Collect available logs at the end of the each test run

[gshif]

At the end of the each test run (or when exit happens because of a failure) all of the available logs and event should be collected:

• for gateway component
• for queryd component (queryd-a)
• for storage component
• for tasks component
• for etcd components
• for etcd-tasks components
• for transpilerde component
• for kafka component
• for influxdb component
• for zookeeper components
• for deployd component

[gshif]

The work is being done in https://github.com/influxdata/idpe/pull/2146

[gshif]

Try collecting logs by querying the monitoring influxdb

[gshif]

https://github.com/influxdata/idpe/pull/2877

• Collecting logs by querying http://influxdb.logging.svc.cluster.local:8086/query?db=telegraf
• All of the containers write their logs into syslog measurement.
• Do not use log collection with kubectl