Re: [sqlsmith] Failed assertion in _hash_kill_items/MarkBufferDirtyHint - Mailing list pgsql-hackers

From Ashutosh Sharma
Subject Re: [sqlsmith] Failed assertion in _hash_kill_items/MarkBufferDirtyHint
Date
Msg-id CAE9k0PmqgcnrBQjZ79Y6J9Z-35LJ1UP19mehYOi-u2zOmupEhw@mail.gmail.com
Whole thread Raw
In response to Re: [sqlsmith] Failed assertion in _hash_kill_items/MarkBufferDirtyHint  (Andreas Seltenreich <seltenreich@gmx.de>)
List pgsql-hackers
Hi Andreas,

On Apr 1, 2017 16:15, "Andreas Seltenreich" <seltenreich@gmx.de> wrote:
Andreas Seltenreich writes:
>>> TRAP: FailedAssertion("!(LWLockHeldByMe(((LWLock*) (&(bufHdr)->content_lock))))", File: "bufmgr.c", Line: 3397)
> I got about one TRAP per hour when testing on 20 nodes with one postgres
> and 5 sqlsmithes on each.

> Ashutosh Sharma writes:
>> [2. reacquire_lock_hashkillitems_if_required.patch]

I've done about 12 hours of testing since applying this patch and no
failed assertions so far.


Thanks for testing my patch. Just FYI, I have slightly changed my patch this morning as per Robert's suggestions. But, still more or less the logic remains the same. Thank you once again.

pgsql-hackers by date:

Previous
From: Andreas Seltenreich
Date:
Subject: Re: [sqlsmith] Failed assertion in _hash_kill_items/MarkBufferDirtyHint
Next
From: Dilip Kumar
Date:
Subject: Re: parallel bitmapscan isn't exercised in regression tests