[chef] Re: Re: Re: Re: Re: Re: Re: Re: Re: Problem running chef-client, just from 2 days ago.

[Lamont Granquist < >]

The "End of File" is getting thrown because the socket connection gets 
torn down ("everything is a file" on unix (but not really, see plan9...)).

On 5/28/14, 11:35 AM, Adam Edwards wrote:
> Hello Carlos -- do you have the output of c:/chef/cache/chef-stacktrace.out
> from this last failure? And if it still repros, can you run with -l debug?
>
> The error implies some sort of invalid file that is being accessed --
> Lamont, can you tell the context from the output we have below?
>
> -Adam
>
> -----Original Message-----
> From: Carlos Camacho 
> [mailto:
>  
> Sent: Tuesday, May 27, 2014 2:55 AM
> To: Lamont Granquist
> Cc: Adam Edwards; 
>
>  
> Subject: Re: [chef] Re: Re: Re: Re: Re: Re: Re: Re: Problem running
> chef-client, just from 2 days ago.
>
> Something really odd just happened.
>
> I restarted the chef server and run the chef-client from Windows and it
> worked fine... A few seconds later i executed the same command again and i
> had the error.. o.O
>
> Check this out....
>
>
> PS C:\Users\Administrator> chef-client
> Starting Chef Client, version 11.12.4
> [2014-05-27T09:43:31+00:00] INFO: *** Chef 11.12.4 ***
> [2014-05-27T09:43:31+00:00] INFO: Chef-client pid: 3952
> [2014-05-27T09:43:39+00:00] INFO: Run List is [role[windows]]
> [2014-05-27T09:43:39+00:00] INFO: Run List expands to []
> [2014-05-27T09:43:39+00:00] INFO: Starting Chef Run for slwnone10
> [2014-05-27T09:43:39+00:00] INFO: Running start handlers
> [2014-05-27T09:43:39+00:00] INFO: Start handlers complete.
> [2014-05-27T09:43:39+00:00] INFO: HTTP Request Returned 404 Object Not
> Found:
> resolving cookbooks for run list: []
> [2014-05-27T09:43:39+00:00] INFO: Loading cookbooks [] Synchronizing
> Cookbooks:
> Compiling Cookbooks...
> Converging 0 resources
> [2014-05-27T09:43:39+00:00] INFO: Chef Run complete in 0.561614 seconds
>
> Running handlers:
> [2014-05-27T09:43:39+00:00] INFO: Running report handlers Running handlers
> complete [2014-05-27T09:43:39+00:00] INFO: Report handlers complete Chef
> Client finished, 0/0 resources updated in 11.029288 seconds
>
>
>
>
> PS C:\Users\Administrator> chef-client
> Starting Chef Client, version 11.12.4
> [2014-05-27T09:43:51+00:00] INFO: *** Chef 11.12.4 ***
> [2014-05-27T09:43:51+00:00] INFO: Chef-client pid: 3580
> [2014-05-27T09:43:59+00:00] INFO: Run List is [role[windows]]
> [2014-05-27T09:43:59+00:00] INFO: Run List expands to []
> [2014-05-27T09:43:59+00:00] INFO: Starting Chef Run for slwnone10
> [2014-05-27T09:43:59+00:00] INFO: Running start handlers
> [2014-05-27T09:43:59+00:00] INFO: Start handlers complete.
> [2014-05-27T09:43:59+00:00] INFO: HTTP Request Returned 404 Object Not
> Found:
> resolving cookbooks for run list: []
> [2014-05-27T09:43:59+00:00] INFO: Loading cookbooks [] Synchronizing
> Cookbooks:
> Compiling Cookbooks...
> Converging 0 resources
>
> Running handlers:
> [2014-05-27T09:43:59+00:00] ERROR: Running exception handlers Running
> handlers complete [2014-05-27T09:43:59+00:00] ERROR: Exception handlers
> complete [2014-05-27T09:43:59+00:00] FATAL: Stacktrace dumped to
> c:/chef/cache/chef-stacktrace.out Chef Client failed. 0 resources updated in
> 10.670475 seconds [2014-05-27T09:43:59+00:00] FATAL: EOFError: end of file
> reached
>
> On Tue, May 27, 2014 at 10:54 AM, Carlos Camacho
> <
>  >
>  wrote:
> > Hello Lamont, Adam.
> >
> > I'm finishing some tests and the PG error is documented here
> > (https://tickets.opscode.com/browse/CHEF-4390).
> >
> > I have a new clue, so please if you can tell me what should be the
> > differences between these two types of connections.
> >
> > If i connect the chef node (Windows) using pptpd and the server, it works
> > fine.
> > If i connect the chef node (Windows)  and the server using the
> > standard network, it fails.
> >
> > Cheers!
> >
> >
> > On Wed, May 21, 2014 at 8:32 PM, Lamont Granquist 
> > <
> >  >
> > wrote:
> > > The postgresql logs look interesting.  The failed unique constraints
> > > might be the problem.  I would expect to see stack traces in the
> > > erchef logs though.
> > >
> > >
> > > On 5/20/14, 12:34 AM, Carlos Camacho wrote:
> > > > Hello,
> > > >
> > > > Adam, Lamont good morning.
> > > >
> > > > Ill give it a try to CHEF-5270 because this error is recent (I don't
> > > > loose any trying as i'm stuck right now) about 2 weeks ago it
> > > > suddenly stop working. Also I'm attaching the chef-client logs in
> > > > debug mode and the chef-server logs. But again i don't see any error
> > > > in the application layer.
> > > >
> > > > Thanks for your help,
> > > >
> > > > Cheers!
> > > >
> > > >
> > > > On Mon, May 19, 2014 at 9:15 PM, Lamont Granquist
> > > > <
> > > >  >
> > > > wrote:
> > > > > On 5/19/14, 12:24 AM, Carlos Camacho wrote:
> > > > > > Hello Julian and Lamont,
> > > > > >
> > > > > > I just check all logs in the server and there is not any apparent
> > > > > > error... It so confusing and frustrating this issue..
> > > > > Well, the server is actually tearing down the connection normally
> > > > > with a
> > > > > FIN|ACK, but then the client is still sending data, which produces
> > > > > FIN|a RST
> > > > > from the server, then the client receives those and sends back
> > > > > FIN|ACK and RST of its own.  So TCP is behaving normally and it
> > > > > does look like this isn't a rackspace issue.  The question is why
> > > > > the server seems to tear down the connection early while the client
> > > > > is still attempting to send data.
> > > > >
> > > > > There should be something either in the nginx or erchef logs on the
> > > > > server on in the -l debug logs or the stacktrace on the client.