I have a brand new server running CentOS 5 with 16 cores and 20GB of RAM. In case it's important the machine is a Xen para-virutalised guest, but it's the only guest on the server and the server has 32 cores and 48GB of RAM.
When I issue:
service mysqld start
I get a timeout and it says it failed. However, if I wait about 45 seconds longer the server then comes up. There are no errors listed in the mysqld log file.
I've never seen mysqld behave like this before, and it can't be that it doesn't have enough resources in this case!
Any ideas where I can even begin to debug this since there are no error messages?