metadata-updates.suite_log.service331.log
[LU-2570] test: metadata-updates: bind: Address family not supported by protocol Created: 03/Jan/13 Updated: 10/Jan/13 Resolved: 10/Jan/13 |
|
| Status: | Resolved |
| Project: | Lustre |
| Component/s: | None |
| Affects Version/s: | Lustre 2.3.0 |
| Fix Version/s: | None |
| Type: | Bug | Priority: | Minor |
| Reporter: | Jay Lan (Inactive) | Assignee: | Bob Glossman (Inactive) |
| Resolution: | Cannot Reproduce | Votes: | 0 |
| Labels: | None | ||
| Environment: |
Server: 2.1.3-1nasS, centos 6.3, 2.6.32_279.2.1.el6 |
||
| Attachments: |
metadata-updates.suite_log.service331.log
|
| Severity: | 3 |
| Rank (Obsolete): | 6002 |
| Description |
|
The metadata-updates failed in Part 5 at Complete test log is attached. |
| Comments |
| Comment by Peter Jones [ 03/Jan/13 ] |
|
Bob Can you please look into this one? Thanks Peter |
| Comment by Bob Glossman (Inactive) [ 03/Jan/13 ] |
|
Jay, from the log you've posted the command that's failing is mpirun_rsh. That doesn't seem to be part of openmpi. I can't find it anywhere. Is this something specific to your environment? Is it a script that invokes mpirun with some options? Without knowing more about it I can't figure out what it's complaining about. |
| Comment by Jay Lan (Inactive) [ 03/Jan/13 ] |
|
I will look into it tomorrow. I originally thought the "bind: Address family not supported by protocol" came from write_disjoint binary. |
| Comment by Jay Lan (Inactive) [ 04/Jan/13 ] |
|
I started to suspect it was caused by running sles11sp2 kernel in an sles11sp1 environment. I do not have an sles11sp2 image to image my test system, so I installed sles11sp2 kernel in an sles11sp1 environment. This maybe a compatibility issue between the kernel and run-time libraries. I will post when I have more information. |
| Comment by Jay Lan (Inactive) [ 04/Jan/13 ] |
|
The same problem still occured in a clean sles11sp2 environment. I will look into the mpirun_rsh next. |
| Comment by Jay Lan (Inactive) [ 10/Jan/13 ] |
|
The problem indeed was caused by our version of mpirun_rsh. Please close this ticket. |
| Comment by Bob Glossman (Inactive) [ 10/Jan/13 ] |
|
Closing due to customer request. Problem was due to a feature of the customer's environment. |