LU-4313 osp: continue waiting for -ENOTCONN

in osp_precreate_reserve, -ENOTCONN should also be treated
as recoverable errors, such as -ENODEV, and continue waiting
the object to be created by OSP once it completes recovery.

This patch is back-ported from the following one:
Lustre-commit: 3a2073d3bf9fab33fa26c5c5f0da2b0821185131
Lustre-change: http://review.whamcloud.com/8415

Signed-off-by: Hongchao Zhang <hongchao.zhang@intel.com>
Change-Id: I1fd13c8bec9b7827396ec922cb985ee642a5e286
Signed-off-by: Jian Yu <jian.yu@intel.com>
Reviewed-on: http://review.whamcloud.com/8468
Tested-by: Jenkins
Tested-by: Maloo <hpdd-maloo@intel.com>
Reviewed-by: Alex Zhuravlev <alexey.zhuravlev@intel.com>
Reviewed-by: Oleg Drokin <oleg.drokin@intel.com>

diff --git a/lustre/osp/osp_precreate.c b/lustre/osp/osp_precreate.c
index ff6beb6..ee0fd5b 100644 (file)
--- a/lustre/osp/osp_precreate.c
+++ b/lustre/osp/osp_precreate.c
@@ -1063,7 +1063,7 @@ int osp_precreate_reserve(const struct lu_env *env, struct osp_device *d)
         *  - OST can allocate fid sequence.
         */
        while ((rc = d->opd_pre_status) == 0 || rc == -ENOSPC ||
-               rc == -ENODEV || rc == -EAGAIN) {
+               rc == -ENODEV || rc == -EAGAIN || rc == -ENOTCONN) {
 
                /*
                 * increase number of precreations