Hi,
Our DS server is getting crashed very frequently, after analysis we found that disk space is eaten by datasets (huge size), since we have used them heavily in our parallel jobs.
My Question: Is there any way to delete datasets once the correp. job is processed successfully? or what is the ideal way to deal with this situation?
As work around we have written a shell script and scheduled it in corn, which delete all dataset files older than 2 days.
Auto Purging of Dataset
Moderators: chulett, rschirm, roy
Auto Purging of Dataset
Thanks,
ds_avatar
ds_avatar
I think that a periodical cleansing or a preload cleansing will also leave your data sets available for some debuging in case you need them.
Roy R.
Time is money but when you don't have money time is all you can afford.
Search before posting:)
Join the DataStagers team effort at:
http://www.worldcommunitygrid.org
Time is money but when you don't have money time is all you can afford.
Search before posting:)
Join the DataStagers team effort at:
http://www.worldcommunitygrid.org
I also thought to use execute command in job sequence, but what will be the selection criteria to pick-up the datasets belong to corresponding job.pandeesh wrote:Once the job is finished , delete them using orchadmin in a after job subroutine or using execute command activity in a sequence.
How come dataset can be used for debug when it is available in non-readable format.roy wrote:I think that a periodical cleansing or a preload cleansing will also leave your data sets available for some debuging in case you need them.
Thanks,
ds_avatar
ds_avatar
Just pass the parameter for dataset name as what you do in job level.(else hard code if you are hard coding in job level)ds_avatar wrote:I also thought to use execute command in job sequence, but what will be the selection criteria to pick-up the datasets belong to corresponding job.
It can be as Data set management option is there in the client tool.ds_avatar wrote:How come dataset can be used for debug when it is available in non-readable format.
IHTH(I Hope This Helps)
pandeeswaran