devices file: do not clear PVID of unread devices

In a certain disconnected state, a block device is present on
the system, can be opened, reports a valid size, reports the
correct device id (wwid), and matches a devices file entry.
But, reading the device can still fail.  In this case,
device_ids_validate() was misinterpreting the read error as
the device having no data/label on it (and no PVID).
The validate function would then clear the PVID from the
devices file entry for the device, thinking that it was
fixing the devices file (making it consistent with the on disk
state.)  Fix this by not attempting to check and correct a
devices file entry that cannot be read.  Also make this case
explicit in the hints validation code (which was doing the
right thing but indirectly.)

1 files changed, 14 insertions, 0 deletions

diff --git a/lib/label/hints.c b/lib/label/hints.c
index 3ce9634f2..95d5d77b8 100644
--- a/lib/label/hints.c
+++ b/lib/label/hints.c
@@ -234,6 +234,7 @@ static int _touch_newhints(void)
 		return_0;
 	if (fclose(fp))
 		stack;
+	log_debug("newhints created");
 	return 1;
 }
 
@@ -504,6 +505,19 @@ int validate_hints(struct cmd_context *cmd, struct dm_list *hints)
 		if (!hint->chosen)
 			continue;
 
+		/* 
+		 * label_scan was unable to read the dev so we don't know its pvid.
+		 * Since we are unable to verify the hint is correct, it's possible
+		 * that the PVID is actually found on a different device, so don't
+		 * depend on hints. (This would also fail the following pvid check.)
+		 */
+		if (dev->flags & DEV_SCAN_NOT_READ) {
+			log_debug("Uncertain hint for unread device %d:%d %s",
+				  major(hint->devt), minor(hint->devt), dev_name(dev));
+			ret = 0;
+			continue;
+		}
+
 		if (strcmp(dev->pvid, hint->pvid)) {
 			log_debug("Invalid hint device %d:%d %s pvid %s had hint pvid %s",
 				  major(hint->devt), minor(hint->devt), dev_name(dev),