Re: RFC - advice actually - on availability reporting for management/PHB

Dear Sir,

I am writing to thank you for your letter and say,

On Tue, Oct 22, 2002 at 03:03:20PM -0700, amol sapkal wrote:
> Hi Sir,
> 
> I am not sure whether I have really understood your
> concern.



> But would like you to know that currently I am working
> on scripts to write all log data to a MySQL DB. This
> is necessarily being done to built up a Problem
> Management facility around NetSaint using a DB.
> 
> 
> 
> Regarding SLAs and downtimes, my experience suggests
> that an end-user is just not satisified with knowing
> the downtimes and uptime percentages. He/She will want
> to also know the reasons for the various downtimes.
> This is very imperative when it comes to understanding
> SLAs. Also the resolution factor is of utmost
> importance.
> 

Agreed. The approach we are trying here is to have Netsaint/Nag host and
service handlers update an RDBMS table with a description of the outage.

Basically the outage description consists of what failed, the time the
it failed, and the time it was fixed.

The DB schema includes fields for Change Notes, Problem Investigation 
reports, Description etc that are added manually by those reponsible 
for the service or host.

The summary field contains the plugin output as a reminder of what 
happened.

Here's an example

Inserted '(undef, Network reachability, ade1, PING CRITICAL - Packet
loss = 30%, RTA = 1238.00 ms, 2002-10-24 09:51:47, 2002-10-24 09:56:37,
No, Yes, Availability monitor)' into db.

['No' means unsecheduled, 'Yes' means auto-entered by 'Availability 
monitor']

> As of now, NetSaint limits itslef to showing the
> uptimes and availability in numerical terms.
> 
> What I am trying to achieve is, to built uptime pages
> and associating them with the comment logfile, which
> can be used by a NOC (N/w Operations Center)to add
> updates. 

Our approach sounds similar: the RDBMS will be updated by Netsaint with 
down records (derived from HARD state changes) and any additional info 
manually added by the host/service admins.

The RDBMS can be viewed in ways that meet the reporting needs (whether 
by PHP pages that present the results of standard queries: all downs for 
this service in this period) or arbitrary queries from ODBC clients.


> 
> This all will be done using,
> 1.The perfdata option in NetSaint (wherein all data
> generated byt the NetSait plugins will be written to
> another logfile. The DB will not access the NetSaint
> logfile, which by default will be a subset of the
> perfdata logfile, but will read its data from this
> large logfile)
> 
> 2.The coments.log file will be read by the DB on a
> regular interval.
>

This sounds like a good idea but here the comment information is often 
either stale or non existent.
 
> 3. When the uptime is generated using seperate cgis
> and this DB, the dowtimes will be associated to the
> comment.log file (any comments keyed in for a link,
> during its downtime will be associated with that
> downtime)
> 
> 3. Also, putting it vaguely, a different config file
> for SLAs needs to be defined, which the DB and cgis
> will lookup for while generating uptime reports.
> 

Agreed. 

I don't have too many ideas about this yet. Having an outage store that 
supports arbitrary queries - that describe the SLA - is the only thing 
that comes to mind at the moment.

> Regarding updation of the DB, it can be done in almost
> realtime, by creating timestamped logfiles
> (23102002.hostname.txt) and making the DB update
> itslef regularly( this can be defined using a perl
> script to parse data from txt file to the DB and
> running it through shell - cron jobs)
> This will also avoid read write access violations if
> any, when trying to read from NetSaint's logfiles.
> 
> 
> 
> Warm Regds,
> 
> Amol Sapkal
> Operations Engineer
> GTL Ltd., Mumbai

Yours sincerely.

-- 
------------------------------------------------------------------------
Stanley Hopcroft
------------------------------------------------------------------------

'...No man is an island, entire of itself; every man is a piece of the
continent, a part of the main. If a clod be washed away by the sea,
Europe is the less, as well as if a promontory were, as well as if a
manor of thy friend's or of thine own were. Any man's death diminishes
me, because I am involved in mankind; and therefore never send to know
for whom the bell tolls; it tolls for thee...'

from Meditation 17, J Donne.


-------------------------------------------------------
This sf.net email is sponsored by: Influence the future 
of Java(TM) technology. Join the Java Community 
Process(SM) (JCP(SM)) program now. 
http://ad.doubleclick.net/clk;4729346;7592162;s?http://www.sun.com/javavote