Not sure what was going on. I typically have fewer than 10 hosts connected to this lab install which was hanging. I updated it to 3.2.2/1.13.3 on Monday and have debug logging on and haven't seen the issue reappear.
*shrug* ----- Original Message ----- From: "Stephen Benjamin" <[email protected]> To: "Foreman Users" <[email protected]> Sent: Tuesday, January 10, 2017 6:13:27 PM Subject: Re: [foreman-users] Web Components Stop Responding ----- Original Message ----- > From: "'Jason B. Nance' via Foreman users" <[email protected]> > To: "Foreman Users" <[email protected]> > Sent: Friday, January 6, 2017 11:41:55 AM > Subject: [foreman-users] Web Components Stop Responding > > Hello Everyone, > > On a Katello 3.2.1 / TFM 1.13.2 server (and noticed on previous versions as > well) I run into the situation where occasionally the web interface and API > stop responding. Attempts to load any page, click on any > button/link/whatever results in the browser just spinning and eventually > timing out. The only log message printed is: > > ==> /var/log/httpd/foreman-ssl_error_ssl.log <== > [Fri Jan 06 10:16:11.926220 2017] [ssl:warn] [pid 24448] [client > 172.16.246.240:56536] AH02227: Failed to set r->user to > 'SSL_CLIENT_S_DN_CN' > > (I see this frequently under normal conditions) > > The system is not under heavy loads during this freeze. Load average is > about 1.2 (on a 6 vCPU / 16G memory virtual server). The only process doing > noticeable work is mongodb. > > Puppet runs on clients also fail during this time because /etc/puppet/node.rb > <fqdn> fails with: > > Could not send facts to Foreman: Net::ReadTimeout > > (no log messages are printed) > > A 'yum install' on a client is able to download the package, but then hangs > on "Uploading Package Profile" (and eventually times out and skips it). > > If I restart [tomcat and] httpd things start responding again. I'm not sure > if the tomcat restart matters, as I did it first and nothing changed but > after I restarted httpd - which took much longer than normal - is when > everything came back to life. There were about 55 connections to httpd at > the time, nearly all in CLOSE_WAIT. > > Anyone have ideas as to what may be happening? This sounds an awful lot like what we fixed in http://projects.theforeman.org/issues/14023. Do you have a lot of puppet clients checking in? You could try to tweak the passenger settings as described here in the Red Hat BZ (will be the default in Foreman 1.14), and see if that helps: https://bugzilla.redhat.com/show_bug.cgi?id=1163452#c11 - Stephen > Thanks, > > j > > -- > You received this message because you are subscribed to the Google Groups > "Foreman 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 https://groups.google.com/group/foreman-users. > For more options, visit https://groups.google.com/d/optout. > -- You received this message because you are subscribed to the Google Groups "Foreman 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 https://groups.google.com/group/foreman-users. For more options, visit https://groups.google.com/d/optout. -- You received this message because you are subscribed to the Google Groups "Foreman 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 https://groups.google.com/group/foreman-users. For more options, visit https://groups.google.com/d/optout.
