0

I'm not sure whether the problem is on client or server. All my machines are Ubuntu 12.04 x64. I'm on Digital Ocean, and I have to configure iptables to "survive".

Server Story

My Chef Server is on 2GB RAM machine with no swap. It's latest Chef 11 from omnibus. I tried accessing Web UI, and it was fine. But after a few hours, it was very slow. Something likely crashed, and probably runit restarted whatever that was. I discovered that my RAM was all used up, only 77mb was free.

I tried reconfigure chef server like this.

# this is /etc/chef-server/chef-server.rb

topology 'standalone'
api_fqdn 'chef.[mydomain]'
lb['fqdn'] = 'chef.[mydomain]'
nginx['server_name'] = 'chef.[mydomain]'
nginx['url'] = 'https://chef.[mydomain]'

# default was 25%
chef_solr['heap_size'] = 300

# default was 200 connections
postgresql['max_connections'] = 30

# default was 25%
postgresql['shared_buffers'] = '128MB'

# default was 50%
postgresql['effective_cache_size'] = '256MB'

That didn't help much. I now have 200-300mb RAM free, but chef server still has problem responding after a while. I also see that my chef client (daemon running every 30m) stops reporting eventually, and I'd find status like "last reported 5 hours ago".

Client Story

In my client log I'd find things like these:

Error connecting to https://chef.[mydomain]/nodes/db.[mydomain] - Connection timed out - connect(2)

My client has iptables configured as follows, just in case this might be a problem.

% iptables -L -v

Chain INPUT (policy DROP 6 packets, 676 bytes)
 pkts bytes target     prot opt in     out     source               destination
 3155 3260K system     all  --  any    any     anywhere             anywhere

Chain FORWARD (policy ACCEPT 0 packets, 0 bytes)
 pkts bytes target     prot opt in     out     source               destination

Chain OUTPUT (policy ACCEPT 1419 packets, 178K bytes)
 pkts bytes target     prot opt in     out     source               destination

Chain system (1 references)
 pkts bytes target     prot opt in     out     source               destination
    0     0 DROP       tcp  --  any    any     anywhere             anywhere             tcpflags: FIN,SYN,RST,PSH,ACK,URG/NONE
    0     0 DROP       tcp  --  any    any     anywhere             anywhere             tcpflags:! FIN,SYN,RST,ACK/SYN state NEW
    0     0 DROP       tcp  --  any    any     anywhere             anywhere             tcpflags: FIN,SYN,RST,PSH,ACK,URG/FIN,SYN,RST,PSH,ACK,URG
    0     0 ACCEPT     all  --  lo     any     anywhere             anywhere
   22  1848 ACCEPT     icmp --  any    any     anywhere             anywhere
  361 37759 ACCEPT     tcp  --  any    any     anywhere             anywhere             tcp dpt:ssh
 2766 3219K ACCEPT     all  --  any    any     anywhere             anywhere             state RELATED,ESTABLISHED

Bootstrapping New Node

I tried bootstrapping a new node, and it failed twice. First time simply timed out, just like the other client. Second time I had this:

SSL3_GET_SERVER_CERTIFICATE:certificate verify failed

So why are my clients timing out?

I've been on this for 3 days, and would appreciate some help.

1 Answer 1

0

This looks like a result of network issues with Digital Ocean, and not Chef's fault.

You must log in to answer this question.

Not the answer you're looking for? Browse other questions tagged .