Page 1 of 1

Re: [Nagios-devel] MSQL Tables and log files mis-match in data

Posted: Wed Mar 27, 2002 12:31 pm
by Guest
Thanks for the pointer on the future Ethan, I am not sure how many of the
Netsaint/Nagios users out here are using the database support, but it seems
to be a reasonable number. Would you envisage us adding database support as
a 'patch' or might you consider providing an API against which we could
implement any other 'storage service' that we might want to play with.

I guess that the alternative might be for someone ;-) to provide a simple
SQL-->Files bridge (mmmm sounds a bit like mSQL v 0.1) for those who still
want/need db support.

Personally I am playing with the database as the files get a little clunky
when writing other analysis code and or interfaces to monitoring/management
systems other than Nagios :-(. With the db tables some of these operations
are much faster (to code and to perform), especially when log-file rotation
(archiving) is enabled.

Where can the community add most value here ?


-----Original Message-----
[]On Behalf Of Ethan
Sent: Wednesday, March 27, 2002 7:27 PM
Subject: Re: [Nagios-devel] MSQL Tables and log files mis-match in data
types, any ideas

Thanks for pointing this out. The inconsistencies should be fixed,
although its not a huge deal. Changing the DB tables at this point
might be a waste of time, as I plan on dropping native DB support in
version 2.0.

On 27 Mar 2002 at 18:32, Mark Gillett wrote:

> Hi All,
> I have been investigating linking some of our in-house tools to Nagios and
> have been working on reading from either the log files or a MySql database
> for retention and status information.
> In these explorations I have discovered some 'maybe' inconsistencies in
> data types used and descriptions of these as below:
> I may be missing something or might be just straight dumb on this - has
> anyone else got any ideas ?
> in
> TABLE hostretention --->
> last_check int(11)
> appears to be a short integer and is described as a time_t since the unix
> epoch in the log-file documentation,
> but in
> TABLE hoststatus --->
> last_check datetime
> appears to be a datetime but is still described as a time_t since the
> epoch in the log-file documentation
> also confusing is that in
> TABLE hoststatus --->
> time_up bigint(20)
> time_down bigint(20)
> time_unreachable bigint(20)
> and are described as no. of seconds elapsed in that state, or since the
> start of the server, but in
> TABLE hostretention --->
> time_up int(11)
> time_down int(11)
> time_undreachable int(11)
> which is described as no. of seconds elapsed as well.
> Am I missing something - does anyone know if there is a reason for this,
> not should I look at harmonising the database descriptions. Also should
> think about updating the xdata code to unift the storage between the files
> and the database in terms of data-types.
> Also - has anyone written a file to database or database to file
> tool ?
> Thanks,
> Mark Gillett
> IXIF Ltd
> UK NMS Support
> _______________________________________________
> Nagios-devel mailing list
> ... gios-devel

Ethan Galstad,
Nagios Developer

Nagios-devel mailing list ... gios-devel

This post was automatically imported from historical nagios-devel mailing list archives
Original poster: