network segments, while only one of the pair is designated as the bbpager, i.e. only one of the pair sends out alerts. But, should the passive bbpager be unable to reach the normally active bbpager, failover kicks in, and the normally passive server begins sending out all alerts. Worst case, we have a split brain condition and get 2 copies of an alert, but in practice, that's only happened a handful of times since 2005.

Can xymon 4.4.0 do these things, or is there a way to implement this behavior with some appropriate add-ons?

Joe

-- Harald Husemann Netzwerk- und Systemadministrator Operation Management Center (OMC) MATERNA GmbH Information & Communications

Westfalendamm 98 44141 Dortmund

Geschäftsführer: Dr. Winfried Materna, Helmut an de Meulen, Ralph Hartwig Amtsgericht Dortmund HRB 5839

Tel: +49 231 9505 222 Fax: +49 231 9505 100 www.annyway.com <http://www.annyway.com/> www.materna.com <http://www.materna.com/>

To unsubscribe from the hobbit list, send an e-mail to hobbit-unsubscribe at hswn.dk

I don't mind. It doesn't have to happen overnight. :-) Who needs sleep?

-----Original Message----- From: Martin Flemming [mailto:martin.flemming at desy.de] Sent: Thursday, March 26, 2009 10:45 AM To: hobbit at hswn.dk Subject: RE: [hobbit] xymon ready to replace bb?

Grmph,

you're right, one place for all is the best solution ..

Me for myself,

i switch from side to side ... and would prefer on central Side, too !

... by the way, the wiki-side from TJ is very well and comprehensive ...

there will be a lot of things to merge into the "offical" Xymonton ...

martin

On Thu, 26 Mar 2009, Galen Johnson wrote:

While I don't want to minimize TJ's work, I don't believe people realize that Xymonton, the Xymon repository for custom tests, is actually built on top of Dokuwiki. This is exactly the type of information the tutorials page was meant to capture. It might be helpful if we had a one stop shop for all things Xymon instead of going one place for scripts and another for docs.

Just my 2 cents (not that I have any vested interest in this ;-) )

=G=

-----Original Message----- From: Martin Flemming [mailto:martin.flemming at desy.de] Sent: Thursday, March 26, 2009 6:20 AM To: hobbit at hswn.dk Subject: Re: [hobbit] xymon ready to replace bb?

Hi, Harald !

...

Xymon with Linux HA cluster with DRBD

did you make some doku's or notes for it ?

.. there is an old thread from Februar but unfortunatley no notes till this moment ...

http://www.hswn.dk/hobbiton/2009/02/msg00146.html

But there is already some place for it :-)

http://en.wikibooks.org/wiki/System_Monitoring_with_Xymon/Administration_Gui...

HA-LAN using LinuxHA and DRBD

[edit]

If you have some notes, please contact T.J. Yang <tj_yang at hotmail.com> to submit it to the wiki

cheers,

   Martin

---

Martin Flemming DESY / IT office : Building 2b / 008a Notkestr. 85 phone : 040 - 8998 - 4667 22603 Hamburg mail : martin.flemming at desy.de

---

On Thu, 26 Mar 2009, Husemann, Harald wrote:

...

Hi,

for the failover part, what about using a real cluster like Linux HA cluster with DRBD? I've setup such a system a few weeks ago for us.

hh

Ralph Mitchell schrieb:

...

Take a look at the Xymon alert man page:

http://www.hswn.dk/hobbit/help/manpages/man5/hobbit-alerts.cfg.5.html

You can specify a script to be executed for any given alert. The ALERT SCRIPTS section in the man page gives details of the environment variables passed in with the the alert information. So, you could write a script to send snmp traps.

I don't know about alert failover.

Ralph Mitchell

On Wed, Mar 25, 2009 at 9:07 PM, J Sloan <joe at tmsusa.com> wrote:

After a bit of a hiatus from bb/hobbit/xymon issues, I'm looking at it again.

I've grabbed a copy of xymon-4.4.0 svn and set it up on a test server here - very nice.

As hope springs eternal, I'm again hoping to replace our aging big brother servers with xymon, since it's the closest thing to a drop-in replacement. Hopefully xymon can be made to do some specific things that bb does, which we depend on here, namely:

1. snmp traps - this is just a basic requirement to send out an snmp trap on an alert. We have no need to receive snmp traps, only to send them, as big brother does. We depend on that behavior to generate tickets by sending traps to netcool whenever there is a bb alert condition.

2. alerting failover - we have redundant bb servers are set up, in pairs. each pair of bb servers monitors certain network segments, while only one of the pair is designated as the bbpager, i.e. only one of the pair sends out alerts. But, should the passive bbpager be unable to reach the normally active bbpager, failover kicks in, and the normally passive server begins sending out all alerts. Worst case, we have a split brain condition and get 2 copies of an alert, but in practice, that's only happened a handful of times since 2005.

Can xymon 4.4.0 do these things, or is there a way to implement this behavior with some appropriate add-ons?

Joe

-- Harald Husemann Netzwerk- und Systemadministrator Operation Management Center (OMC) MATERNA GmbH Information & Communications

Westfalendamm 98 44141 Dortmund

Geschäftsführer: Dr. Winfried Materna, Helmut an de Meulen, Ralph Hartwig Amtsgericht Dortmund HRB 5839

Tel: +49 231 9505 222 Fax: +49 231 9505 100 www.annyway.com <http://www.annyway.com/> www.materna.com <http://www.materna.com/>

To unsubscribe from the hobbit list, send an e-mail to hobbit-unsubscribe at hswn.dk

To unsubscribe from the hobbit list, send an e-mail to hobbit-unsubscribe at hswn.dk

Gruss

    Martin Flemming

Martin Flemming DESY / IT office : Building 2b / 008a Notkestr. 85 phone : 040 - 8998 - 4667 22603 Hamburg mail : martin.flemming at desy.de

Hi Martin,

I can of course write a doc about our setup to share my experiences. As I saw, there was a bit of a discussion about the right place for it, so, question to the list: Where shall I upload it? Or shall I send it directly to the list, and someone (TJ?) picks it + includes it?

hh

Martin Flemming schrieb:

Hi, Harald !

...

Xymon with Linux HA cluster with DRBD

did you make some doku's or notes for it ?

.. there is an old thread from Februar but unfortunatley no notes till this moment ...

http://www.hswn.dk/hobbiton/2009/02/msg00146.html

But there is already some place for it :-)

http://en.wikibooks.org/wiki/System_Monitoring_with_Xymon/Administration_Gui...

HA-LAN using LinuxHA and DRBD

[edit]

If you have some notes, please contact T.J. Yang <tj_yang at hotmail.com> to submit it to the wiki

cheers,

    Martin

---

Martin Flemming DESY / IT office : Building 2b / 008a Notkestr. 85 phone : 040 - 8998 - 4667 22603 Hamburg mail : martin.flemming at desy.de

---

On Thu, 26 Mar 2009, Husemann, Harald wrote:

...

Hi,

for the failover part, what about using a real cluster like Linux HA cluster with DRBD? I've setup such a system a few weeks ago for us.

hh

Ralph Mitchell schrieb:

...

Take a look at the Xymon alert man page:

http://www.hswn.dk/hobbit/help/manpages/man5/hobbit-alerts.cfg.5.html

You can specify a script to be executed for any given alert. The ALERT SCRIPTS section in the man page gives details of the environment variables passed in with the the alert information. So, you could write a script to send snmp traps.

I don't know about alert failover.

Ralph Mitchell

On Wed, Mar 25, 2009 at 9:07 PM, J Sloan <joe at tmsusa.com> wrote:

After a bit of a hiatus from bb/hobbit/xymon issues, I'm looking at it again.

I've grabbed a copy of xymon-4.4.0 svn and set it up on a test server here - very nice.

As hope springs eternal, I'm again hoping to replace our aging big brother servers with xymon, since it's the closest thing to a drop-in replacement. Hopefully xymon can be made to do some specific things that bb does, which we depend on here, namely:

1. snmp traps - this is just a basic requirement to send out an snmp trap on an alert. We have no need to receive snmp traps, only to send them, as big brother does. We depend on that behavior to generate tickets by sending traps to netcool whenever there is a bb alert condition.

2. alerting failover - we have redundant bb servers are set up, in pairs. each pair of bb servers monitors certain network segments, while only one of the pair is designated as the bbpager, i.e. only one of the pair sends out alerts. But, should the passive bbpager be unable to reach the normally active bbpager, failover kicks in, and the normally passive server begins sending out all alerts. Worst case, we have a split brain condition and get 2 copies of an alert, but in practice, that's only happened a handful of times since 2005.

Can xymon 4.4.0 do these things, or is there a way to implement this behavior with some appropriate add-ons?

Joe

-- Harald Husemann Netzwerk- und Systemadministrator Operation Management Center (OMC) MATERNA GmbH Information & Communications

Westfalendamm 98 44141 Dortmund

Geschäftsführer: Dr. Winfried Materna, Helmut an de Meulen, Ralph Hartwig Amtsgericht Dortmund HRB 5839

Tel: +49 231 9505 222 Fax: +49 231 9505 100 www.annyway.com <http://www.annyway.com/> www.materna.com <http://www.materna.com/>

To unsubscribe from the hobbit list, send an e-mail to hobbit-unsubscribe at hswn.dk

-- Harald Husemann Netzwerk- und Systemadministrator Operation Management Center (OMC) MATERNA GmbH Information & Communications

Westfalendamm 98 44141 Dortmund

Geschäftsführer: Dr. Winfried Materna, Helmut an de Meulen, Ralph Hartwig Amtsgericht Dortmund HRB 5839

Tel: +49 231 9505 222 Fax: +49 231 9505 100 www.annyway.com <http://www.annyway.com/> www.materna.com <http://www.materna.com/>

I would send it here. This is pushed to people and stored by Henrik who couldn't be more tied to Xymon. Great begining and will last forever :)

On 3/28/09, Martin Flemming <martin.flemming at desy.de> wrote:

Hi, Harald !

.. good question ... but it should not be a religious question .. :-)

.. but i think, in both ways, you need a registration for the rights to append your notes ...

So obvisously, it should the best way for all Xymon's to have one and only Wiki/Howto Side ...

For Xymonton would be Galen the right person ....

...

Or shall I send it directly to the list

hmm, maybe the quickest solution ...

thanks & cheers,

martin

On Fri, 27 Mar 2009, Husemann, Harald wrote:

...

Hi Martin,

I can of course write a doc about our setup to share my experiences. As I saw, there was a bit of a discussion about the right place for it, so, question to the list: Where shall I upload it? Or shall I send it directly to the list, and someone (TJ?) picks it + includes it?

hh

Martin Flemming schrieb:

...

Hi, Harald !

...

Xymon with Linux HA cluster with DRBD

did you make some doku's or notes for it ?

.. there is an old thread from Februar but unfortunatley no notes till this moment ...

http://www.hswn.dk/hobbiton/2009/02/msg00146.html

But there is already some place for it :-)

http://en.wikibooks.org/wiki/System_Monitoring_with_Xymon/Administration_Gui...

HA-LAN using LinuxHA and DRBD

[edit]

If you have some notes, please contact T.J. Yang <tj_yang at hotmail.com> to submit it to the wiki

cheers,

     Martin

______________________________________________________ Martin Flemming DESY / IT office : Building 2b / 008a Notkestr. 85 phone : 040 - 8998 - 4667 22603 Hamburg mail : martin.flemming at desy.de

---

On Thu, 26 Mar 2009, Husemann, Harald wrote:

...

Hi,

for the failover part, what about using a real cluster like Linux HA cluster with DRBD? I've setup such a system a few weeks ago for us.

hh

Ralph Mitchell schrieb:

...

Take a look at the Xymon alert man page:

http://www.hswn.dk/hobbit/help/manpages/man5/hobbit-alerts.cfg.5.html

You can specify a script to be executed for any given alert. The ALERT SCRIPTS section in the man page gives details of the environment variables passed in with the the alert information. So, you could write a script to send snmp traps.

I don't know about alert failover.

Ralph Mitchell

On Wed, Mar 25, 2009 at 9:07 PM, J Sloan <joe at tmsusa.com> wrote:

After a bit of a hiatus from bb/hobbit/xymon issues, I'm looking at it again.

I've grabbed a copy of xymon-4.4.0 svn and set it up on a test server here - very nice.

As hope springs eternal, I'm again hoping to replace our aging big brother servers with xymon, since it's the closest thing to a drop-in replacement. Hopefully xymon can be made to do some specific things that bb does, which we depend on here, namely:

1. snmp traps - this is just a basic requirement to send out an snmp trap on an alert. We have no need to receive snmp traps, only to send them, as big brother does. We depend on that behavior to generate tickets by sending traps to netcool whenever there is a bb alert condition.

2. alerting failover - we have redundant bb servers are set up, in

pairs. each pair of bb servers monitors certain network segments, while only one of the pair is designated as the bbpager, i.e. only one of the pair sends out alerts. But, should the passive bbpager be unable to reach the normally active bbpager, failover kicks in, and the normally passive server begins sending out all alerts. Worst case, we have a split brain condition and get 2 copies of an alert, but in practice, that's only happened a handful of times since 2005.

Can xymon 4.4.0 do these things, or is there a way to implement this behavior with some appropriate add-ons?

Joe

To unsubscribe from the hobbit list, send an e-mail to hobbit-unsubscribe at hswn.dk

-- Josh Luthman Office: 937-552-2340 Direct: 937-552-2343 1100 Wayne St Suite 1337 Troy, OH 45373

Those who don't understand UNIX are condemned to reinvent it, poorly. --- Henry Spencer

Husemann, Harald wrote:

Hi,

for the failover part, what about using a real cluster like Linux HA cluster with DRBD? I've setup such a system a few weeks ago for us.

That would be a rather disruptive and expensive change. Since bb's failover capability has been working perfectly since 2005, it would be a hard sell in this economic climate to convince management to shell out money to build 5 clusters. They'll tell me "forget it, let's stay with big brother"

It would be tricky to do in our situation also because a cluster would typically be built from machines in the same lan. Trying to build a cluster of machines on completely different networks as ours are, would be a bit more challenging, since neither side could take over an IP from the other.

I will admit, that if I was building something new from scratch, your idea would be the way to go. But unfortunately the company has become dependent on some specific features of big brother, thus my search for a drop-in replacement.

Joe

On Thu, Mar 26, 2009 at 5:53 PM, Malcolm Hunter <malcolm.hunter at gmx.co.uk>wrote:

...

The other (the slave one) first checks the status of the other server (a simpel wget of the status page can be enough) and only sends out the alert if this page is not green.

So, basically, both servers are triggering on the same alert, but the slave server only sends out the alert if the primary server is not green.

Wouldn't there be more involved in this? What if the primary server's hobbit daemon was down, but the web service was still running? The secondary server would want to report the hobbit daemon being down, but wouldn't because the primary server's page was still green and hadn't been updated.

The hobbit pages have a time stamp in the title, so the slave can grab the page, grep for <title>, extract the date, convert to seconds using gnudate. Then the slave gets its own timestamp in seconds, subtracts the primary's time and complains if there's more than a couple of minutes difference.

That timestamp doesn't update if the hobbit daemon is down.

Ralph Mitchell

-----Oorspronkelijk bericht----- Van: McDonald, Dan [mailto:Dan.McDonald at austinenergy.com] Verzonden: vrijdag 27 maart 2009 0:40 Aan: hobbit at hswn.dk Onderwerp: Re: [hobbit] xymon ready to replace bb?

On Thu, 2009-03-26 at 18:31 -0500, Ralph Mitchell wrote:

On Thu, Mar 26, 2009 at 5:53 PM, Malcolm Hunter <malcolm.hunter at gmx.co.uk> wrote: > The other (the slave one) first checks the status of the other server (a > simpel wget of the status page can be enough) and only sends out the alert > if > this page is not green. Why not point the master's client to the slave, and if bbd goes red/purple, then take over....

Do you have a sample configuration for this? I'd be interested...

On Friday 27 March 2009 00:25:53 Stef Coene wrote:

For the failover, configure both servers to report to each other it's situation. Configure both servers to trigger a script when there is an alert. Except one of them is the master and is always sending out the alert. The other (the slave one) first checks the status of the other server (a simpel wget of the status page can be enough) and only sends out the alert if this page is not green.

I would rather:

bb $MASTERDISPLAY 'query $MASTERDISPLAY.bbd'|grep -q ^green || enable_hobbitd_alert

(where implementation of enable_hobbitd_alert is left as an exercise for the reader)

Hi everyone,

I've setup a custom script to return several values that are being entered into the same RRD by xymon. I'd like to graph those values, but the values differ by thousands so I can't really put them on the same graph. Is there any way to generate multiple graphs on a single test page? I see there are several vmstat tests like vmstat1, vmstat2, etc. Would I be able to use that format to give me more than one graph?

Also, in hobbitserver.cfg there is a value for mrtg like this: mrtg::1. I can't find any explanation of the significance of the double colon. Can anyone fill me in?

BTW, I just changed from BB commercial edition to Xymon. Xymon fixes everything I always disliked about BB, not to mention that Quest has virtually halted development on BB since it purchased it.

Thanks,

Scot Kreienkamp

La-Z-Boy Inc.