client-messges
mds1-messges
mds2-messges
[LU-3875] Lustre client hanging due to :client_bulk_callback() and :ptlrpc_expire_one_request from OST to MDS nid time out Created: 04/Sep/13 Updated: 09/Jan/20 Resolved: 09/Jan/20 |
|
| Status: | Resolved |
| Project: | Lustre |
| Component/s: | None |
| Affects Version/s: | Lustre 2.1.3 |
| Fix Version/s: | None |
| Type: | Bug | Priority: | Major |
| Reporter: | hithesh kumar | Assignee: | WC Triage |
| Resolution: | Cannot Reproduce | Votes: | 0 |
| Labels: | None | ||
| Attachments: |
client-messges
mds1-messges
mds2-messges
|
| Severity: | 4 |
| Rank (Obsolete): | 10056 |
| Comments |
| Comment by hithesh kumar [ 04/Sep/13 ] |
|
The mds1 is the MDS server for MDT. mds2 is the failover node for mds1 and also serves around 7 OST's for which mds1 is the failover node. we are facing lustre mount hang problem for which i have attached client /var/log/messages of client, mds2(acts as OSS for 7 ost and MDT failover node). Lustre 2.1.3 has been installed on all MDS and OSS servers, Lustre patchless client is installed. |
| Comment by hithesh kumar [ 04/Sep/13 ] |
|
173.16.1.50 is mds1, 173.16.1.51 is mds2, 173.16.1.52 is oss1, 173.16.1.53 is oss2 servers, we have around 34 client machine whose ip ranges from 173.16.1.220 to 173.16.1.254. Native Centos 6.1 infiniband rpms are installed. |
| Comment by Andreas Dilger [ 09/Jan/20 ] |
|
Close old bug |