------- Comment From dougm...@us.ibm.com 2018-04-23 16:04 EDT------- Looking closer at the logs of the crash in comment #81, I see that there are 3 calls into qlt_unreg_sess(), for the same port, in a span of less than 2 seconds. Between the time that the first instance of qlt_free_session_done() is queued and it actually executes, the port has a re-login event. It's difficult to tell exactly what all the login messages mean, but it appears that the login procedure continues even though qlt_free_session_done() appears to have logged out the port. There are then two more qlt_unreg_sess() calls during the login sequence, but both of kworker threads associated with those seem get stuck in the "wait for logout complete" loop - at least according to the stack traces in the dump. qla24xx_fcport_handle_login() seems to be getting called repeatedly, or at least as many times as qlt_unreg_sess() gets called. Unclear if that is normal, or a reaction to simultaneous attempts to logout and shutdown the port.
It also seems strange that more mutexing is not happening, although I can't see conclusive proof that these contradictory events are happening simultaneously. I do see that qlt_free_session_done(), among others, makes calls to qla2x00_fcport_event_handler() which would also handle the login. There may be some interesting stack traces in that core file, that might give a better picture of what is going on. Was that core file preserved someplace that we can look at? -- You received this bug notification because you are a member of Ubuntu Bugs, which is subscribed to Ubuntu. https://bugs.launchpad.net/bugs/1762844 Title: ISST-LTE:KVM:Ubuntu1804:BostonLC:boslcp3: Host crashed & enters into xmon after moving to 4.15.0-15.16 kernel To manage notifications about this bug go to: https://bugs.launchpad.net/ubuntu-power-systems/+bug/1762844/+subscriptions -- ubuntu-bugs mailing list ubuntu-bugs@lists.ubuntu.com https://lists.ubuntu.com/mailman/listinfo/ubuntu-bugs
