Details
-
Bug
-
Resolution: Incomplete
-
Major
-
5.0.1
-
None
-
Untriaged
-
Unknown
Description
During basic demo test, I noticed that the ops/sec suffers greatly when adding one node to just the data service. This has been tested a number of times before over previous versions with much less impact.
Screenshot attached, logs are:
https://s3.amazonaws.com/cb-engineering/perry/rebalanceops/collectinfo-2018-03-15T100918-ns_1%40ec2-34-217-79-250.us-west-2.compute.amazonaws.com.zip
https://s3.amazonaws.com/cb-engineering/perry/rebalanceops/collectinfo-2018-03-15T100918-ns_1%40ec2-52-36-156-116.us-west-2.compute.amazonaws.com.zip
https://s3.amazonaws.com/cb-engineering/perry/rebalanceops/collectinfo-2018-03-15T100918-ns_1%40ec2-54-218-36-208.us-west-2.compute.amazonaws.com.zip
https://s3.amazonaws.com/cb-engineering/perry/rebalanceops/collectinfo-2018-03-15T100918-ns_1%40ec2-54-218-59-234.us-west-2.compute.amazonaws.com.zip
https://s3.amazonaws.com/cb-engineering/perry/rebalanceops/collectinfo-2018-03-15T100918-ns_1%40ec2-54-244-57-146.us-west-2.compute.amazonaws.com.zip
https://s3.amazonaws.com/cb-engineering/perry/rebalanceops/collectinfo-2018-03-15T100918-ns_1%40ec2-54-245-47-44.us-west-2.compute.amazonaws.com.zip
This was running a pillowfight load:
cbc-*pill*owfight -u Administrator -P password -U http://ec2-52-36-156-116.us-west-2.compute.amazonaws.com:8091/travel-sample?unsafe_optimize=true -m 20 -M 20 -I 10000 -t 1 -r 1
cbc version
cbc:
Runtime: Version=2.8.5, Changeset=e9c46be209881f7a8c583523a7a58a2dbd88c110
Headers: Version=2.8.5, Changeset=e9c46be209881f7a8c583523a7a58a2dbd88c110
IO: Default=libevent, Current=select, Accessible=select
SSL Runtime: OpenSSL 1.0.2k-fips 26 Jan 2017
SSL Headers: OpenSSL 1.0.2k-fips 26 Jan 2017
Snappy: 1.1.1
Tracing: NOT SUPPORTED
System: Linux-4.15.3-300.fc27.x86_64; x86_64
CC: GNU 4.8.5; -O2 -g -pipe -Wall -Wp,-D_FORTIFY_SOURCE=2 -fexceptions -fstack-protector-strong --param=ssp-buffer-size=4 -grecord-gcc-switches -m64 -mtune=generic -fno-strict-aliasing -pthread
CXX: GNU 4.8.5; -O2 -g -pipe -Wall -Wp,-D_FORTIFY_SOURCE=2 -fexceptions -fstack-protector-strong --param=ssp-buffer-size=4 -grecord-gcc-switches -m64 -mtune=generic -fno-strict-aliasing -pthread
Attachments
Issue Links
- relates to
-
MB-26791 Latency of GET operations exceeds minutes during swap rebalance (9 nodes, 1TB)
-
- Closed
-