[LU-12567] ha-all-ior-ssf-fpp-mdtest: IOR fpp: cannot close file. ERROR: Input/output error

• Status: Resolved
• Project: Lustre
• Fix Version/s: Lustre 2.15.0
• Type: Bug
• Priority: Major
• Reporter: Alexander Boyko
• Assignee: Alexander Boyko
• Resolution: Fixed
• Labels: patch
• Severity: 3

Description

The next error at client log

2019-06-06 09:26:41 [72389.747032] LustreError: 10999:0:(osc_request.c:1076:check_write_rcs()) Unexpected # bytes transferred: 0 (requested 16777216)

      if (req->rq_bulk != NULL &&                                                        
            req->rq_bulk->bd_nob_transferred != requested_nob) {                          
                CERROR("Unexpected # bytes transferred: %d (requested %d)\n",             
                       req->rq_bulk->bd_nob_transferred, requested_nob);                  
                return(-EPROTO);                                                          
        } 
static inline int osc_recoverable_error(int rc)
{
        return (rc == -EIO || rc == -EROFS || rc == -ENOMEM ||
                rc == -EAGAIN || rc == -EINPROGRESS);
}

This error is not recoverable and fail the IO.
It looks like reply came for original request while we are resending it. So request has a reply, but bulk has 0 bytes.
The patch https://review.whamcloud.com/#/c/32905 fix the issue, but original ticket is closed. So I resubmitting to this https://review.whamcloud.com/#/c/35571

Comments

[Gerrit Updater]

"Oleg Drokin <[e-mail]>" merged in patch https://review.whamcloud.com/35571/
Subject: LU-12567 ptlrpc: handle reply and resend reorder
Project: fs/lustre-release
Branch: master
Current Patch Set:
Commit: f7f31f8f969f410cca0b4b8b02f81391148e01f2

[Peter Jones]

Landed for 2.15