Description of problem:

iscsiuio needs the following fixes:

1. iscsiuio cannot log from the same context that also needs to perform recovery operations, because if the logging disk and disk needing recovery are the same iscsiuio could block on the writing of log data and not be able to perform the recovery.

2. iscsiuio needs to set the oom values so that it is not killed, because if we need to write out data to free up memory and we need iscsiuio to also perform recovery iscsiuio must stay alive.

3. iscsiuio needs to set mlockall so that it is not paged out. Again, if it is paged out to a disk it later needs to perform some recovery operations for then this is not acceptable.


These issues were found while working on this bz
https://bugzilla.redhat.com/show_bug.cgi?id=741179

I did not clone it because the the libvirt, multipath and kernel people do not need to be bothered with this iscsi issue.


Version-Release number of selected component (if applicable):


How reproducible:


Steps to Reproduce:
1.
2.
3.
  
Actual results:


Expected results:


Additional info: