[LU-11243] Assertion and hang upon lod_add_device failure

Duplicate — Mon, 1 Apr 2019 14:12:47 +0000

See following assertion:

    (lod_lov.c:361:lod_add_device()) lustre-OSTe42a-osc-MDT0000: can't set up pool, failed with -12
    10059:0:(osp_dev.c:473:osp_disconnect()) ASSERTION( imp != ((void *)0) ) failed:
    10059:0:(osp_dev.c:473:osp_disconnect()) LBUG
    CPU: 1 PID: 10059 Comm: llog_process_th Kdump: loaded Tainted: G
    
    Problem is obd_disconnect() will cleanup @imp and set NULL.
    |->osp_obd_disconnect
        |->class_manual_cleanup
           |->class_process_config
                 |->class_cleanup
                      |->obd_precleanup
                            |->osp_device_fini
                                  |->client_obd_cleanup
    
    While ldo_process_config() will try to access @imp again:
    |->ldo_process_config
      |->osp_shutdown
       |->osp_disconnect
         LASSERT(imp != NULL) --->fail here
    
    Another problem is if we failed before obd_connect().
    we will hang on with mount:
    |->ldo_process_config
      |->osp_shutdown
       |->osp_disconnect
        |->ptlrpc_disconnect_import
         |->rc = l_wait_event(imp->imp_recovery_waitq,
                !ptlrpc_import_in_recovery(imp), &lwi);

Since connect is not called, imp state will keep DISCONN.

Whamcloud Community JIRA

[LU-11243] Assertion and hang upon lod_add_device failure