Poor condor_schedd responsiveness in MRG 1.2
Issue
-
Poor condor_schedd responsiveness
-
Using "netstat -ap| grep schedd" shows the UDP command queue for the schedd had backed up.
-
Logs show many entries such as "ERROR: receiving new UDP message but found a long message still waiting to be closed (consumed=0). Closing it now"
Environment
- MRG 1.2 condor-7.4.3-0.3.el5
Subscriber exclusive content
A Red Hat subscription provides unlimited access to our knowledgebase, tools, and much more.
