hostname and canonical name for this host are not consistent when checked from java process

On Thu, Jun 13, 2013 at 8:10 AM, Stefan wrote:

Hello!

I receive the above mentioned error despite being quite sure my
configuration is ok.

I have a DNS service up and running. Normal resolving and loopback. tested
and ok.
Additionally I have configured /etc/hosts with all IP Adresses
(e.g. 192.168.1.160 romeo0.lan romeo0)

The full error message is:
"The hostname and canonical name for this host are not consistent when
checked from a Java process."



hostname
=> romeo0

hostname -f
=> romeo0.lan

host -v -t A romeo0.lan
=>

;; ->>HEADER<<- opcode: QUERY, status: NOERROR, id: 37626


dig romeo0.lan

On Thursday, June 13, 2013 3:13:16 PM UTC+2, bdb wrote:

Stefan,

You're in luck, I just had this same question and one of our developers
(A.B.) provided some detailed information on this:

The DNS resolution check uses the Java InetAddress class (much like what
Hadoop daemons typically do) to:
1. get the local address (addr = InetAddress.getLocalHost()) and uses the
retrieved address to
2. get the hostname (addr.getHostName()) and the FQDN / canonicalHostName
(addr.getCanonicalHostName())
3. Verify that 'hostname' can be used to derive 'canonicalHostName' by
using the info in /etc/resolv.conf
-- 3a. If /etc/resolv.conf has an entry of the form "domain
isp.domain.name", 'hostname'.isp.domain.name should be
'canonicalHostName' for the check to pass.
-- 3b. If /etc/resolv.conf has an entry of the form "search domain1.name
domain2.name .. domainN.name", 'canonicalHostName' should be one of
'hostname'.domain1.name, 'hostname'.domain2.name .. or
'hostname'.domainN.name

Here are a couple of things to check/verify:
- Check /etc/resolv.conf on the hosts that are not 'healthy'. See what the
domain and search lines are.romeo0.lan is not part of /etc/resolv.conf on
those hosts and/or different from what it is on the 'healthy' hosts.
- Look for an INFO message of the form ""hostname <hostname> differs from
the canonical name <canonicalname>" in the agent logs
(/var/log/cloudera-scm-agent/*.log) on the unhealthy hosts. That should
give you the exact names being used for the check performed in step 3 above.

Please let us know if this helps to determine the source of the health
check failure, or if you have any questions or need clarification on
anything.

Thanks,

*Brian Burton*
*Customer Operations Engineer*
<http://www.cloudera.com>


On Thu, Jun 13, 2013 at 8:10 AM, Stefan <stefan...@gmail.com <javascript:>

On Thu, Jun 13, 2013 at 10:06 AM, Stefan wrote:

Brian,
thanks for your reply. I checked the logs and found that the java process
reads:
hostname = romeo
canonical name = romoe0.lan

Yes I am aware of this and I believe this is correct. I configured the
hostname as a non fully qualified name according to instructions here:

http://www.cloudera.com/content/cloudera-content/cloudera-docs/CDH4/4.2.0/CDH4-Installation-Guide/cdh4ig_topic_11_1.html

Of course I can rename the hostname to romeo0.lan

Should I do that? Still a bit confused, maybe I am missing something here.

btw: I have no search or domain entries in resolve.conf because I assumed
they are optinal.

Stefan

On Thursday, June 13, 2013 5:34:16 PM UTC+2, bdb wrote:

Stefan,

Your setup is correct. I'll check to see what the behavior is with no
entries in /etc/resolv.conf and get back to you.

Thanks,

*Brian Burton*
*Customer Operations Engineer*
<http://www.cloudera.com>


On Thu, Jun 13, 2013 at 10:06 AM, Stefan <stefan...@gmail.com<javascript:>

On Thu, Jun 13, 2013 at 12:26 PM, Stefan wrote:

Hi Brian,

I changed the hostename in etc/sysconfig/network to the FQDN (romeo0.lan)
and now it works.

I dont know, maybe the support page I was referring to can be
misunderstood.

Thanks for helping me with all the infos that allowed me to solve the
issue.
Stefan

On Friday, June 14, 2013 7:57:55 AM UTC-7, bdb wrote:

Stefan,

Glad you were able to get it working. Thanks for closing the loop and
letting us know the fix.

*Brian Burton*
*Customer Operations Engineer*
<http://www.cloudera.com>


On Thu, Jun 13, 2013 at 12:26 PM, Stefan <stefan...@gmail.com<javascript:>

On Friday, 14 June 2013 16:57:55 UTC+2, bdb wrote:

Stefan,

Glad you were able to get it working. Thanks for closing the loop and
letting us know the fix.

*Brian Burton*
*Customer Operations Engineer*
<http://www.cloudera.com>


On Thu, Jun 13, 2013 at 12:26 PM, Stefan <stefan...@gmail.com<javascript:>

On 30 June 2013 12:25, Manal Helal wrote:

Hi,

I am having the same problem in both ubunto and CentOS data nodes, and I
don't have a "search" line in the /etc/resolv.conf in the problematic
nodes, so I added it because it is the only different line between the
healthy and unhealthy nodes.

# Generated by NetworkManager
search 8.8.4.4 CentOS
nameserver 8.8.8.8


Is there a way to solve it permanently, it comes back, because the file
keeps getting re-written,


thank you very much,

On Mon, Jul 1, 2013 at 2:50 PM, Philip Langdale wrote:

As it says, NetworkManager generates that file. You'd want to configure
network manager to use a specific server and search domain, etc.

--phil