ambari-issues mailing list archives

Site index · List index
Message view « Date » · « Thread »
Top « Date » · « Thread »
From "Andrew Onischuk (JIRA)" <j...@apache.org>
Subject [jira] [Commented] (AMBARI-14926) ambari cached hdfs.py conflicts with python hdfs lib resulting into monitoring errors
Date Fri, 25 Mar 2016 11:54:25 GMT

    [ https://issues.apache.org/jira/browse/AMBARI-14926?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=15211730#comment-15211730
] 

Andrew Onischuk commented on AMBARI-14926:
------------------------------------------

There is 20 instances hanging arnd on Hadoop QA. 
Without and with my patch 1 test fails:
{noformat}
Failed tests:
ERROR: test_get_pxf_protocol_version (test_alerts_api_status.TestAlertsApiStatus)
----------------------------------------------------------------------
Traceback (most recent call last):
  File "/home/user/ambari/ambari-common/src/test/python/mock/mock.py", line 1199, in patched
    return func(*args, **keywargs)
  File "/home/user/ambari/ambari-server/src/test/python/stacks/2.3/PXF/test_alerts_api_status.py",
line 51, in test_get_pxf_protocol_version
    version = api_status._get_pxf_protocol_version()
TypeError: _get_pxf_protocol_version() takes exactly 1 argument (0 given)
{noformat}

No new tests are failing.

> ambari cached hdfs.py conflicts with python hdfs lib resulting into monitoring errors
> -------------------------------------------------------------------------------------
>
>                 Key: AMBARI-14926
>                 URL: https://issues.apache.org/jira/browse/AMBARI-14926
>             Project: Ambari
>          Issue Type: Bug
>          Components: ambari-agent
>    Affects Versions: 2.2.0
>         Environment: Linux CentOS 7 x86_64
>            Reporter: matteo brancaleoni
>            Assignee: Andrew Onischuk
>         Attachments: AMBARI-14926.patch
>
>
> On systems with datanode and journal node monitored by ambari-agent,
> if in the standard python path the hdfs lib is installed ( https://pypi.python.org/pypi/hdfs/
) it makes datanode + journalnode monitoring fail:
> * datanodes are seen up in global live datanodes
> * but the service is down into the host service list (restart will work, but are marked
as stopped)
> looking into the logs I get:
> DEBUG 2016-02-04 16:54:49,160 PythonReflectiveExecutor.py:47 - Running command reflectively
['/usr/bin/python2',
>  u'/var/lib/ambari-agent/cache/common-services/HDFS/2.1.0.2.0/package/scripts/journalnode.py',
>  'SECURITY_STATUS',
> --
>  '/var/lib/ambari-agent/tmp']
> DEBUG 2016-02-04 16:54:49,170 PythonReflectiveExecutor.py:61 - Reflective command failed
with exception:
> Traceback (most recent call last):
>   File "/usr/lib/python2.6/site-packages/ambari_agent/PythonReflectiveExecutor.py", line
55, in run_file
>     imp.load_source('__main__', script)
>   File "/var/lib/ambari-agent/cache/common-services/HDFS/2.1.0.2.0/package/scripts/journalnode.py",
line 30, in <module>
>     from utils import service
>   File "/var/lib/ambari-agent/cache/common-services/HDFS/2.1.0.2.0/package/scripts/utils.py",
line 37, in <module>
>     from zkfc_slave import ZkfcSlave
>   File "/var/lib/ambari-agent/cache/common-services/HDFS/2.1.0.2.0/package/scripts/zkfc_slave.py",
line 21, in <module>
>     from hdfs import hdfs
> ImportError: cannot import name hdfs
> I think that ambari is including its cache files after the python path, so if an hdfs
module is present into python path makes the scripts goes crazy.
> (basically using the system hdfs py module and not his own)
> maybe cache files path must be included before the python path or imported with local
imports?



--
This message was sent by Atlassian JIRA
(v6.3.4#6332)

Mime
View raw message