We've been running low on memory for a while now, waiting on the new hardware to come in and be added to the cluster. Today I've finished adding one of the new pieces of hardware into the cluster. virt5 has 185GB of RAM, but otherwise is similarly spec'd to the old hardware.
Even though this new node has a lot more RAM, once it has the same number of instances as the other hosts, it'll stop being the preferred node, since we are using a scheduler that schedules based on the number of instances running on a node. Once all of the hardware has been replaced by the new hardware, we'll be able to run a much larger number of instances than we are currently able to. As of now we have space for another 40 instances, or so. If you notice any issues building instances, let me know. - Ryan _______________________________________________ Labs-l mailing list [email protected] https://lists.wikimedia.org/mailman/listinfo/labs-l
