Re: [Devmon] HP Proliant DL380

[Colin C. <col...@gm...>]

On Thu, Dec 9, 2010 at 8:31 PM, Buchan Milne <bg...@st...> wrote:
> On Thursday, 9 December 2010 02:02:53 Colin Coe wrote:
>
>> I grabbed the code out of the SVN on SF and rebuilt.  It looked a lot
>> better but what I found was on one of the Proliants with a failed
>> drive but no spare,
>
> What template were you using? This probably indicates you should be using (for
> now) compaq-servernohspare, until the questions below are resolved ...
>
>> the test came back clear not red.  I made this
>> change:
>> # diff -u dm_tests.pm.orig dm_tests.pm
>> ----
>> --- dm_tests.pm.orig    2010-12-09 08:07:04.000000000 +0800
>> +++ dm_tests.pm 2010-12-09 08:18:14.000000000 +0800
>> @@ -1820,9 +1820,9 @@
>>
>>         # Make sure we have leaf data for our primary oid
>>          if(!defined $oids->{$pri}{'val'}) {
>> -          do_log("Missing repeater data for $pri for $test msg", 0);
>> -          $msg .= "&clear Missing repeater data for primary OID $pri\n";
>> -          $worst_color = 'clear';
>> +          do_log("Warning: missing repeater data for $pri for $test msg",
>> 0); +#          $msg .= "&clear Missing repeater data for primary OID
>> $pri\n"; +#          $worst_color = 'clear';
>>            next;
>>          }
>>
>> ----
>> and it now comes back red although I don't know what the potential
>> badness of this change is (not knowing the code).
>
> This isn't the right fix, but it is difficult to know what the correct fix is.
> What should occur when a table is totally empty? If there has been one
> complete table on the test, do we ignore failures on the rest? Do we report
> that the tables are empty? Do we try and allow customisation of what happens
> based on the template (e.g., in this case, warn that there are no hot spares
> or not, including allowing the admin to change a threshold)?
>
> Maybe for now the best option is to keep the $msg line?
>
> Regards,
> Buchan
>

Hi Buchan

Many thanks for the pointers.  Using 'servernohspare' does look like
it has fixed my problems.

Another question though, it appears that the 'servernohspare' test
does not show the controller status where plain 'server' does.  Is
this intentional?

Thanks again

CC

-- 
RHCE#805007969328369