On Thu, 2 Feb 2017 19:16:13 +0000, Bart Van Assche wrote: > On Thu, 2017-02-02 at 20:10 +0100, David Disseldorp wrote: > > Current master - the REPORT_SUPPORTED_OPCODES skip can be ignored... > > the full test name needs a prefix: > > --test=ALL.iSCSITMF.LUNResetSimpleAsync > > or > > iSCSI.iSCSITMF.LUNResetSimpleAsync > > Even if I run that test a few hundred times in a loop "echo w > > /proc/sysrq-trigger" still doesn't report a lockup. I would appreciate it if > you could provide me sufficient information to reproduce the behavior you > reported. But do you see the stuck core_tmr_lun_reset and iscsit_cause_connection_reinstatement threads? I could have worded my report better. The "lockup" I was referring to was the initiator awaiting TMF response and never getting one - which doesn't happen with mainline. This is easiest to reproduce with: iscsi-test-cu --test=ALL.iSCSITMF.LUNResetSimpleAsync ... and then iscsi-test-cu --test=ALL.MultipathIO.Reset ... (you'll need to provide your IQN twice for MPIO) If you run MultipathIO.Reset before iSCSITMF.LUNResetSimpleAsync then it runs fine (ignoring the test failure), if you run it afterwards then it blocks indefinitely, as LIO never sends a TMF response. Cheers, David -- To unsubscribe from this list: send the line "unsubscribe target-devel" in the body of a message to majordomo@xxxxxxxxxxxxxxx More majordomo info at http://vger.kernel.org/majordomo-info.html
