Re: cache lookup failed for function 1 on fmgr_info_cxt_security, fmgr.c:184

Fix it, change the lanvalidator value of 1 to the value of fmgr_c_validator.
Thanks a lot Tom, that was fantastic. Not sure how the lanvalidator
value changed to 1.

Thanks,
Soni

On 23/04/2019 10:39, Soni wrote:
> Here it is:
>
> select * from pg_language;
>  lanname  | lanowner | lanispl | lanpltrusted | lanplcallfoid |
> laninline | lanvalidator | lanacl
> ----------+----------+---------+--------------+---------------+-----------+--------------+--------
>
>  internal |       10 | f       | f            |             0
> |         0 |         2246 |
>  sql      |       10 | f       | t            |             0
> |         0 |         2248 |
>  plpgsql  |       10 | t       | t            |         13793 |    
> 13794 |        13795 |
>  c        |       10 | f       | f            |             0
> |         0 |            1 |
> (4 rows)
>
> Thanks,
> Soni
>
> On 23/04/2019 10:31, Tom Lane wrote:
>> Soni <diptatapa(at)gmail(dot)com> writes:
>>> Here's the backtrace. Please let me know if it helps.
>>> #4  OidFunctionCall1Coll (functionId=functionId(at)entry=1,
>>> collation=collation(at)entry=0, arg1=arg1(at)entry=75471) at fmgr.c:1325
>> hm, there's "function 1" ...
>>
>>> #5  0x0000000000532054 in ProcedureCreate (procedureName=<optimized
>>> out>, procNamespace=procNamespace(at)entry=2200, replace=<optimized out>,
>>>       returnsSet=returnsSet(at)entry=0 '\000',
>>> returnType=returnType(at)entry=2950, proowner=10,
>>> languageObjectId=languageObjectId(at)entry=13,
>>>       languageValidator=languageValidator(at)entry=1,
>> ... and that's where it's coming from.  Seems like you have corrupt
>> data in pg_language.  Might be entertaining to see what
>> "select * from pg_language" gives.
>>
>>             regards, tom lane