Testing MEDIATEX (see Conceptual model) with about 50,000 archives.
Download the Open Icons Library and build the related catalog and extraction rules files. This last about 15 minutes.
If not already done, please configure the
mdtx-hello collection: see Scenario 2.
Grow the cache size by modifying /etc/mediatex/mdtx.conf
# default cache parameters cacheSize 1 Go
Empty the catalog and extraction meta-data files:
$ cat /dev/null > ~mdtx-hello/git/catalog000.txt $ cat /dev/null > ~mdtx-hello/git/extract000.txt $ mediatex upgrade+
Uploading the icons and build the HTML catalog:
$ mediatex upload+ file open_icon_library-standard.tar.gz \ catalog icons.cat rules icons.ext to coll hello
Archive are now injected. You can browses the icons catalog at this URL:
mdtx login and the password you provide before. Now MEDIATEX
(see Conceptual model) should be able to help you conserving or migrating theses archives.
The audit process show bellow is not requiered but only provides a way to double check all goes wright (MEDIATEX (see Conceptual model) already do that by providing a score for collections).
It take time to extract archives from this big TAR container. To speed up the audit process that simulate all archive retrieval, you can first manually extract all archives:
# /etc/init.d/mediatexd stop $ mediatex su $ cd ~mdtx/cache/mdtx-hello $ tar -zxf 2016-06/OpenIconsLibrary.tar.gz # /etc/init.d/mediatexd start
The server will take too long time to scan the cache directory when (re-)starting. (This should become optional in further versions). This last about 4 minutes.
Audit the collection:
$ mediatex audit coll hello for email_address $ mediatex srv extract # tail -f /var/log/mediatex
You should be notified by mail that the audit report is uploaded to the collection. The audit process is not optimized and probably will not. This last about 4 hours.