Re: [msql-list] FreeBSD 7 issues

[David J. Hughes]

Thanks for the feedback Doug. I too am looking into this on a fresh
FreeBSD 7.0 system. It certainly is odd that the malloc in the new
FreeBSD can behave in a way that causes a failure like this when we
haven't seen it on any other version of UNIX / Linux with normal or
debugging malloc implementations. This is truly a strange issue.

David
..

On 09/04/2008, at 6:24 PM, Doug Hardie wrote:

> I am beginning to get a handle on the problems that I am
> encountering. What is happening is one or more of the broker child
> processes is becoming corrupt. It no longer returns a valid
> response but decides that one of the data elements in the query is
> not in the table definition. It almost seems periodic when the
> problem occurs. I have 5 backends running and get n out of five
> failures with 5 - n correct responses. Restarting msql causes all
> failures to end. I don't know yet how to make the problems occur
> on demand. When the problem occurs it also affects msql with
> simple queries. The value of n has been between 1 and 3 so far. I
> am still trying to figure out how to make the problem occur. It
> doesn't happen very often (days between).
> -----------------------------------------------------------------
> This is the Mini SQL Mailing List operated by Hughes Technologies
> To unsubscribe, go to http://www.Hughes.com.au/extras/email/
>
-----------------------------------------------------------------
This is the Mini SQL Mailing List operated by Hughes Technologies
To unsubscribe, go to http://www.Hughes.com.au/extras/email/