Created on 02-05-2016 03:19 PM - edited 09-16-2022 03:02 AM
Hi all,
I'm a new Cloudera user, I'm installing a small cluster of CDH-5-5.1.11 on different hosts on Cent OS 7.2.
In the automatic process, during parcel installation stage, I get "urlopen error timed out" on all hosts excepted the CLoudera manager server.
Everything has been checked (firewall is ok, files can be downloaded on the hosts directly from the internet or from the manager with direct curl). I can't find a relevant solution on the forum.
Any ideas
Details below :
- part files stay at 0 in the cache directory on the hosts
- cloudera agent log file gives :
[06/Feb/2016 00:16:04 +0000] 11288 Thread-13 downloader INFO Starting download of: http://gw.domain.com:7180/cmf/parcel/download/CDH-5.5.1-1.cdh5.5.1.p0.11-el7.parcel
[06/Feb/2016 00:16:04 +0000] 11288 Thread-13 downloader INFO /opt/cloudera/parcel-cache/CDH-5.5.1-1.cdh5.5.1.p0.11-el7.parcel.part exists, add Range header: bytes=0-
[06/Feb/2016 00:16:34 +0000] 11288 Thread-13 downloader ERROR Unexpected exception during download
Traceback (most recent call last):
File "/usr/lib64/cmf/agent/src/cmf/downloader.py", line 276, in _download
response = self.opener.fetch_url(request)
File "/usr/lib64/cmf/agent/src/cmf/downloader.py", line 69, in fetch_url
return urllib2.urlopen(request)
File "/usr/lib64/python2.7/urllib2.py", line 154, in urlopen
return opener.open(url, data, timeout)
File "/usr/lib64/python2.7/urllib2.py", line 431, in open
response = self._open(req, data)
File "/usr/lib64/python2.7/urllib2.py", line 449, in _open
'_open', req)
File "/usr/lib64/python2.7/urllib2.py", line 409, in _call_chain
result = func(*args)
File "/usr/lib64/python2.7/urllib2.py", line 1244, in http_open
return self.do_open(httplib.HTTPConnection, req)
File "/usr/lib64/python2.7/urllib2.py", line 1214, in do_open
raise URLError(err)
URLError: <urlopen error timed out>
[06/Feb/2016 00:16:34 +0000] 11288 Thread-13 downloader INFO Finished download [ url: http://gw.domain.com:7180/cmf/parcel/download/CDH-5.5.1-1.cdh5.5.1.p0.11-el7.parcel, state: exception, total_bytes: 0, downloaded_bytes: 0, start_time: 2016-02-06 00:16:04, download_end_time: , end_time: 2016-02-06 00:16:34, code: 600, exception_msg: <urlopen error timed out>, path: None ]
Created 02-10-2016 07:30 AM
Hi @cjervis,
Thank you for your proposal.
I checked the thread you mentionned when I got the issue but it's not related to my case. I tried to restart the agent with sudo as mentionned by the author but it didn't work.
Your post forced me to replay all the steps and I found the answer so I share to avoid an other beginner to find in this hole :
the hostname for the cloudera-server was initially configured as a public FQDN. After that, I changed it with an internal FQDN and all the other internal DNS stuff, but cloudera manager kept continuing to share this public IP with its agent but the firewall was blocking the trafic.
I found this in the download field of the agent log where the public FQDN was mentioned. I just had to restart everything and now it seems to be ok.
Thank you very much for your tip
Created 02-10-2016 03:24 AM
I tried several things but none was successful and I couldn't find an answer in the community. I feel alone...
Someone who could have an idea ?
Created 02-10-2016 05:36 AM
I'm not an expert but perhaps some further information may get you the help you need. What have you already tried to solve the issue that did not work for you? Also, what steps did you take during the installation?
The additional information helped another user such as this one who had a permissions issue.
Created 02-10-2016 07:30 AM
Hi @cjervis,
Thank you for your proposal.
I checked the thread you mentionned when I got the issue but it's not related to my case. I tried to restart the agent with sudo as mentionned by the author but it didn't work.
Your post forced me to replay all the steps and I found the answer so I share to avoid an other beginner to find in this hole :
the hostname for the cloudera-server was initially configured as a public FQDN. After that, I changed it with an internal FQDN and all the other internal DNS stuff, but cloudera manager kept continuing to share this public IP with its agent but the firewall was blocking the trafic.
I found this in the download field of the agent log where the public FQDN was mentioned. I just had to restart everything and now it seems to be ok.
Thank you very much for your tip
Created 02-10-2016 07:32 AM
Awesome! I'm happy to see you figured out the issue. Best of luck as you continue your project.