I have two FC5 machines destined to run zimbra. One of them is working absolutely fine with 3.1.1, running all components.

The second one has, store, snmp, mta and spell.

The kernels are identical:
2.6.15-1.2054_FC5smp #1 SMP Tue Mar 14 16:05:46 EST 2006 i686 i686 i386 GNU/Linux

The ram is identical (8GB). The processors are identical (Dual 2.8GHz Xeon)

However, the machine with fewer components on it fails to start the mail store because ...
/opt/zimbra/logs/mysqld.log:
Code:
060602 13:34:13  mysqld started
060602 13:34:13  InnoDB: Error: cannot allocate 2547007488 bytes of
InnoDB: memory with malloc! Total allocated memory
InnoDB: by InnoDB 29721352 bytes. Operating system errno: 12
InnoDB: Check if you should increase the swap file or
InnoDB: ulimits of your operating system.
InnoDB: On FreeBSD check you have compiled the OS with
InnoDB: a big enough maximum process size.
InnoDB: We keep retrying the allocation for 60 seconds...
InnoDB: Fatal error: cannot allocate the memory for the buffer pool
060602 13:35:13 [ERROR] Can't init databases
060602 13:35:13 [ERROR] Aborting
and
Code:
[root@zs2 log]# cat /opt/zimbra/tomcat/logs/catalina.out
zmtomcatstart: info: stale pid 6059 in pid file: No such process
Error occurred during initialization of VM
Could not reserve enough space for object heap
Could not create the Java virtual machine.
ulimits are unlimited (idenitcal on both machines)

has been tried with and without swap enabled.

[root@zs2 log]# cat /proc/meminfo
MemTotal: 8293120 kB
MemFree: 7975384 kB

The only difference between the two machines is that the working one has a hardware RAID controller, and the other one is Software raid based.

We've run full hardware diags on the non-working machine, and all is reported fine.

I am at something of a loss as to why this machine can't allocate the RAM. Does anyone have any pointers?

Many thanks,
Simon