I am back in my own personal Hell that appears when I have to debug something in the unix/linux world. I am here hoping to piggy back on the experiences of others who have been thrown into the deep end and survived.
My usual approach to debugging is to look at the messages and codes produced during the problem... then looking up what the message REALLY means (if you can find any documentation to help translation). That usually gets me close enough to at least guess about what went wrong and how to fix it. The present problem is that the TOMCAT JAVA app server that has been running for MANY years has now gone on strike. The 'S TOMCAT' command is met with a U4038 abend and instant death. I do know that U4038 is the generic "something went wrong" from the LE language environment modules. That is ALL that I get. No unix log files, so monster JAVA stack dumps with module names spread out over pages of dump output. NO dump output of any kind. Nothing in the Unix syslog. Nothing written to the JES spool for this started task. and so on. This post is a 'hail mary' that hopefully reaches the other poor sap who remembers how they got out of a similar situation, TLDR summary: A working JAVA app now fails to start with no know trail of how and why it failed. Where is a poor sysprog to look to find the problem? ---------------------------------------------------------------------- For IBM-MAIN subscribe / signoff / archive access instructions, send email to [email protected] with the message: INFO IBM-MAIN
