Datastage server file system full

Post questions here relative to DataStage Enterprise/PX Edition for such areas as Parallel job design, Parallel datasets, BuildOps, Wrappers, etc.

Moderators: chulett, rschirm, roy

Post Reply
priyadarshikunal
Premium Member
Premium Member
Posts: 1735
Joined: Thu Mar 01, 2007 5:44 am
Location: Troy, MI

Post by priyadarshikunal »

Is the file system full or getting filled when you run jobs?

It can get full even of you don't land your data on the server.

Those are like

&PH& - phantom logs
RT_LOGnnn - job logs
oracle logs in scratch disk
some leftover file in scratch disk like tsort*, lookup* which can consume a lot fo space.
Priyadarshi Kunal

Genius may have its limitations, but stupidity is not thus handicapped. :wink:
RAI ROUSES
Participant
Posts: 176
Joined: Wed Nov 15, 2006 5:48 am

Post by RAI ROUSES »

Hi Priyadarshi,
thanks for your help.

We have same big datasets from the test taht we are doing to parallel, i had try to removed them using the dataset managment option, but it return the message :
(DSD.GetScriptDir) Cannot open file RT_SCTEMP

How can i safe removed the dataset ?

The projects where we have created the datastets it's a project test, if we removed the project it also removed the datasets ?
after we removed the project we can manualy removed the datasets ?


Eache datastage server projects have 70M to 80M, for me it's normal.
I had clean the &PH& directory and the RT_LOG files, but the have a small size.

Where is defined the scratch disk ? What can i removed from them ?

merci
8) Rai 8)
mf_arts
Participant
Posts: 21
Joined: Wed Oct 19, 2005 5:10 am
Location: Dublin

Post by mf_arts »

Hi Rai,

The scratch disk is defined on configuration file as below example:
{
node "node1"
{
fastname "Servername"
pools ""
resource disk "/uX1/Datasets/" {pools ""}
resource scratchdisk "/uX2/Scratch/" {pools ""}
}
}

For a better performance scratch disk should be on a different resource and if you need to have several sorts also create an additional resource on file system. To change or create a configuration file :
Designer--> Tools-->Configuration File or either create a new or edit the default (system created with at least one node) then save and check (compilation of configuration file).

Regards
Mario Fernandes

An expert is one who knows more and more about less and less. - Nicholas Murray Butler
ray.wurlod
Participant
Posts: 54607
Joined: Wed Oct 23, 2002 10:52 pm
Location: Sydney, Australia
Contact:

Post by ray.wurlod »

Read the manual on orchadmin command. It has additional options that allow you to force things (such as moving/deleting Data Sets) to occur.
IBM Software Services Group
Any contribution to this forum is my own opinion and does not necessarily reflect any position that IBM may hold.
PhilHibbs
Premium Member
Premium Member
Posts: 1044
Joined: Wed Sep 29, 2004 3:30 am
Location: Nottingham, UK
Contact:

Re: Datastage server file system full

Post by PhilHibbs »

RAI ROUSES wrote:What we have to do to managed correctly the file system of datastage server ?
I recommend a disk usage analysis utility such as SpaceMonger - it's for Windows, so you'll have to point it at a SAMBA drive mapping. It will tell you where all your disk space has gone, unless it's in directories that the share user can't access (for example, Server Edition creates some temporary directories that can only be read by the user that ran the job, but those should be on your data partition anyway if you have specified a Directory for temporary sort files). There are probably equivalents for Unix, such as the du command, but SpaceMonger has a nice clear visual layout.
http://en.wikipedia.org/wiki/Category:D ... s_software
Phil Hibbs | Capgemini
Technical Consultant
Post Reply