2014 HPCwire Awards

The StratoStor project I’ve been working on for the past 10 months just got a “Top 5 New Products or Technologies to Watch” award from HPCwire announced at this week’s SuperComputing 2014 (SC14) conference in New Orleans.

HPC = High Performance Computing, HPCwire is a news bureau for all things regarding High Performance Computing, and SC14 is where every major vendor of HPC equipment and products shows off their wares, so getting this bit of recognition from the readers of HPCwire is really nice.

So THANK YOU HPCwire readers, for this award.

https://www.hpcwire.com/2014-hpcwire-readers-choice-awards/23/

2014 HPCwire Awards

Validating Distributed Application Workloads

This is the talk I gave at RICON this year on Validating Distributed Application Workloads. It’s about how we set up test environments at Seagate for validating storage system performance at the petabyte scale. This talk centers around the testing done to validate performance of a 2PB rack running Riak CS.

Increase a VM’s available memory with virsh

If you try to increase the amount of available memory using the obvious command it fails with an error message:

# virsh setmem <vm name> 16G --live
error: invalid argument: cannot set memory higher than max memory

The physical host in this case has 128G RAM and 32 CPUs. Plenty of capacity. To increase the maximum amount of memory that can be allocated to the VM:

# virsh setmaxmem <vm name> 16G --config

There are also –live and –current options which claim to affect the running/current domain. These options do not actually work. You have to use the –config option (changes take effect after next boot) and then power off the machine by logging in and running “poweroff”.

Once the machine is off set the actual memory with:

# virsh setmem <vm name> 16G --config

Then start the vm:

# virsh start <vm name>

Once the VM starts up it will have more memory.

Hope you find this useful.

Increase a VM’s vcpu count with virsh

You have a virtual machine you created with virsh. You want to increase the number of vcpus in the virtual machine, so you use the obvious command:

virsh setvcpus --count 8 <vm name>

… and get the irritating error message:

error: invalid argument: requested vcpus is greater than max allowable vcpus for the domain: 8 > 2

This is virsh telling you that you can’t increase the number of vcpus to a number larger than what you started with.

Although virsh doesn’t support increasing the number of vcpus while the VM is running, you can change the number of vcpus if you’re willing to reboot the VM. All you need to to is to edit the virsh XML file with:

virsh edit <vm name>

Look for the line “vcpu placement” and increase the value to the number of vcpus that you want. I changed the vcpus from 2 to 8 here:

<vcpu placement='static'>8</vcpu>

Save the file.

Shutdown the VM:

virsh shutdown <vm name>

Wait until the VM’s status is “shut down”.

virsh list --all

Destroy the VM:

virsh destroy <vm name>

Start up the VM:

virsh start <vm name>

Once the VM starts you’ll have more vcpus running.

Hope you find this useful.

Getting rid of the “redirecting to systemctl” message in OpenSUSE

On OpenSUSE systems running systemd all rcX scripts now redirect start, stop, reload, restart, etc. service commands to systemctl. The messages that  used to appear on STDOUT telling you that a command is successful (or not) are now logged, but are no longer displayed on STDOUT.

That I can deal with, but every call to an rcX script now generates the message “redirecting to systemctl” to STDERR. I have a lot of scripts that call rcX scripts, and they interpret STDERR messages as “something just broke”.

The culprit is the new /etc/rc.status script that ships with OpenSUSE. It spews out the “redirecting to systemctl” message to STDERR for every operation that you do. The following command will modify the script and remove this stupid message:

if ( grep -q 'redirecting to systemctl' /etc/rc.status ) ; then
    # Save a copy of the original file
    cp -p /etc/rc.status /etc/rc.status.orig;

    # OpenSUSE 12.1:
    perl -i.bak -pe 's,echo "redirecting to systemctl" >/dev/stderr,,;' /etc/rc.status;

    # OpenSUSE 12.3:
    perl -i.bak -pe 's,echo "redirecting to systemctl \${SYSTEMCTL_OPTIONS} \$1 \${_rc_base}" 1>&2,,;' /etc/rc.status;
fi

This works for OpenSUSE 12.1 and 12.3. I did not have a 12.2 system available to test with.

Hope you find this useful.