Archiving can be triggered by doing the following steps:
Because archiving does not happens immediately the status (called 'Archiving Status' in data set tables) of the data sets will be changed to BACKUP_PENDING or ARCHIVE_PENDING.
To make archived data sets available again repeat the steps, but choose 'Unarchive'.
If you want to disallow archiving, choose 'Lock'. Remember that you can do this only for available data sets. The 'Archiving Status' will change to 'AVAILABLE (LOCKED)'. To make archiving possible again, choose 'Unlock'.
Instead of triggering archiving only requesting archiving is possible. The maintenance task ArchivingByRequestTask is required. It triggers the actual archiving.
Archiving can be automated by the Auto Archiver. This is a maintenance task which triggers archiving of data sets fullfulling some conditions (e.g. not accessed since a while). Note that the auto archiver doesn't archives itself. It just automates the selection of data sets to be archived. For all configuration parameters see AutoArchiverTask.
An archiving policy selects from the unarchived data sets candidates (which are either data sets not accessed since some days or data sets marked by a tag) the data sets to be archived. If not specified all candidates will be archived.
The policy can be specified by
policy.class property. It has to be the fully-qualified name of a Java class implementing
ch.systemsx.cisd.etlserver.IAutoArchiverPolicy. All properties starting with
policy. specifying the policy further.
Description: Policy which tries to find a group of data sets with a total size from a specified interval. This is important in case of Multi Data Set Archiving. Grouping can be defined by space, project, experiment, sample, data set type or a combination of those. Groups can be merged if they are too small. Several grouping keys can be specified.
Searching for an appropriate group of data sets for auto archiving is logged. If no group could be found an admin is notified via email (email address specified in
log.xml). The email contains the searching log.
|minimal-archive-size||The total size (in bytes) of the selected data sets has to be equal or the larger than this value. Default: 0|
The total size (in bytes) of the selected data sets has to be equal or the less than this value. Default: Unlimited
Comma separated list of grouping keys. A grouping key has the following form:
A basic key is from the following vocabulary:
class = ch.systemsx.cisd.etlserver.plugins.AutoArchiverTask interval = 10 days archive-candidate-discoverer.class = ch.systemsx.cisd.etlserver.plugins.TagArchiveCandidateDiscoverer archive-candidate-discoverer.tags = /admin-user/archive policy.class = ch.systemsx.cisd.etlserver.plugins.GroupingPolicy policy.minimal-archive-size = 30000000000 policy.maximal-archive-size = 150000000000 policy.grouping-keys = Space#DataSetType, Experiment#Sample:merge
In this example the candidates are unarchived data sets which have been tag by the user
admin-user with the tag
archive. The policy tries to find a group of data set with total size between 30 Gb and 150 Gb. It first looks for groups where all data sets are of the same type and from the same space. If no group is found it tries to find groups where all data sets are from the same experiment and sample (data set with no samples are assigned to
no_sample). If no matching groups are found and at least two groups are below the minimum the policy tries to merge groups to a bigger group until the bigger group match the size condition. If no group can be found an email will be sent describing in detail the several steps of finding a matching group.