[Spacewalk-list] Proxy throws http-error 500 while preparing Sles Upgrade SP3 to SP4

[Bosch, Fabian (BITBW)]

HI

I'm trying to Upgrade SLES to Patchlevel 4 via Spacewalk. Let's leave aside, that the Migrationpatch delivered by SUSE needs some crazy workarounds concerning the release-key. (you can visit https://forums.suse.com/showthread.php?6674-Cannot-update-SLES-11-SP3&p=29121#post29121 if you are interested)

My problem is, that I'm not able to install all update-packages at once for SP3 to prepare the system for the migration via Spacewalk-Proxy.
Sometimes, when I schedule all available patches (could be few hundreds), after some time, the zypper logs an error 500 at the proxy.

My guess was, that the squids cache reached its limit and doesn't delete the old entrys right. But neither the free space on the virtual hard drive nor the cache_dir's limit was reached.

Has anyone got some of these errors too? What logfiles should I view in addition to the following?


The errors I get on the proxys side

Squid access log is logging some TCP_MISS with the Spacewalk-Channels fpr Sles 11

Squid cache log says:


2015/08/21 13:21:44| Starting Squid Cache version 3.1.23 for x86_64-redhat-linux-gnu...

2015/08/21 13:21:44| Process ID 1585

2015/08/21 13:21:44| With 1024 file descriptors available

2015/08/21 13:21:44| Initializing IP Cache...

2015/08/21 13:21:44| ipcacheAddEntryFromHosts: Bad IP address '.27.0.0.1'

2015/08/21 13:21:44| DNS Socket created at [::], FD 7

2015/08/21 13:21:44| DNS Socket created at 0.0.0.0, FD 8

2015/08/21 13:21:44| Adding nameserver 10.xxx.xxx.130 from /etc/resolv.conf

2015/08/21 13:21:44| User-Agent logging is disabled.

2015/08/21 13:21:44| Referer logging is disabled.

2015/08/21 13:21:44| Unlinkd pipe opened on FD 13

2015/08/21 13:21:44| Local cache digest enabled; rebuild/rewrite every 3600/3600 sec

2015/08/21 13:21:44| Store logging disabled

2015/08/21 13:21:44| Swap maxSize 0 + 262144 KB, estimated 20164 objects

2015/08/21 13:21:44| Target number of buckets: 1008

2015/08/21 13:21:44| Using 8192 Store buckets

2015/08/21 13:21:44| Max Mem  size: 262144 KB

2015/08/21 13:21:44| Max Swap size: 0 KB

2015/08/21 13:21:44| Using Least Load store dir selection

2015/08/21 13:21:44| Current Directory is /

2015/08/21 13:21:44| storeDirWriteCleanLogs: Starting...

2015/08/21 13:21:44|   Finished.  Wrote 0 entries.

2015/08/21 13:21:44|   Took 0.00 seconds (  0.00 entries/sec).

FATAL: No port defined

Squid Cache (Version 3.1.23): Terminated abnormally.

CPU Usage: 0.022 seconds = 0.015 user + 0.007 sys Maximum Resident Size: 39632 KB Page faults with physical i/o: 0 Memory usage for squid via mallinfo():

        total space in arena:    3072 KB

        Ordinary blocks:         3035 KB     42 blks

        Small blocks:               0 KB      1 blks

        Holding blocks:          1104 KB      4 blks

        Free Small blocks:          0 KB

        Free Ordinary blocks:      36 KB

        Total in use:            4139 KB 135%

        Total free:                36 KB 1%

This is underlining my thought of a full cache - but as said neither the cache_dir limit was reached nor the hard drive is full.

Httpd error log says:


[Fri Aug 21 15:17:10 2015] [notice] suEXEC mechanism enabled (wrapper: /usr/sbin/suexec) [Fri Aug 21 15:17:10 2015] [warn] Init: (spacewalk.foo.my.de:443) You configured HTTP(80) on the standard HTTPS(443) port!

[Fri Aug 21 15:17:10 2015] [notice] Digest: generating secret for digest authentication ...

[Fri Aug 21 15:17:10 2015] [notice] Digest: done [Fri Aug 21 15:17:10 2015] [warn] Init: (spacewalk.foo.my.de:443) You configured HTTP(80) on the standard HTTPS(443) port!

[Fri Aug 21 15:17:10 2015] [notice] Apache/2.2.15 (Unix) DAV/2 mod_ssl/2.2.15 OpenSSL/1.0.1e-fips mod_wsgi/3.2 Python/2.6.6 configured -- resuming normal operations [Fri Aug 21 15:17:30 2015] [error] Spacewalk 1556 2015/08/21 15:17:30 +02:00: ('Error opening connection', 'http://spacewalk.foo.my.de/XMLRPC/GET-REQ/tools-opensuse13.1/repodata/repomd.xml?head_requests=no', error(111, 'Connection refused')) [Fri Aug 21 15:17:30 2015] [error] Exception reported from izegbpspw1.foo.my.de [Fri Aug 21 15:17:30 2015] [error] Time: Fri Aug 21 15:17:30 2015 [Fri Aug 21 15:17:30 2015] [error] Exception type <class 'socket.error'> [Fri Aug 21 15:17:30 2015] [error] [Fri Aug 21 15:17:30 2015] [error] Exception Handler Information [Fri Aug 21 15:17:30 2015] [error] Traceback (most recent call last):

[Fri Aug 21 15:17:30 2015] [error]   File "/usr/share/rhn/proxy/rhnShared.py", line 130, in _connectToParent

[Fri Aug 21 15:17:30 2015] [error]     self.responseContext.getConnection().connect()

[Fri Aug 21 15:17:30 2015] [error]   File "/usr/lib/python2.6/site-packages/rhn/connections.py", line 129, in connect

[Fri Aug 21 15:17:30 2015] [error]     HTTPConnection.connect(self)

[Fri Aug 21 15:17:30 2015] [error]   File "/usr/lib/python2.6/site-packages/rhn/connections.py", line 107, in connect

[Fri Aug 21 15:17:30 2015] [error]     httplib.HTTPConnection.connect(self)

[Fri Aug 21 15:17:30 2015] [error]   File "/usr/lib64/python2.6/httplib.py", line 742, in connect

[Fri Aug 21 15:17:30 2015] [error]     self.timeout)

[Fri Aug 21 15:17:30 2015] [error]   File "/usr/lib64/python2.6/socket.py", line 567, in create_connection

[Fri Aug 21 15:17:30 2015] [error]     raise error, msg

[Fri Aug 21 15:17:30 2015] [error] error: [Errno 111] Connection refused [Fri Aug 21 15:17:30 2015] [error]


But we can ignore the openSuse-channel because this channel has nothing to do with the migration and will be fixed separately.





Regards,

Fabian Bosch
Referat 34 - Open Source

IT Baden-Württemberg (BITBW)
Krailenshaldenstraße 44
70469 Stuttgart
Telefon: +49 711 8910-94476
Telefax: +49 711 8910-17575
E-Mail: fabian.bosch at bitbw.bwl.de<mailto:fabian.bosch at bitbw.bwl.de>
Internet: www.bitbw.de

-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://listman.redhat.com/archives/spacewalk-list/attachments/20150824/6f27fc28/attachment.htm>