Member since
07-25-2017
18
Posts
3
Kudos Received
2
Solutions
My Accepted Solutions
Title | Views | Posted |
---|---|---|
4070 | 10-20-2017 02:52 PM | |
5641 | 07-26-2017 11:41 AM |
10-20-2017
02:52 PM
try include "someother.conf"
... View more
09-07-2017
02:33 PM
1 Kudo
1) You can find the parameter names on https://www.cloudera.com/documentation/enterprise/properties/5-12-x/topics/cm_props_cmserver.html 2) The parameters are pass through in director conf script. As long as you add the entries in cloudera manager configs section, it will be passed to cloudera manager. HTH
... View more
07-27-2017
04:37 PM
Per application log, the instance was allocated, private ip address assigned, but somehow PluggableComputeProvider complainsed that 0 instance found. It is because allocate and find are two consecutive calls in director, though allocation succeeded, later describe call can still return NOT_FOUND because of eventual consistency of EC2 calls. Retry will help. To guarantee success, you can set min count to 0, and if the instance is shown not created in director, modify cluster and select retry for the unsatisfied instance.
... View more
07-27-2017
02:10 PM
Sorry I apologize for the previous comment. In log: [2017-07-26 13:42:31] INFO [pipeline-thread-5] - c.c.director.aws.ec2.EC2Provider: << Instance i-0e91d6d581c37c4b5 got IP 10.16.113.60 [2017-07-26 13:42:31] INFO [pipeline-thread-5] - c.c.l.bootstrap.AllocateInstances: Waiting for 0 instances to start running It means that instance was successfully created(private ip address assigned), but a later describe failed to find the instance because of eventual consistency of EC2. Retry with the same conf should eventually succeed. To avoid failing the pipeline, you can set the minimum count to 0, and choose retry in modify cluster if the same problem happens again. Director is well aware of the problem, and is working uopn the it.
... View more
07-27-2017
02:01 PM
1 Kudo
You would like to check cm logs, /var/log/cloudera-manager-server, to see whether there is any trace there.
... View more
07-26-2017
11:41 AM
1 Kudo
It means that director failed to allocate instances required. There should be exceptions logged in application.log why the allocation was failed. You might be able to tell the reason from CloudTrail as well.
... View more
07-25-2017
05:19 PM
You can use API call: PUT /api/v2/users/{username}/password { "oldPassword" : "foo", "newPassword": "bar" } https://www.cloudera.com/documentation/director/latest/topics/director_server_user_management.html#concept_u24_yd2_xq
... View more