InnoDB: Error: unlock row could not find a 4 mode lock on the record

2012-01-04 Thread kk r
Hi All, InnoDB repeatedly prints this message in the error log: # InnoDB: Error: unlock row could not find a 4 mode lock on the record Im not sure if the error message is critical or not, but its definitely not nice to have in the error log. After some exploration, I found this error is relevant

Re: Common Pattern for parent-child INSERTs?

2012-01-04 Thread Claudio Nanni
Jan, Just thinking out loud. In relational theory views are just like any entity, in this case the referential integrity would be with the same logical entity, and in the moment of the insert the referential integrity constraint is violated because the new [names_id] will be present only after the

Re: Common Pattern for parent-child INSERTs?

2012-01-04 Thread Jan Steinman
Thanks, Claudio. What you suggested is essentially what I'm doing. I just thought if this were something common, someone would have a better way of doing it. I would LOVE to be able to simply insert into a names-addresses-phones VIEW, but I haven't been able to make that work. On 4 Jan 12, at 1

Re: Common Pattern for parent-child INSERTs?

2012-01-04 Thread Claudio Nanni
Hi Jan, I am not sure to understand what your question is, what do you mean with inserting updating 2-3 tables? I guess treat the 3-tables join as one single 'object' ? Since you have the referential integrity constraint on the [addresses] and [phones] table you need to follow this basic pattern:

Common Pattern for parent-child INSERTs?

2012-01-04 Thread Jan Steinman
Having been steeped in object-orientation, I have a nasty habit of creating parent-child tables that have a 1:1 relationship where the child extends the parent, sometimes to a depth of three or more. For example: CREATE TABLE names TYPE InnoDB id INT NOT NULL AUTO INCREMENT PRIMARY KEY, n

Re: (off topic) why PATH

2012-01-04 Thread Reindl Harald
Am 05.01.2012 00:52, schrieb Jan Steinman: >> From: Reindl Harald >> >> however: both "chroot" and "virtualization" has nothing to do with this >> whole topic, really nothing > > But that shouldn't keep someone from going off on an anti-Mac rant, no? :-) > The OP was on a Mac, and it really do

Re: (off topic) why PATH

2012-01-04 Thread Jan Steinman
> From: Reindl Harald > > however: both "chroot" and "virtualization" has nothing to do with this whole > topic, really nothing But that shouldn't keep someone from going off on an anti-Mac rant, no? :-) The OP was on a Mac, and it really doesn't help to tell them they have the wrong computer

Re: why does basic MySQLdump on db table fail with "Permission denied", or else it creates a zero KB file

2012-01-04 Thread Govinda
>> ..which I just (earlier this morning) changed to this: >> export >> PATH="/usr/local/bin:/usr/local/sbin:/usr/local/mysql/bin:/usr/local/mysql/bin/mysqldump:$PATH" > You are missing a point, that the proper thing for PATH is directory (or > effective directory), not runfile in directory. This

Re: why does basic MySQLdump on db table fail with "Permission denied", or else it creates a zero KB file

2012-01-04 Thread Hal�sz S�ndor
;>>> 2012/01/03 11:52 -0500, Govinda ...which strikes me as odd (again, showing how new I am to driving from the CL), because I do NOT see any entry like this: /usr/local/mysql/bin/mysqldump Is "mysql" a symbolic link? ..which I just (earlier this morning) changed to this: