On Thu, 2013-02-28 at 10:16 -0800, r.yeo wrote: > Architecture of PM is Ubuntu 12.04 in a VM with 8GB of RAM (4.7GB > used) with Apache 2.2 as the front end. > > No network bottlenecks that I have been able to identify and the nodes > this is happening on seem to be random. Different buildings on > campus, different vlans, etc. I have nodes on the same network > segment getting identical configs that are performing flawlessly. > > recurse => true is not being called in any module and no directories > in my file manifests. > > My typical module with files has about 5 files and they are all > smaller then 5KB. > > I'll keep digging and see if I can find any similarities on the client > machines that might be causing this. It really seems client side at > this point since looking in dashboard it does seem to be very > consistent on the nodes that are having issues and the nodes that are > good even though they are getting identical configs.
Could it be that you try to impose a larger concurrency (ie number of puppet agents trying to access your puppetmaster) than what is configured in your passenger setup? Like if you have 30 agents trying to access the master at the same time, but you can accomodate only 25, the other 5 will have to wait for every request they're sending to your master VM. -- Brice Figureau Follow the latest Puppet Community evolutions on www.planetpuppet.org! -- You received this message because you are subscribed to the Google Groups "Puppet Users" group. To unsubscribe from this group and stop receiving emails from it, send an email to [email protected]. To post to this group, send email to [email protected]. Visit this group at http://groups.google.com/group/puppet-users?hl=en. For more options, visit https://groups.google.com/groups/opt_out.
