[Beowulf] LSI Megaraid stalls system on very high IO?

[mathog]

On 14 May 2015 09:41:07 +1000 Christopher Samuel wrote:
> On 14/05/15 08:57, mathog wrote:
> 
> [...]
>> cat /sys/block/dm*/queue/scheduler
>> none
>> none
> [...]
>> How is it possible to mount a volume, Raid or no, with no scheduler???
> 
> I think you're looking in the wrong place; you need to look at the sys
> files for the disk (sda, sdb, or variations if your RAID controller
> registers them differently) rather than the Device Mapper (DM) entries.
>  They don't have IO schedulers associated with them.

Hmm...

Ah, I see the mix up.  There is a "/dev/sda" entry in /etc/fstab for an 
ntfs
volume, but /etc/mtab shows

/dev/sda1 /boot ext4 rw 0 0

and fdisk run on /dev/sda lists it as a single 8 Tb disk.  So sda is the 
raid after all.  sda is using cfq scheduling.

> As for swap, could it be someone built it with a kickstart that defined
> it that way?

Pretty sure he just put in the Centos install DVD and let it default for 
most things, most likely including swap.

Thanks,

David Mathog
mathog at caltech.edu
Manager, Sequence Analysis Facility, Biology Division, Caltech