messages
t324_vmcore_crash.log
[LU-2996] ptlrpc doesn't response under the heavy memory pressure Created: 20/Mar/13 Updated: 11/Mar/16 Resolved: 11/Mar/16 |
|
| Status: | Resolved |
| Project: | Lustre |
| Component/s: | None |
| Affects Version/s: | None |
| Fix Version/s: | None |
| Type: | Bug | Priority: | Minor |
| Reporter: | Shuichi Ihara (Inactive) | Assignee: | Emoly Liu |
| Resolution: | Incomplete | Votes: | 0 |
| Labels: | None | ||
| Environment: |
Lustre-1.8.8, CentOS6.1 |
||
| Attachments: |
messages
t324_vmcore_crash.log
|
| Severity: | 3 |
| Rank (Obsolete): | 7306 |
| Description |
|
Under tha heavy memory pressure, client doesn't response in a timely manner, but even system goes to be back normal, a lot of call traces that caused by ptlrpc. Attached are client's syslog and stack trace after we got crashdump. |
| Comments |
| Comment by Peter Jones [ 21/Mar/13 ] |
|
Hi Emoly Could you please help with this issue? Thanks Peter |
| Comment by Emoly Liu [ 21/Mar/13 ] |
|
OK, I will check the logs. |
| Comment by Emoly Liu [ 25/Mar/13 ] |
|
From the log we can see that there were two lustre file systems mounted on this client node and each lustre filesystem had tens of OSTs at least. During the log time, the client node (lustre client) was restarted five times. Ihara, to understand the issue clearly, could you please introduce more details about the system environment and related operations? Thanks. |
| Comment by Shuichi Ihara (Inactive) [ 02/Apr/13 ] |
|
Emoly, yes, two Lustre filesystem are mounted |
| Comment by Shuichi Ihara (Inactive) [ 06/Apr/13 ] |
|
Emoly, any updates? |
| Comment by John Fuchs-Chesney (Inactive) [ 25/Feb/16 ] |
|
Hello Ihara, This ticket has become very old – do you need any more work on it or may we mark it as resolved? Many thanks, |
| Comment by John Fuchs-Chesney (Inactive) [ 11/Mar/16 ] |
|
Resolving as incomplete. |