BookmarkSubscribeRSS Feed
danewt
Obsidian | Level 7

Something appears to have gone off the rails with the APM ETL process in our SAS 9.4M8 Environment Manager Service Architecture. The nightly Audit, Performance, and Measurement (APM) job is now reporting errors and artifactusagedetails.sas7bdat is unusually large (22GB). The size of that file typically hovers around 4-5GB but, sometime in mid-NOV2025, the size started growing in fits and bursts. 

The runLog process seems to be choking on the large size of artifactusagedetails. Here is an extract from a recent failed runLog log file:

ERROR: Insufficient space in file ARTIFACT.ARTIFACTUSAGEDETAILS.DATA.
ERROR: File ARTIFACT.ARTIFACTUSAGEDETAILS.DATA is damaged. I/O processing did not complete.
NOTE: The SAS System stopped processing this step because of errors.
NOTE: Due to ERROR(s) above, SAS set option OBS=0, enabling syntax check mode. 
      This prevents execution of subsequent data modification statements.
NOTE: There were 11560138 observations read from the data set ARTIFACT.ARTIFACTUSAGEDETAILS.
      WHERE (datetime>2078956800) and (datetime<='  16JAN2026:23:59:00'DT);
WARNING: The data set ARTIFACT.ARTIFACTUSAGEDETAILS may be incomplete.  When this step was stopped there were 11560137 observations 
         and 35 variables.
NOTE: Compressing data set ARTIFACT.ARTIFACTUSAGEDETAILS decreased size by 86.71 percent. 
      Compressed is 96034 pages; un-compressed would require 722509 pages.
WARNING: Data set ARTIFACT.ARTIFACTUSAGEDETAILS was not replaced because this step was stopped.

Can anyone suggest how to clean up artifactusagedetails.sas7bdat?

1 REPLY 1
SASKiwi
PROC Star

I've never used APM but it is interesting to note that you are getting > 86% compression on the ARTIFACTUSAGEDETAILS dataset. That suggests it contains a lot of empty space. To reduce it in size one obvious thought would be to reduce the history kept. You will need to analyse the dataset in more detail to understand why it has grown so much. Comparing it to an earlier and smaller version would be useful.  

suga badge.PNGThe SAS Users Group for Administrators (SUGA) is open to all SAS administrators and architects who install, update, manage or maintain a SAS deployment. 

Join SUGA 

Get Started with SAS Information Catalog in SAS Viya

Learn how to explore data assets, create new data discovery agents, schedule data discovery agents, and much more.

Find more tutorials on the SAS Users YouTube channel.

Discussion stats
  • 1 reply
  • 499 views
  • 0 likes
  • 2 in conversation