Re: FailedAssertion("!(PrivateRefCount[i] == 0)", File: "bufmgr.c", Line: 1741

On Wed, May 30, 2012 at 1:47 PM, Robert Haas <robertmhaas(at)gmail(dot)com> wrote:
>> The process holding the AccessExclusiveLock is the startup process. It's
>> holding the lock on behalf of the transaction in the master. But something's
>> wrong, and the AccessExclusiveLock doesn't stop a regular backend from
>> acquiring the AccessShareLock on the table. I suspect the fast-path locking
>> patch, because this works on 9.1.
>
> Yeah, apparently so.  gdb says that FastPathStrongRelationLocks on the
> standby is all-zeros even after that record has been replayed.  Not
> sure how that's possible yet.

Ah. The problem is that FastPathTag() expects that locks on database
objects will only be taken by backends with a non-zero value for
MyDatabaseId. Apparently the can-i-use-the-fastpath test and the
do-i-need-to-force-other-people-out-of-the-fastpath test need to be a
bit more asymmetrical than they are at present.

--
Robert Haas
EnterpriseDB: http://www.enterprisedb.com
The Enterprise PostgreSQL Company