Commit f8d4db63 authored by Klaus Aehlig's avatar Klaus Aehlig
Browse files

Add shelltest for the --ignore-dynu option of hbal

In the example, we have two nodes and 6 instances, all using external
storage. There are four small instances on one node, and two big
instances, requiring twice the amount of resources of a small instance,
on the other node. So, with respect to static data, the cluster
is already perfectly balanced. Using the default assumption of
dynamic usage, however, the cluster can be improved by moving
one small instance to the other node. The reason is, that by default,
all instances are assumed to use the same resources, and this measure
dominates, as all static resources are greatly underutilized.
Signed-off-by: default avatarKlaus Aehlig <>
Reviewed-by: default avatarMichele Tartara <>
parent 14b4e045
......@@ -1181,6 +1181,7 @@ TEST_FILES = \
test/data/htools/hail-invalid-reloc.json \
test/data/htools/hail-node-evac.json \
test/data/htools/hail-reloc-drbd.json \
test/data/htools/ \
test/data/htools/ \
test/data/htools/ \
test/data/htools/ \
......@@ -127,3 +127,13 @@ diff -u $T/simu-rebal-merged.tiered $T/simu-rebal.tiered.original
./test/hs/hbal -t$TESTDATA_DIR/ -G group-02
>>>/Group size 0 nodes, 0 instances/
>>>= 0
# By default, hbal should assume equal, non-zero utilisation
./test/hs/hbal -t$TESTDATA_DIR/
>>>/Solution length=1/
# ...but the --ignore-dynu option should be honored
./test/hs/hbal -t$TESTDATA_DIR/ --ignore-dynu
>>>/Cluster is already well balanced/
Markdown is supported
0% or .
You are about to add 0 people to the discussion. Proceed with caution.
Finish editing this message first!
Please register or to comment