dcae collector does not listen port 8080 (anymore)

-1

Last Friday I was able to startup DCAE + the additional VMs tracing the traffic between firewall and DCAE collector

  ##########
T 10.1.0.11:33548 -> 10.0.0.5:8080 [AP]
POST /eventListener/v3 HTTP/1.1.
Authorization: Basic Og==.
User-Agent: libcurl-agent/1.0.
Host: 10.0.0.5:8080.
Accept: */*.
Content-type: application/json.
Content-Length: 816.
.

##
T 10.1.0.11:33548 -> 10.0.0.5:8080 [AP]
{"event": {"commonEventHeader": {"domain": "measurementsForVfScaling", "eventId": "42", "functionalRole": "vFirewall", "lastEpochMicrosec": 1498819483942164, "priority": "Normal", "reportingEntityName": "demofwl01fwl", "sequence": 42, "sourceName": "Dummy VM name - No Metadata available", "startEpochMicrosec": 1498819473942164, "version": 1.2, "reportingEntityId": "No UUID available", "sourceId": "Dummy VM UUID - No Metadata available"}, "measurementsForVfScalingFields": {"measurementInterval": 10.000000, "vNicUsageArray": [{"bytesIn": 4343, "bytesOut": 0, "packetsIn": 101, "packetsOut": 0, "vNicIdentifier": "eth1", "broadcastPacketsIn": 0, "broadcastPacketsOut": 0, "multicastPacketsIn": 0, "multicastPacketsOut": 0, "unicastPacketsIn": 0, "unicastPacketsOut": 0}], "measurementsForVfScalingVersion": 1.1}}}
##
T 10.0.0.5:8080 -> 10.1.0.11:33548 [AP]
HTTP/1.1 200 OK.
Server: Apache-Coyote/1.1.
X-Rathravane: ~ software is craft ~.
Content-Type: application/json;charset=ISO-8859-1.
Content-Length: 17.
Date: Fri, 30 Jun 2017 10:44:43 GMT.
.

#
T 10.0.0.5:8080 -> 10.1.0.11:33548 [AP]
Message Accepted

(Don't get confused by 10.0.0.5 as the IP address - this was given to the collector by Openstack; see my other question on this)

Since I restarted DCAE and the associated DCAE VMs, later on Friday, I cannot see a listener at port 8080 anymore. I cannot see an immediate error in cloud-init-output.log

Does anybody know what happend to the listener, or which of the docker images is supposed to listen to 8080?

Comment

CommentAdd your comment...

2 answers

2
1
0
kranthi guttikonda
Jul 03, 2017
Josef Reisinger Following are the details from my setup

root@zldcbm01njcoll00:~# docker ps
CONTAINER ID IMAGE COMMAND CREATED STATUS PORTS NAMES
2eae3ff6feb8 nexus3.onap.org:10001/openecomp/dcae-collector-common-event:1.1-STAGING-latest "/opt/app/manager/..." 4 days ago Up 4 days 0.0.0.0:8080->8080/tcp, 0.0.0.0:8443->8443/tcp, 0.0.0.0:32768->9999/tcp dcae_docker-common-event_bm01
4d0f34ed963f nexus3.onap.org:10001/openecomp/dcae-dmaapbc:1.1-STAGING-latest "./bin/dmaapbc deploy" 4 days ago Up 4 days 0.0.0.0:18080->8080/tcp, 0.0.0.0:18443->8443/tcp dcae_docker-databus-controller_bm01
76f425ef31e7 attos/dmaap "/bin/sh -c ./star..." 4 days ago Up 4 days 0.0.0.0:3904-3905->3904-3905/tcp dockerfiles_dmaap_1
3cbf0d0711d9 dockerfiles_kafka "start-kafka.sh" 4 days ago Up 4 days 0.0.0.0:9092->9092/tcp dockerfiles_kafka_1
1c0d77fb4609 wurstmeister/zookeeper "/bin/sh -c '/usr/..." 4 days ago Up 4 days 22/tcp, 2888/tcp, 3888/tcp, 0.0.0.0:2181->2181/tcp dockerfiles_zookeeper_1

root@zldcbm01njcoll00:~# netstat -uanltp | grep 8080
tcp6 0 0 :::8080 :::* LISTEN 5075/docker-proxy
tcp6 0 0 :::18080 :::* LISTEN 4353/docker-proxy
root@zldcbm01njcoll00:~#

root@zldcbm01njcoll00:~# docker images
REPOSITORY TAG IMAGE ID CREATED SIZE
dockerfiles_kafka latest 0f1fba0e1e46 4 days ago 260MB
nexus3.onap.org:10001/openecomp/dcae-collector-common-event 1.1-STAGING-latest c7057753203a 5 days ago 539MB
anapsix/alpine-java latest aee4f5af1664 5 days ago 124MB
attos/dmaap latest b0ae220fcf1f 12 days ago 747MB
nexus3.onap.org:10001/openecomp/dcae-dmaapbc 1.1-STAGING-latest 1fcf5b48d63b 7 weeks ago 328MB
wurstmeister/zookeeper latest 351aa00d2fe9 7 months ago 478MB
root@zldcbm01njcoll00:~#
Comment
Josef Reisinger
Jul 03, 2017
Thanks kranthi guttikonda
Seems as if DCAE start is not 100% reliable ... I left the system where it was over the weekend and there was not 8080 listener. Shortly after I asked the question I restarted DCAE again (shutting the controller container down) and I could see the docker container listenting on 8080 after a while - This wasn't there before the restart or I could at least not find anything in cloud-init logs which indicated a docker container is missing.
Messages are now coming from vFW into DCAE collector and I see some output from

curl http://collector-ip>:3904/events/unauthenticated.SEC_MEASUREMENT_OUTPUT/group3/sub1?timeout=3000

Well, the next big task now is to understand DCAE I guess (fear). Any hints are greatly appreciated.
kranthi guttikonda
Jul 03, 2017
Josef Reisinger I am also trying to find out proper way to use DCAE
Are you able to access zldcbm01njcdap02 or cdap02 VM on port 9999 from browser? Check if you are able to login automatically. Unfortunately it redirects me to login page. I am having hard time in almost all stages because of http_proxy

Adding the HortonWorks public GPG key to apt failure in CDAP VMs
Josef Reisinger
Jul 03, 2017
@kranthi guttikonda I respond in the above mentioned question
CommentAdd your comment...
1
0
-1
Vijay Venkatesh Kumar
Jul 05, 2017
Adding details responded on the discuss list.

The listener runs inside the dcae-collector-common-event service on the DCAE col VM.
Once you are inside the container – you could run below to confirm
1)      Run -> netstat -an | grep 8080
Should get something like below
tcp        0      0 0.0.0.0:8080            0.0.0.0:*               LISTEN
Also if you tail on VES manager logs (/opt/app/manager/logs/manager.log) – it would report health status info.
2)      If #1 doesn’t give any result and manager reports VESCollecter_Is_Not_Running, then you could bring it up manually “/opt/app/VESCollector/bin/VESrestfulCollector.sh start” command.

After executing #2 you could check #1 to confirm if the listener is actually up. If it is - then events can be sent via the Fw demo appl. If issue still - pls provide logs so we can troubleshoot further.
Comment
CommentAdd your comment...