Blain Nelson wrote:
> I'm using the precompiled binary for linux on a intel 686. It's been
> working just fine for me for several weeks, but, today, I decided to get
> cute and set it up to load mysql on boot. So I went to /etc/rc.d (which
> existed, where the init.d that the book talked about did not), and moved
> into it the file mysql.server from usr/local/mysql/support-files.
> On boot, mysql wasn't loaded, so I tried to get it to load the
> old-fashioned way I had been doing it, and it began giving me a new kind
> of error thing than I'd seen before.
> It would say something to the effect of
> "Number of processes running: 2
> start mysqld process  hanging -- killed
> killing 117 -- no such process"
> and then maybe a line or two more, and then repeat this with the process
> number going up by 31 or 32 each time. It would include a time stamp on
> each start line, with it being 2 seconds after the one previous.
> If I just left it looping, the numbers go on into the thousands, and it
> doesn't seem to stop. ^C will put a few command prompts across the
> screen, but doesn't stop the process. ^Z does nothing either. The only
> solution I found was to open another virtual terminal and reboot.
> It now does this anytime I try to load mysql (using
> /usr/local/mysql/bin/safe_mysqld &, as I was told in the book).
> So, did I break this? Is there a quick fix? How many lashes must I get
> with the wet noodle for this?
> Your help will be appreciated.
You should cd to "/usr/local/mysql" and then start "bin/safe_mysqld &".
If you start it from root, then mysqld can't find the databases and aborts.
As you start it with safe_mysqld this process will try to start mysqld again and again and
If you want to stop this, you have to kill the safe_mysqld process (no need to reboot).
PS: Sorry for the late answer, I was really busy.