In the hot-plug storage OS, if we tear down the target storage, there is a race between the path removing from the path list and handling the path-failed udev event. Therefore, we need to check null path before handle path-failed event.
Signed-off-by: Guan Junxiong <[email protected]> --- multipathd/main.c | 10 ++++++---- 1 file changed, 6 insertions(+), 4 deletions(-) diff --git a/multipathd/main.c b/multipathd/main.c index 31ce923..84bb2c5 100644 --- a/multipathd/main.c +++ b/multipathd/main.c @@ -1073,7 +1073,7 @@ uev_pathfail_check(struct uevent *uev, struct vectors *vecs) { char *action = NULL, *devt = NULL; struct path *pp; - int r; + int r = 1; action = uevent_get_dm_action(uev); if (!action) @@ -1090,15 +1090,17 @@ uev_pathfail_check(struct uevent *uev, struct vectors *vecs) lock(&vecs->lock); pthread_testcancel(); pp = find_path_by_devt(vecs->pathvec, devt); + if (!pp) + goto out_lock; r = io_err_stat_handle_pathfail(pp); - lock_cleanup_pop(vecs->lock); - if (r) condlog(3, "io_err_stat: %s: cannot handle pathfail uevent", pp->dev); +out_lock: + lock_cleanup_pop(vecs->lock); FREE(devt); FREE(action); - return 0; + return r; out: FREE(action); return 1; -- 2.6.4.windows.1 -- dm-devel mailing list [email protected] https://www.redhat.com/mailman/listinfo/dm-devel
