I am running Redhat Enterprise 5.5 on my systems. I currently upgraded to 4.2.5 with no problem. Today I tried to update a server with the Splunk 4.3. I also updated the *Nix app (4.5) on this server at the same time. When I go to start splunk I get this error after site checks and during the config checks. From reading so far, it could be a python error of some sort?
ImportError: No module named site
I noticed this error on the attempt to upgrade 4.2.5 to 4.3:
error: unpacking of archive failed on file /opt/splunk/lib/python2.7/site-packages/splunk/appserver/mrsparkle/locale/zh_CH/LC_MESSAGES/messages.po;4f0efb70: cpio: read
It looks like there is an integrity problem with previous RPM file "splunkforwarder-4.3-115073-linux-2.6-x86_64.rpm" I downloaded on the 11th of January 2011
Trying to extract the file using rpm2cpio indicates the file was truncated:
rpm2cpio splunkforwarder-4.3-115073-linux-2.6-x86_64.rpm |cpio -idmv
cpio: premature end of file
I redownloaded the file from Splunk and compared md5sums, and it showed that the previous version I downloaded was corrupt. The newly downloaded file extracted correctly.