About jklaverstijn

jklaverstijn · ‎12-14-2017

Yep I joined as soon as I read your comment Juan. Thanks for pointing it out. It was educational but did not help me further on this one.

jklaverstijn · ‎12-13-2017

Dear fellow admins, We are running SAS 9.4M4 with SAS Grid manager on Linux (RHEL 7.3). This works fine as it is but we are facing a next step in the scale of our implementation. We are planning for what we call a multi-tenant environment where several organisational units will work in blissful isolation from each other. This separation used to be achieved by using separate server contexts until the grid options sets came along. The way grid options sets work taking identity and group membership into account if a perfect match for our multi-tenancy. It allows us to control which queues are being used and what SAS options are in effect. Very powerful stuff and it helps with the separation as far as workspace servers and stored process servers is concerned (I think these are all IOM servers). The last piece of the puzzle is the LSF batch processing. The batch datastep server is not an IOM server so a batch job cannot be controlled with a grid options set. I have read this article on integrating the schedule manager with grid manager. This explains how to use sasgsub to run the batch jobs. This command would allow us to attach grid options sets. The sasgsub command wants metadata credentials via the metauser/metapass options. the article says METAPASS should be changed from _PROMPT_ to the encrypted value of the password corresponding to the METAUSER value But how can we obtain a metapass value from the sasbatch.sh script in a secure fashion? We don't know the user passwords so this sounds a bit too easy. And no, we do not want to run all batch jobs with a single administrative user; it must work for individual users. Has anyone cracked this or found a better way? I am looking forward to your input. Regards, -- Jan

jklaverstijn · ‎12-13-2017

Hi Manny, I have done some work on reading in some of the process manager and lsf status files. These are event based so you would have to do some work to transform a series of events to a current state. The files I found interesting and still comprehensible are $LSFTOP/work/dna_cluster/logdir/lsb.events, $LSFTOP/work/dna_cluster/logdir/lsb.acct and $JSTOP/work/history/history.log. They are all easy prey for a datastep. IBM has documented them in their knowledge center: https://www.ibm.com/support/knowledgecenter/en/SSWRJV_10.1.0/lsf_config_ref/lsb.events.5.html https://www.ibm.com/support/knowledgecenter/en/SSWRJV_10.1.0/lsf_config_ref/lsb.acct.5.html A more ready made approach would be to use the EOM DI Monitor. This is a web application that helps you monitor the SAS batch jobs in LSF. This is an open source development by EOM. Get in touch with Bart Heinsius for details. We use it to great satisfaction. Hope this helps, -- Jan

jklaverstijn · ‎12-02-2017

Yes very much as I suggested. Kudo's for using an informat instead of a format. Wouldn't use the name _n_ for the loop index however as this is a special variable. regards, -- Jan

jklaverstijn · ‎12-02-2017

And on question 2: since your variable is character, the use of "<>" is not appropriate. Also, the variable your testing must be repeated after the AND: AssessorDecision ne "Withdrawn" and AssessorDecision ne ' ' Regards, - Jan.

jklaverstijn · ‎12-02-2017

An entirely different approach could be to transpose your data. If you have so many similar columns consider transforming that to a dataset with just one variable plus an extra indicator for the suffix that you would remove. The aforementioned technique with arrays and do loops could be used for the purpose of transposing if proc transpose doesn't cut it. You would then not have to repeat the mapping for each variable in each row but just the single variable in each row. This wide to long transposition usually gives you a dataset that is easier to analyze so you may benefit from this approach in other scenario's as well. - Jan.

jklaverstijn · ‎12-02-2017

It is always a good idea to suspect optimizations exist if you notice repetition in your code. Without trying to do the coding itself I can give some pointers: 1) Variable lists with a numeric postfix (but also others) are good candidates stick in an array. You can then loop over that array. See this older but stil relevant SGF paper. It is easy to add or change your arrays at a later stage. 2) These groups of codes that map to a single flag can be handled by a format. Your "if value in ('a', 'b', ...) then" can be transfored into "flag = put(value, $mappingfmt.)". The proc format at the top of your code would be the single place to add or change mappings. Hope this helps you on your way, -- Jan

jklaverstijn · ‎11-30-2017

does it have any issues? -- probably. bugs, -- certainly it is good to have over M4 ? -- very likely SAS makes no secret that a product this elaborate and complex has bugs. It's the ones that bite you that you should be concerned about. We tend to follow an 9.4Mn update based on the changes & enhancements and not the bugs. E.g. we moved from 9.4M3 to 9.4M4 for Kerberos support in Teradata, a new version of Business Rules Manager and the new JSON support. 9.4M5 may become highly relevant if you also deploy SAS Viya. In all it's a site specific decision. based on this you may conclude that what SAS now calls a maintenance release is in fact a bit more than that. Also keep an eye on the version bumps in some clients. For example Data Integration Studio will receive updates in sync with maintenance releases. Same can be true for Enterprise Miner and SAS Studio. Enterprise Guide seems to have gone rogue and follow its own lifecycle. Bugs and issues are generally dealt with in hotfixes that you apply on demand. These hotfixes are also rolled up in the next maintenance release so upgrading will deal with all issues you would otherwise fix manually and some you never realized you had in the first place (the majority). As an extra you may expect new bugs with a new version as well. These are a fact of life. Don't loose sleep over them. Some actually made my life more interesting. When problems arise the learning begins. Again you will normally notice only a small percentage of them. Keep an eye on the hotfix pages; SAS is very good at communicating problems and their solutions. So my advice is to fix your issues based on the hotfixes and upgrade based on the C&E docs. Hope this helps, - Jan.

jklaverstijn · ‎11-29-2017

Without the full code and logs there's no way of telling. Please run your program with OPTIONS MPRINT SYMBOLGEN; and provide us the details. Also, make sure it's formatted properly by pasting into a code segment (click on the {i} or running man icon). And to test your connection I would suggest to insert the proper values verbatim instead of via a macro var reference. If that works you can take the next step and use macro's. Regards, - Jan.

jklaverstijn · ‎11-21-2017

I apologize for coming back to an accepted solution but this one is not airtight. It will only work if the account used to reschedule has administrative powers in LSF. If an arbitrary user tries to redeploy a flow that is already scheduled by another arbitrary user, this will not work as described. A warning will be issued by SMC and two flows with the same name but different users will now exist in Flow Manager. As the message indicates the existing flow needs to be removed by hand as it was not overwritten. The current scheduling user does not necessarily have that privilege. Only an LSF administrator or the owning user can remove a flow. This has to be done using the Flow Manager application that is part of your SAS distribution. Hope this helps, - Jan.

jklaverstijn · ‎11-20-2017

The cleanwork utility is your way to go and others have already replied on how it works. On a side note I would be interested in why your work grows so quickly. In the ideal situation, when SAS terminates in an intended fashion ( a user quits the SAS session; a SAS job end) the work library is deleted automatically. Any residual work locations are due to SAS sessions that end in a less fortunate way because they crash or are killed by some administrative action. If you have many of these this may require some investigation. Start with checking the value of the WORKTERM option. By default it is WORKTERM; the opposite, NOWORKTERM, will cause your WORK libraries to persist. I have never seen that being the default but you never know. If WORKTERM is the default, dig deeper. Regards, - Jan.

jklaverstijn · ‎11-19-2017

Can you provide us with some information or examples on what you have and what you want? And what you have tried sofar? Now there's very little to go by. - Jan

jklaverstijn · ‎11-18-2017

A flow once scheduled by a user is tied to that user. Unless you have a full LSF license, the one that comes with SAS limits your possibilities somewhat. For starters you cannot change flow definitions from Flow Manager. Rescheduling a flow as someone else would at best result in two flows with the same name but different users. The old one would still be there. In your case, assuming you cannot logon as the former (assuming you ment "former" when you said "formal") user, you can unschedule the existing flow using administrative accounts and re-schedule as someone else. Unscheduling can be done from SAS Management Console as a user with administrative powers (eg sasadm@saspw). You will likely be prompted (sasadm@saspw usually has no account on the OS) for a logon to the Platform scheduling server. Use lsfadmin or some other account that is defined as an LSF Administrator. This takes care of removing schedule info both from metadata and LSF. Once that is done, reschedule with another account. Sometimes things don't go as planned. It can happen that a flow is unscheduled as far as the metadata are concerned but it still exists in LSF. This conflict can be a pesky one. Use lsfadmin or equivalent to remove the flow directly in Flow Manager. There are many pointers in the response from @JuanS_OCS on how to tie accounts to a authentication domain directly or via a group membership. As you have already scheduled quite some flows we can only hope you figured this one out by now.

jklaverstijn · ‎11-18-2017

Looking at parentheses the ! is interpreted as a parameter for %sysfunc and not of countw. You're better of with %sysfunc(countw(&Beds.,!)); Hope this helps, - Jan

jklaverstijn · ‎11-15-2017

Have you attempted a metadata Analyze&Repair from SAS Management Console? If not have a go and let us know what happened. Regards, - Jan.

Online Status	Offline
Date Last Visited	‎09-22-2025 12:07 PM

Re: Why does SAS randomly report "Library WORK does not exist" during ...

Empty Web App Servers

Re: Email notifications from subflows

Re: Email notifications from subflows

Re: Automating SAS Package deployment using Import/Export Batch Tool

Re: Automating SAS Package deployment using Import/Export Batch Tool

Email notifications from subflows

Re: Wolfensäs: The FPS game that was written in SAS

Re: SAS Upgrade from 9.4 M7 to 9.4 M8 failing

Re: How to attach a file from a server in SAS Email

Re: Before you update to RHEL 8.8, please read.....

Re: Reference column type in data step OR suppress invalid numeric dat...

Re: How to set the umask for CAS?

Re: Convert Character ISO datetime into datatime

Re: SAS EG Stored Process wont update table

Re: Email notifications from subflows

Re: SAS Enterprise Guide to point to Unicode SAS

Re: SAS Upgrade from 9.4 M7 to 9.4 M8 failing

Re: Which element of the architecture effectively sends an e-mail when...

Re: How to set the umask for CAS?

Re: Integration of LSF batch scheduler and grid options sets

Integration of LSF batch scheduler and grid options sets

Re: sas-grid batch Job status in a table format

Re: data steps with nested do loops

Re: data steps with nested do loops

Re: data steps with nested do loops

Re: data steps with nested do loops

Re: SAS 9.4 M5 reliability

Re: Connect to SQLSVR( DATAsrc="&varName") ... won't connect

Re: How to redeploy in Flow Manager with different account?

Re: Workspace Cleanup

Re: code

Re: How to redeploy in Flow Manager with different account?

Re: strange response of countw

Re: sas metadata server backup failed

SAS Global Forum 2017

SAS Global Forum 2018

SAS Global Forum 2016