http://bugzilla.novell.com/show_bug.cgi?id=521596 Summary: device mapper multipath: IO's not failed when all paths fail Classification: openSUSE Product: openSUSE 11.1 Version: Final Platform: x86-64 OS/Version: SLES 11 Status: NEW Severity: Normal Priority: P5 - None Component: Kernel AssignedTo: bnc-team-screening@forge.provo.novell.com ReportedBy: Eddie.Williams@starband.net QAContact: qa@suse.de Found By: --- User-Agent: Mozilla/5.0 (X11; U; Linux i686; en-US; rv:1.9.0.11) Gecko/2009060310 Ubuntu/8.10 (intrepid) Firefox/3.0.11 I have the updated kernel 2.6.27.23-0.1.1 and updated multipath tools 0.4.8-40.4.1 installed. With an EMC Clariion attached to a server if I pull all paths to the storage the IO's are queued indefinitely. Well, for the couple of hours that I waited. I modified the configured (captured the config file using "multipath -t" to create multipath.conf) for the Clariion to fail IO's instead of queue them, aka changed "no_path_retry 60" to "no_path_retry fail" and reloaded multipath (ran "multipath -F" and then ran "multipath"). I had the same result, aka pulling all paths the IO's were queued indefinitely. If I ran the dmsetup command to force the IO's to be failed this worked. Perhaps when all of the paths are actually removed from the multipath configuration (udev behavior) that multipath gets confused with no paths configured? This worked on SLES 10. Reproducible: Always Steps to Reproduce: 1. enable multipath 2. connect clariion array (2 paths in my config) 3. mount LUN 4. start an untar 5. pull both cables from back of server 6. IO's are queued indefinitely Actual Results: IO's are queued indefinitely Expected Results: IO's should fail eventually -- Configure bugmail: http://bugzilla.novell.com/userprefs.cgi?tab=email ------- You are receiving this mail because: ------- You are on the CC list for the bug.