admin on August 4th, 2010

On this release I’ve done  just some minor changes in particular I’ve added a better exception management, and now Orabbix’s log is even more detailed.

Download Orabbix 0.9.4 here

admin on July 28th, 2010
On this release I’ve added more and more informations on error,
I’ve added default values for:
  • MaxWait
  • MaxIdle
  • MaxActive
  • OrabbixDaemon.Sleep
  • OrabbixDaemon.MaxThreadNumber (this one is calculated and based from number of monitored databases)
in case that Orabbix can’t read values from config properties file (because are missing or invalid) the daemon will go ahead
New features:
  • you can specify for each database different values of:
  1. MaxWait
  2. MaxIdle
  3. MaxActive
specific for each database, so each database can have a different connection pool configuration.
Modified the template excluded all that is not strictly “Orabbix” from template (so i’ve removed cpu performances etc.. and other tests where data don’t come from Orabbix)
admin on July 26th, 2010
Orabbix 0.9.2 is out!
Orabbix Logo
On this release:
  • Added more information on error that should be really appreciate in case of problem with configuration files,logfiles are more verbose, or better should be easily understandable.
  • Added a features that send “0” (zero) or the values NoDataFound when a database is not available so you will see HIT/Ratio  and other graps that fall down to zero and don’t continue to show a value that is not real.
If there are problems or if someone find bugs or problems please let me know!

Tags: , , ,

In a large enviroment like we are there are many groups that use zabbix and it’s used as first level infrastructure monitor and for an indeep and storicized monitor.
Nowdays people that monitor the first level of infrastructure are used to look the “Dashboard” that is really useful to have a “qualitative” view of the state of all the monitored systems. The problem appear when people of first level monitoring continue to CALL YOU to ask “Do you know who is the responsible of this server ?” for this reason I’ve wrote this patch.
With this patch you need to populate “extended properties” inside each  host (anyway is useful to have in one system all the data about your monitored infrastracture) the responsible, the number to call if you see an error, the reference of and service mail etc..
So people that see in “last 20 issue” a trouble just need to clich over  servername and in the section links they are going to find “inventory” with this link you are dragged into inventory of the selected server and the first level will find all the infos that they need and YOU DON’T RECEIVE MORE BORING PHONE CALLING (WOOOW)
The Patch is the following:

the patch is a modify on line 747 of “include/blocks.inc.php”

$menus.= “[‘”.S_LATEST_DATA.”‘,\”javascript : redirect(‘latest.php?groupid=0&hostid=”.$trigger[‘hostid’].”‘)\”, null,{‘outer’ : [‘pum_o_item’],’inner’ : [‘pum_i_item’]}],”;
$menus = rtrim($menus,’,’);

and just add this line:

$menus.= “[‘”.S_INVENTORY.”‘,\”javascript : redirect(‘hostprofiles.php?hostid=”.$trigger[‘hostid’].”‘)\”, null,{‘outer’ : [‘pum_o_item’],’inner’ : [‘pum_i_item’]}],”;

so your file will be so:

$menus.= “[“.zbx_jsvalue(S_LINKS).”,null,null,{‘outer’ : [‘pum_oheader’],’inner’ : [‘pum_iheader’]}],”;
$menus.= “[‘”.S_LATEST_DATA.”‘,\”javascr

ipt : redirect(‘latest.php?groupid=0&hostid=”.$trigger[‘hostid’].”‘)\”, null,{‘outer’ : [‘pum_o_item’],’inner’ : [‘pum_i_item’]}],”;
$menus.= “[‘”.S_INVENTORY.”‘,\”javascript : redirect(‘hostprofiles.php?hostid=”.$trigger[‘hostid’].”‘)\”, null,{‘outer’ : [‘pum_o_item’],’inner’ : [‘pum_i_item’]}],”;
$menus = rtrim($menus,’,’);

the result is
like in picture:

New Link

Tags: ,

admin on July 19th, 2010
I’ve implemented this easy way to keep monitored the health of your standby’s databases.
I use this because you don’t have a way to check if there is a block corruption on standby databases until they became essential for your disaster recovery.
i use rman on each standby database:

on a .bat program

set ORACLE_SID=
set NLS_DATE_FORMAT=YYYY/MM/DD HH24:MI:SS
rman target / nocatalog @rman_check_logical_database.rms msglog rman_check_logical_database.log

and the rman script is:

run {
# Andrea Dalle Vacche
# BACKUP VALIDATE ver. 1.0 19/07/2010
#
backup validate check logical database;
}
exit;

I’ve put this “.bat” on a scheduled task that run every day and after i execute this query on standby:

select * FROM v$database_block_corruption;

NOTE: (got from oracle literature) The V$DATABASE_BLOCK_CORRUPTION view indicates which blocks in a datafile were marked corrupt since the most recent BACKUP  or BACKUP VALIDATE command was run.

So if the query don’t return rows is everything right, otherwise you’ll get the block/blocks that is/are corrupted.

Tags: ,

By continuing to use the site, you agree to the use of cookies. more information

The cookie settings on this website are set to "allow cookies" to give you the best browsing experience possible. If you continue to use this website without changing your cookie settings or you click "Accept" below then you are consenting to this. To know all the policy detais click here.

Close