[prev in list] [next in list] [prev in thread] [next in thread] 

List:       linux-ha-dev
Subject:    [Linux-ha-dev] Re: [Linux-HA] a simple question about service monitor
From:       "Leandro Ariel Gomez Chavarria" <lgomez () cencosud ! com ! ar>
Date:       2003-11-13 16:32:35
[Download RAW message or body]

Hi Henry

I did something like that for samba.

Basically, what I did it's a script to launch samba, and after that it
re-launch itself as a demon and monitor the processes each 10 secs. Also
monitors the status of the disks (I use a SAN IBM FastT700 and Volume
Manager). If the demon find a samba demon down it relaunch it and if it
can't after 3 times force a swicht of the service. Also if there are
some kind of problem with the disks force an inmediate switch.

I use heartbeat 1.1.3 because the hb_standby functionalities, it works
really fine.

Besides, I have an Active-Active configuration, because I need 2
different samba instances, I give service to 2 different "areas" of the
company, actually are 2 different countries ;-).

Another functionality of this script it's to make switches at my will.

All you need it's just a bit of imagination and a lot of work.

good luck

Leandro.-

>>> qxiong@hotmail.com 11/11/03 04:26am >>>
Hi Alan and Linux-HA community,
since currently heartbeat package only monitor node, rather individual
service, thus, I am planing to implement a monitor to check whether or
not a
service is running well, for this purpose, the haresource will be like
this:
node1 10.0.0.1 service1
node1 10.0.0.2 service2

In node1, there are totally two resource groups, each group has only
one
service and its own virtul IP
The monitor will periodically check the availability of those services.
If
monitor detect that one service ,eg. service1, fail, HA will force
10.0.0.1
with service1 move to node2. Thus,
10.0.0.1 service1 runs in node2
10.0.0.2 service2 runs in node1
my question is:
1. Can I implement this monitor to do the above funtion in conjunct
with
current Linux HA?
2. If do, What kind of API/function in HA do  I need to call to update
local 
node status --
e.g. hold all resources or some resource? or notify heartbeat that
10.0.0.1 
servce1 will be moved to mode2?

Thank you in advance!
Henry

_________________________________________________________________
From Beethoven to the Rolling Stones, your favorite music is always
playing 
on MSN Radio Plus. No ads, no talk. Trial month FREE!  
http://join.msn.com/?page=offers/premiumradio 

_______________________________________________
Linux-HA mailing list
Linux-HA@lists.linux-ha.org 
http://lists.linux-ha.org/mailman/listinfo/linux-ha
_______________________________________________________
Linux-HA-Dev: Linux-HA-Dev@lists.linux-ha.org
http://lists.linux-ha.org/mailman/listinfo/linux-ha-dev
Home Page: http://linux-ha.org/
[prev in list] [next in list] [prev in thread] [next in thread]