Issue #12623 has been updated by Chris Price.
Confirmed that the use_srv_records setting also alleviates this problem. In my acceptance testing environment, I logged into my CentOS6 node and changed the /etc/sysconfig/network line back to its original value: <pre> HOSTNAME=pe-centos6.localdomain </pre> Rebooted, and 'hostname' now returns a pseudo-fqdn again: <pre> [root@pe-centos6 ~]# hostname <b>pe-centos6.localdomain</b> [root@pe-centos6 ~]# </pre> Facter interprets this as the fqdn again: <pre> [root@pe-centos6 ~]# facter fqdn pe-centos6.localdomain [root@pe-centos6 ~]# </pre> And running this command: <pre> time puppet agent --test --debug </pre> regularly takes over three minutes to execute. It very visibly hangs on this line: <pre> debug: Searching for SRV records for domain: localdomain </pre> at least twice during each execution of the agent. If I then edit the /etc/puppet/puppet.conf and adding the line: <pre> use_srv_records=false </pre> and then run the agent, the agent execution time has dropped from over 3 minutes to less than 5 seconds. ---------------------------------------- Bug #12623: Long timeout for SRV DNS rsolution https://projects.puppetlabs.com/issues/12623#change-54876 Author: Chris Price Status: Investigating Priority: Normal Assignee: Chris Price Category: network Target version: Affected Puppet version: development Keywords: Branch: This issue has come up for me several times when setting up nodes for acceptance tests. It is probably most easily reproducible with a fresh copy of our CentOS6 VM. In /etc/sysconfig/network there is a line that looks like this: <pre> HOSTNAME=pe-centos6.localdomain </pre> This causes facter to return that same string for "fqdn". Then, when you attempt to use this VM as an agent node for acceptance tests, it blocks for an *extremely* long time during the 03_ValidateSignCerts phase. Running the offending commands outside of the acceptance test framework: <pre> puppet master --dns_alt_names="puppet,$(hostname -s),$(hostname -f)" --verbose --no-daemonize --logdest=/var/lib/puppet/log/puppetmaster.log --debug --trace </pre> and <pre> puppet agent --test --debug </pre> Will reproduce the slowness, and you'll see the following in the agent output: <pre> debug: Searching for SRV records for domain: localdomain debug: Found 0 SRV records for: _x-puppet-report._tcp.localdomain </pre> There may be a delay of 5 minutes or more in between those two lines being printed, however. The code that this occurs in is in lib/puppet/network/resolver.rb, in the method each_srv_record. However, this code is simply calling into the ruby Resolve::DNS.getresources() method. Reviewing the documentation for this method, I don't see a way to specify a timeout interval. I'm also not entirely sure whether reducing the timeout is the correct solution, or if there is some other alternative. As a workaround, simply fixing the "HOSTNAME" line in /etc/sysconfig/network so that it does not contain the domain name seems to dramatically improve the performance. -- You have received this notification because you have either subscribed to it, or are involved in it. To change your notification preferences, please click here: http://projects.puppetlabs.com/my/account -- You received this message because you are subscribed to the Google Groups "Puppet Bugs" group. To post to this group, send email to [email protected]. To unsubscribe from this group, send email to [email protected]. For more options, visit this group at http://groups.google.com/group/puppet-bugs?hl=en.
