3

I have changed the logging level in my WebSphere Application Server v8.5 from *=info to *=all, and restarted the server from eclipse. The server went down but did not start again, I tried rebooting the machine a couple of times and tried starting/stopping the server from cmd, but nothing worked.

I'm suspecting that it might be due to excessive logging (since I changed the level to all, it might be too much load on the server to start?)

My question is, is there any way to change to logging level back from any configuration or something like that? Since the server did not start so I can't change it back from admin console. Any suggestions would be much appreciated.

[EDIT] I checked the logs and found that WebSphere is logging a huge amount of trace logs and that's why the server takes too long to start.

Ahmed Anwar
  • 688
  • 5
  • 25
  • Is your server part of a managed (ND) cell, or is it a "standalone" server? Also note that your trace strings aren't formatting too well... you might want `*=info` (backticks) though the question is clear enough now. – Scott Kurz Oct 03 '17 at 15:59
  • It is a standalone server (development machine actually) And you're right it is *=info not *.info, I edited the question. My problem is that the server took an hour so far and did not start yet, so I need to change the logging level from anywhere that's not WebSphere admin console since I can't access it while the server is not started. – Ahmed Anwar Oct 03 '17 at 16:05

1 Answers1

6

The trace specification is defined in server.xml, located within your profile directory in config/cells/(cellname)/nodes/(nodename)/servers/(servername).

For safety's sake, I'd recommend backing up the file first, then open it in an editor and search for "startupTraceSpecification". Swap that from *=all back to *=info, and you should be good to go.

Jarid
  • 1,878
  • 7
  • 11