Going on a limb here...: I believe I have occurred similar issue (i.e. two transactions
go into an indefinite wait). Though, very infrequent occurrence.
My only explanation at that time was that there is some "loophole" when the
deletes/inserts had some impact also on the table indexes. In our case, the
deletes/inserts statements were invoked by a stored procedure.
From: Johan De Meersman [mailto:vegivamp@stripped]
Sent: Monday, May 14, 2012 9:28 AM
To: Baron Schwartz
Subject: Re: Deadlock due lockwait. How can I tell mysql to wait longer?
----- Original Message -----
> From: "Baron Schwartz" <baron@stripped>
> Because it can be resolved by rolling back just one of them. Why
> destroy ALL the work people are trying to accomplish, if you could
> just throw away some of it?
What I fail to understand, Baron, is how there can be a deadlock here - both transactions
seem to be hanging on a single-table, single-row update statement. Shouldn't the oldest
transaction already have acquired the lock by the time the youngest came around; and
shouldn't the youngest simply wait until the eldest finished it's update?
Or is this a problem with the consistent view that I'm not seeing?
Bier met grenadyn
Is als mosterd by den wyn
Sy die't drinkt, is eene kwezel
Hy die't drinkt, is ras een ezel
MySQL General Mailing List
For list archives: http://lists.mysql.com/mysql
To unsubscribe: http://lists.mysql.com/mysql
The information contained in this e-mail and any attached
documents may be privileged, confidential and protected from
disclosure. If you are not the intended recipient you may not
read, copy, distribute or use this information. If you have
received this communication in error, please notify the sender
immediately by replying to this message and then delete it
from your system.