pcp
[Top] [All Lists]

Re: Oracle connection debugging (was Re: [pcp] Handling Oracle PMDA Late

To: Nathan Scott <nathans@xxxxxxxxxx>
Subject: Re: Oracle connection debugging (was Re: [pcp] Handling Oracle PMDA Latencies)
From: Marko Myllynen <myllynen@xxxxxxxxxx>
Date: Fri, 15 Apr 2016 10:06:07 +0300
Cc: pcp developers <pcp@xxxxxxxxxxx>
Delivered-to: pcp@xxxxxxxxxxx
In-reply-to: <57108708.3080906@xxxxxxxxxx>
Organization: Red Hat
References: <56F25541.9020602@xxxxxxxxxx> <874656864.33839965.1458781570723.JavaMail.zimbra@xxxxxxxxxx> <56F36E49.8010501@xxxxxxxxxx> <675902814.33867352.1458794396829.JavaMail.zimbra@xxxxxxxxxx> <570D1333.2040109@xxxxxxxxxx> <899654573.39808794.1460523158800.JavaMail.zimbra@xxxxxxxxxx> <570F511E.5000605@xxxxxxxxxx> <1512930308.40394593.1460673441009.JavaMail.zimbra@xxxxxxxxxx> <57108708.3080906@xxxxxxxxxx>
Reply-to: Marko Myllynen <myllynen@xxxxxxxxxx>
User-agent: Mozilla/5.0 (X11; Linux x86_64; rv:38.0) Gecko/20100101 Thunderbird/38.7.0
Hi,

On 2016-04-15 09:15, Marko Myllynen wrote:
> On 2016-04-15 01:37, Nathan Scott wrote:
>> ----- Original Message -----
>>> On 2016-04-13 07:52, Nathan Scott wrote:
>>>> ----- Original Message -----
>>>>> [...]
>>>>> connect.pl says "OK", after PMDA installation there are no oracle
>>>>> metrics available and the Oracle PMDA log is silent.
>>>>
>>>> Hmm, I've not seen anything like that.  Usually theres a big warning
>>>> in the pmcd/oracle.log file ... and usually its that initial connect
>>>> that fails.
>>>>
>>>>> Are there any recommended sanity checks that could be done in this case?
>>>>
>>>> I'd verify the connection with sqlplus, issuing queries like those
>>>> listed in the pmdaoracle.pl code (near the start - bunch of 'em you
>>>> could try there).  Those should all show columns/values returned by
>>>> sqlplus.
>>>
>>> Here are some additional details and perhaps a clue:
>>>
>>
>> Ah - does this patch help?
>>  
>>  # $ENV{PCP_PERL_DEBUG} = 'LIBPMDA';
>>  $pmda = PCP::PMDA->new('oracle', 32);
>> +$pmda->connect_pmcd;
>>  $pmda->set_user($os_user);
> 
> Unfortunately not, the error remains the same. But enabling the above
> debug I see the following (with the patch):

To follow-up our IRC discussion:

> And finally this:
> 
> [Fri Apr 15 09:08:48] pmdaoracle(125624) Error: pmdaFetch: Unavailable
> metric PMID 32.12.4[1]
> [Fri Apr 15 09:08:48] pmdaoracle(125624) Error: pmdaFetch: Unavailable
> metric PMID 32.12.4[3]
> [Fri Apr 15 09:08:48] pmdaoracle(125624) Error: pmdaFetch: Unavailable
> metric PMID 32.12.4[7]
> 
> [Fri Apr 15 09:08:48] pmdaoracle(125624) Error: pmdaFetch: Unavailable
> metric PMID 32.0.73[0]
> [Fri Apr 15 09:08:48] pmdaoracle(125624) Error: pmdaFetch: Unavailable
> metric PMID 32.0.79[0]

On different test runs the error is printed for a varying set of
metrics. From the command line e.g. this works (to test 32.0.73):

echo 'select statistic#, name, value from v$sysstat;' | sqlplus
scott/tiger@orcl

After "service pmcd restart" (I'm on RHEL 6.8 Beta so HUPing pmcd is not
an option) I see some metrics being available, however there's steady
flow of the above kind of errors printed in the log (with occasional
errors from the line 430 - DBI->connect()).

Thanks,

-- 
Marko Myllynen

<Prev in Thread] Current Thread [Next in Thread>