Re: pg_xlogdump fails to handle WAL file with multi-page XLP_FIRST_IS_CONTRECORD data

On 23 March 2016 at 18:04, Pavan Deolasee <pavan(dot)deolasee(at)gmail(dot)com> wrote:

>
>
> On Wed, Mar 23, 2016 at 1:13 PM, Michael Paquier <
> michael(dot)paquier(at)gmail(dot)com> wrote:
>
>>
>> + /*
>> + * Compute targetRecOff. It should typically be greater than short
>> + * page-header since a valid record can't , but can also be zero
>> when
>> + * caller has supplied a page-aligned address or when we are
>> skipping
>> + * multi-page continuation record. It doesn't matter though
>> because
>> + * ReadPageInternal() will read at least short page-header worth
>> of
>> + * data
>> + */
>> This needs some polishing, there is an unfinished sentence here.
>>
>> + targetRecOff = tmpRecPtr % XLOG_BLCKSZ;
>> targetRecOff, pageHeaderSize and targetPagePtr could be declared
>> inside directly the new while loop.
>>
>
> Thanks Michael for reviewing the patch. I've fixed these issues and new
> version is attached.
>
>
Can you describe the process used to generate the sample WAL segment?

I'd like to turn it into a TAP test to go along with the patch.

--
Craig Ringer http://www.2ndQuadrant.com/
PostgreSQL Development, 24x7 Support, Training & Services