On 06/18/2014 11:32 AM, Dan Prince wrote:
> Would this fix (or something similar) help nodepool to allocate things
> more efficiently?
>
> https://review.openstack.org/#/c/88223/
That's an interesting approach. Just looping around the same little
test from [1] with 20 nodes across two providers, and 50 ubuntu nodes
and 1 fedora, we get
---
(py27)$ cat run_test.sh
#!/bin/sh
RUNS=500
FEDORA_COUNT=0
for i in `seq 1 $RUNS`
do
python ./test.py 10 10 1 50 | grep -q fedora
if [ $? -eq 0 ]; then
FEDORA_COUNT=$((FEDORA_COUNT+1))
fi
done
echo $FEDORA_COUNT out of $RUNS got a fedora node
(py27)$ sh ./run_test.sh
25 out of 500 got a fedora node
---
Without the patch, we never see a fedora node. So it would probably
help in that dire situation we were in a few days ago, although we
could tweak it considerably.
(p.s. I want to turn this into a test-case, once we know what sort of
result we're looking for)
-i
[1] https://gist.github.com/ianw/541d2b7ad8983acbf225
_______________________________________________
OpenStack-Infra mailing list
OpenStack-Infra@lists.openstack.org
http://lists.openstack.org/cgi-bin/mailman/listinfo/openstack-infra
