Thao, Thanks for raising these.
I haven't seen the Homestead issue you describe. * Are you able to share the install/run output logs? * Also, what do you see if you run "/etc/init.d/homestead run" from the homestead node's command-line? * Which version of clearwater-docker are you running? There are three issues it might be related to, but they've been fixed - which version of clearwater-docker are you running against? * https://github.com/Metaswitch/clearwater-docker/pull/9 covers an upgrade to work correctly with Cassandra 2.0, but this went in about 3 weeks ago. * https://github.com/Metaswitch/clearwater-docker/issues/10 covers the fact that we used not to enforce a strict ordering of start-up process. I've never seen this affect Cassandra or Homestead, though. * If you were running from the "latest" repo rather than the standard "stable" repo (although I don't think you are), you might have hit the issue that is addressed by https://github.com/Metaswitch/homestead/pull/234 - basically, we weren't specifying a dependency that is normally satisfied by standard Ubuntu VMs, but not by Docker's stripped-down base image. This is now fixed in the new "latest" repo. Also, I noticed that "monit status" reports output on your box - did you start monit manually? (Monit doesn't run well under Docker, so we use supervisord instead.) I think the Ellis issue is https://github.com/Metaswitch/clearwater-docker/issues/12, which was fixed under https://github.com/Metaswitch/clearwater-infrastructure/pull/202. This is broken in the "stable" repo, but is fixed in the "latest" repo. To switch to the "latest" repo, edit https://github.com/Metaswitch/clearwater-docker/blob/master/base/Dockerfile#L19 to replace "http://repo.cw-ngv.com/stable"; with " http://repo.cw-ngv.com/latest";. You'll then need to re-build all the images (probably by deleting the images using "docker rmi" and then re-running the build process). One other question: which platform are you running Docker on? We've seen problems on some CoreOS systems relating to a bug in OverlayFS that breaks UNIX domain sockets - see https://github.com/coreos/bugs/issues/311 for details. I hope that helps - please let me know how you get on! We only recently made our Docker support public, and don't yet have great regression tests for it - we're hoping to flesh this out in the near future. Thanks, Matt From: [email protected] [mailto:[email protected]] On Behalf Of Thao HT Nguyen Sent: 30 June 2015 00:31 To: [email protected] Subject: [Clearwater] clearwater-docker: issues with Homestead: port 8888 not available (and other issues) Hello, I would like to deploy and run the clearwater-docker by applying the docker commands suggested in the README.md on https://github.com/Metaswitch/clearwater-docker/tree/master/. I had several issues, so let's start with the first ones (I will email other issues once I pass through these ones). _________________________________________________________________________________________________________ ISSUES WITH HOMESTEAD: Cassandra (port 8888) not running. I found out that homestead port 8888 was not available. Here is the list of processes captured on homestead: root 1 0 0 23:01 ? 00:00:00 /usr/bin/python /usr/bin/supervisord -c /etc/supervisor/supervisord.conf root 7 1 0 23:01 ? 00:00:00 /usr/sbin/sshd -D root 147 1 0 23:01 ? 00:00:00 /bin/sh /etc/init.d/homestead-prov run root 148 1 0 23:01 ? 00:00:00 nginx: master process /usr/sbin/nginx -g daemon off; root 151 147 0 23:01 ? 00:00:01 /usr/share/clearwater/homestead/env/bin/python -m metaswitch.crest.main --worker-processes 1 www-data 152 148 0 23:01 ? 00:00:00 nginx: worker process www-data 153 148 0 23:01 ? 00:00:00 nginx: worker process www-data 154 148 0 23:01 ? 00:00:00 nginx: worker process www-data 155 148 0 23:01 ? 00:00:00 nginx: worker process root 186 1 4 23:01 ? 00:00:54 /usr/lib/jvm/java-7-openjdk-amd64//bin/java -ea -javaagent:/usr/share/cassandra/lib/jamm-0.2.5.jar -XX:+CMSClassUnloadingEnabled -XX:+UseThreadPriorities -XX:ThreadPriorityPolicy=42 -Xms1001M -Xmx1001M -Xmn250M -XX:+HeapDumpOnOutOfMemoryError -Xss256k -XX root 970 7 0 23:19 ? 00:00:00 sshd: root@pts/0 root 981 970 0 23:19 pts/0 00:00:00 -bash root 1003 1 0 23:19 ? 00:00:00 monit root 1030 1 0 23:19 ? 00:00:00 /usr/sbin/ntpd -p /var/run/ntpd.pid -g -u 103:106 root 1099 1 0 23:19 ? 00:00:00 /bin/bash /usr/share/clearwater/bin/clearwater_diags_monitor root 1104 1099 0 23:19 ? 00:00:00 inotifywait -e create -qq . root 5045 1003 2 23:21 ? 00:00:00 [poll_homestead-] <defunct> cassand+ 5188 1 99 23:21 ? 00:00:04 /usr/lib/jvm/java-7-openjdk-amd64//bin/java -ea -javaagent:/usr/share/cassandra/lib/jamm-0.2.5.jar -XX:+CMSClassUnloadingEnabled -XX:+UseThreadPriorities -XX:ThreadPriorityPolicy=42 -Xms1001M -Xmx1001M -Xmn250M -XX:+HeapDumpOnOutOfMemoryError -Xss256k -XX root 5383 981 0 23:21 pts/0 00:00:00 ps -ef Here is part of the "monit status": Process 'homestead_process' status Does not exist monitoring status Monitored data collected Mon, 29 Jun 2015 23:20:14 Process 'cassandra_process' status Does not exist monitoring status Monitored data collected Mon, 29 Jun 2015 23:20:15 _________________________________________________________________________________________________________ ISSUES WITH ELLIS (or maybe the issues comes from the "base" build?) - Redeployed from scractch last week - The provisioning portal (ellis) doesn't display the domain name anymore. [cid:[email protected]] I found during the "base" build that following error: Step 18 : RUN /etc/init.d/clearwater-auto-config-docker restart ---> Running in 38123f560712 /etc/init.d/clearwater-auto-config-docker: 82: /etc/init.d/clearwater-auto-config-docker: home_domain: not found ---> c43db9b6e245 Removing intermediate container 38123f560712 Step 19 : RUN /etc/init.d/clearwater-infrastructure restart ---> Running in 12a56bd101d0 * Restarting clearwater-infrastructure clearwater-infrastructure Configuring monit for only localhost access ...done. ________________________________ Please let me know how I can fix those issues. Thanks in advance Thao
_______________________________________________ Clearwater mailing list [email protected] http://lists.projectclearwater.org/listinfo/clearwater
