Re: "failed to find parent tuple for heap-only tuple" error as anERRCODE_DATA_CORRUPTION ereport()

On 3/1/18 2:19 PM, Peter Geoghegan wrote:
> On Thu, Mar 1, 2018 at 11:15 AM, David Steele <david@pgmasters.net> wrote:
>> On 12/15/17 5:31 PM, Peter Geoghegan wrote:
>>> Commit d70cf811, from 2014, promoted an Assert() within
>>> IndexBuildHeapScan() to a "can't happen" elog() error, in order to
>>> detect when a parent tuple cannot be found for some heap-only tuple --
>>> if this happens, then it indicates corruption. I think that we should
>>> make it a full ereport(), with an errcode of ERRCODE_DATA_CORRUPTION,
>>> to match what Andres just added to code that deals with freezing (he
>>> promoted Assert()s to errors, just like the 2014 commit, though he
>>> went as far as making them ereport()s to begin with). Attached patch
>>> does this.
>>>
>>> I propose a backpatch to 9.3, partially for the sake of tools like
>>> amcheck, where users may only be on the lookout for
>>> ERRCODE_DATA_CORRUPTION and ERRCODE_INDEX_CORRUPTED.
>>
>> This patch applies, passes testing, and appears very straight-forward to
>> me.  Are there any tests for these conditions currently, or are you only
>> doing that in amcheck?
> 
> The enhanced amcheck in the current CF only tests these conditions
> indirectly, by pretending to be a CREATE INDEX statement. It matters
> to amcheck because these conditions are pretty plausible symptoms of
> corruption, and I can imagine someone missing them because they are
> not either ERRCODE_DATA_CORRUPTION or ERRCODE_INDEX_CORRUPTED.
> 
> If amcheck didn't exist, then I'd still think that this patch was worthwhile.

Yes, I agree.  My thrust was more to discover if there is any testing
for these conditions being done in core.  It sounds like no, but I don't
think it's the responsibility of this patch to add them.

I'll mark it Ready for Committer.

-- 
-David
david@pgmasters.net