- Make disk cleaner on mac clean without requesting permission to acces path install#
- Make disk cleaner on mac clean without requesting permission to acces path update#
- Make disk cleaner on mac clean without requesting permission to acces path password#
Make disk cleaner on mac clean without requesting permission to acces path update#
Update HDFS parameter configuration description
Templeton data on hdfs needs to be readable by all usersĬontent of /user/templeton dir-bash-3.2$ hadoop dfs -ls /user/templetonFound 4 items-rw-r-r- 3 templeton hdfs 107373 19:53 /user/templeton/hadoop-streaming.jar-rw- 3 templeton hdfs 35352096 19:53 /user/templeton/-rw- 3 templeton hdfs 47909478 19:53 /user/templeton/-rw- 3 templeton hdfs 127652 19:53 /user/templeton/ugi.jarOnly templeton user can use the jars Nagios Admin Contact should be checked to ensure it is always an email address
Make disk cleaner on mac clean without requesting permission to acces path password#
Make the Naigos password (and re-type password) different so as you cause a validation error.This will let the user move on to the next screen by ignoring all other validation errors on this page. Init Wizard: Advanced Config validation errors can be bypassed Maybe we can hide the heading when no such service is present. On the manage services page we have a section for client only services that have no long running processes.If the user has no such component there is a heading but no content. Manage services section will have any empty section when no client only components installed
In /etc/hive/conf/hive-env.sh add -export HIVE_AUX_JARS_PATH=/usr/lib/hcatalog/share/hcatalog/hcatalog-0.4.0.jarThe user defined authorization class should implement interface .ql.
In /etc/hive/conf/hive-site.xmlUpdate hive config to enable authorization htmli and such that would be nice to clean up. html files whose extension should be changed to. There's some images in the html/ directory (that should be in images/) there's. There's a bunch of (temporary playground) files that got wrongly committed to the HMC codebase so this is to remove all of them and get things into a cleaner state (at least on the face of things).Ĭurrently there is no support for uninstall of mysql package. Like the title says: overwrite index.php with the contents of clusters.php making sure to add a link to the AddNodesWizard as well. We need to support data cleanup so that a cluster can be re-installed in case of failures. Support data cleanup if installation fails. Refactor puppet kick loop to easily change retries and timeouts. Puppet agent timeout should be increases as sometimes (possibly due to a bug in ruby) puppet master takes long time to compile and send back the catalog.
Make the daemon names and other field names consistentįollowing names need to be consistent: Hdfs -> HDFS Mapreduce -> MapReduce Zookeeper -> ZooKeeper HADOOP -> HadoopĮnable LZO should show checkbox instead of textĬurrently the enable lzo option shows a text box that needs to be filled with true/false. Remove the word alert from all the Nagios alerts descriptions. So Cpation of the table should change indicating the corresponding service name. Support filtering hosts based on discovery statusĪpi to get hosts should allow filtering out bad hostsĪlert table on the right needs to be tied visually/verbally to the context/content it is displayingĬurrently in Dashboard when service entry is clicked right side alerts table caption does not indicate that it is showing alerts related to that service. PHP Notice: Undefined variable: manifest in /usr/share/hmc/php/puppet/genmanifest/hostsConfig.php This causes problems during service start in later stages.
Make disk cleaner on mac clean without requesting permission to acces path install#
Remove import of mysql puppet module from manifest because this module is deprecated.Īdd check to verify jdk path after installĪfter the jdk install we do not validate the path. Remove import of mysql puppet module from manifest. Hcat client should be installed at the templeton server node. There is no checks for safemode when we reconfigure the cluster. We should move it to .rest.agent to exclude agent entities to expose to public. Hence schema xsd generation for public rest api will include agent entities beans. Remove agent entities beans from public schema xsdĪgent entities beans are located ambari-client:. If the heartbeat was not received correctly by the controller then it should retry. Preserve cluster id blueprint name and blueprint revision in agent local diskĪfter agent execute commands on behave of the controller agent should store cluster id blueprint name and blueprint revision in the local disk to keep track of the current deployment state on the agent.Īgent should retry heartbeat message if controller did not receive the heartbeat
It would be nice to have transition state: STARTING and STOPPING. For running a server (daemon) the current states are: START and STARTED.