RE: Silent deadlock possible in current sources
I'm inclined to think that that is the correct solution and the new
approach is simply broken. But, not knowing what Vadim had in mind
while making this change, I'm going to leave it to him to fix this.
Thanks, Tom! I'll take care about this...
Although this specific lockup mode didn't exist in 7.0.*, it does
suggest a possible cause of the deadlocks-with-no-deadlock-report
behavior that a couple of people have reported with 7.0: maybe there
^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
is another logic path that allows a deadlock involving two
buffer locks, or a buffer lock and a normal lock. I'm on the
warpath now ...
Buffer locks were implemented in 6.5.
Vadim
"Mikheev, Vadim" <vmikheev@SECTORBASE.COM> writes:
Although this specific lockup mode didn't exist in 7.0.*, it does
suggest a possible cause of the deadlocks-with-no-deadlock-report
behavior that a couple of people have reported with 7.0: maybe there^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
is another logic path that allows a deadlock involving two
buffer locks, or a buffer lock and a normal lock. I'm on the
warpath now ...
Buffer locks were implemented in 6.5.
Yeah, but we've only heard about silent deadlocks from people running
7.0. I'm speculating that some "unrelated" 7.0 change is interacting
badly with the buffer lock management. Haven't gone digging yet, but
I will.
regards, tom lane