[Spacewalk-list] Spacewalk Proxies failing to serve large packages

[Stephen Herr]

Oh, I bet you're running out of space in /tmp and hitting this behavior:
https://bugzilla.redhat.com/show_bug.cgi?id=1158193

Proxy 2.0 and on stores large files in /tmp instead of in-memory, since
people were having problems where they ran out of memory and then the
Out-Of-Memory killer would kill httpd. Proxy 2.3 adds an config option
that allows you to better control the in-memory vs on-disk size limit,
but the best solution is probably just to extend your /tmp partition.

-Stephen


On 05/11/2015 02:39 PM, Matthew Madey wrote:
> Thanks for the responses. No, there are no other proxies between the
> Spacewalk Proxy and client, and I don't have any timeout settings
> specified in rhn.conf, yum.conf, or up2date. 
> If I register the client directly to the Spacewalk server, I'm able to
> download the package without an issue.. so the problem definitely lies
> in the caching mechanism of the Spacewalk proxy. What concerns me is the
> below error I see on the Spacewalk Proxy when attempting to retrieve the
> package. 
> 
> [Fri May 08 18:42:28 2015] [error] [client 172.20.165.201] mod_wsgi
> (pid=9403): Exception occurred processing WSGI script
> '/usr/share/rhn/wsgi/xmlrpc.py'.
> [Fri May 08 18:42:28 2015] [error] [client 172.20.165.201] IOError:
> failed to write data
> 
> 
> 
> On Mon, May 11, 2015 at 11:21 AM, Michael Calmer <mc at suse.de
> <mailto:mc at suse.de>> wrote:
> 
>     Hi,
> 
>     Am Freitag, 8. Mai 2015, 17:52:29 schrieb Matthew Madey:
>     > I'm seeing an issue with packages 300MB and larger being served by
>     > Spacewalk proxies. When I run yum install on a client, it gets to
>     97% of
>     > the download or so and fails with the below message:
>     >
>     > Error Downloading Packages:
>     >  somepackage.x86_64: failed to retrieve
>     getPackage/somepackage.x86_64.rpm
>     > from rhel-x86_64-server-6-datacenter
>     > error was [Errno 14] PYCURL ERROR 18 - "transfer closed with
>     1499666 bytes
>     > remaining to read"
>     >
>     > Looking at the Spacewalk proxy, I see the below error:
>     >
>     > [Fri May 08 18:42:28 2015] [error] [client 172.20.165.201] mod_wsgi
>     > (pid=9403): Exception occurred processing WSGI script
>     > '/usr/share/rhn/wsgi/xmlrpc.py'.
>     > [Fri May 08 18:42:28 2015] [error] [client 172.20.165.201]
>     IOError: failed
>     > to write data
>     >
>     > Here's what I've tried so far with no luck:
>     >
>     > 1. Increase max_object_size parameter in squid.conf to 500MB
>     > 2. Turn on KeepAlive parameter in httpd.conf
>     > 3. Increase KeepAlive timeout in httpd.conf to 120 seconds
>     > 4. Increase Timeout in httpd.conf to 120 seconds
> 
>     What about the timeout parameter in rhn.conf (server and proxy) and
>     in up2date
>     conf (for yum - or was it in a special yum config file?)
> 
>     --
>     Regards
> 
>             Michael Calmer
> 
>     --------------------------------------------------------------------------
>     Michael Calmer
>     SUSE LINUX GmbH, Maxfeldstr. 5, D-90409 Nuernberg
>     T: +49 (0) 911 74053 0 <tel:%2B49%20%280%29%20911%2074053%200>
>     F: +49 (0) 911 74053575 <tel:%2B49%20%280%29%20911%2074053575>  -
>     e-mail: Michael.Calmer at suse.com <mailto:Michael.Calmer at suse.com>
>     --------------------------------------------------------------------------
>     SUSE Linux GmbH, GF: Felix Imendörffer, Jane Smithard, Dilip Upmanyu,
>                          Graham Norton, HRB 21284 (AG Nürnberg)
> 
> 
>     _______________________________________________
>     Spacewalk-list mailing list
>     Spacewalk-list at redhat.com <mailto:Spacewalk-list at redhat.com>
>     https://www.redhat.com/mailman/listinfo/spacewalk-list
> 
> 
> 
> 
> _______________________________________________
> Spacewalk-list mailing list
> Spacewalk-list at redhat.com
> https://www.redhat.com/mailman/listinfo/spacewalk-list
>