xfs_scrub: create a new category for unfixable errors

There's nothing that xfs_scrub (or XFS) can do about media errors for
data file blocks -- the data are gone.  Create a new category for these
unfixable errors so that we don't advise the user to take further action
that won't fix the problem.

[sandeen: this error counter is only used for media errors today, but
there are tests in the code to accommodate potential future new types
of unfixable errors.]

Signed-off-by: Darrick J. Wong <darrick.wong@oracle.com>
Signed-off-by: Eric Sandeen <sandeen@sandeen.net>
Reviewed-by: Eric Sandeen <sandeen@redhat.com>

7 files changed, 30 insertions, 8 deletions

diff --git a/scrub/common.c b/scrub/common.c
index 32ef8cd297..29d08f3ea0 100644
--- a/scrub/common.c
+++ b/scrub/common.c
@@ -43,7 +43,7 @@ xfs_scrub_excessive_errors(
 		return false;
 
 	pthread_mutex_lock(&ctx->lock);
-	errors_seen = ctx->corruptions_found;
+	errors_seen = ctx->corruptions_found + ctx->unfixable_errors;
 	pthread_mutex_unlock(&ctx->lock);
 
 	return errors_seen >= ctx->max_errors;
@@ -61,6 +61,10 @@ static struct {
 		.string = "Corruption",
 		.loglevel = LOG_ERR,
 	},
+	[S_UNFIXABLE] = {
+		.string = "Unfixable Error",
+		.loglevel = LOG_ERR,
+	},
 	[S_WARN]   = {
 		.string = "Warning",
 		.loglevel = LOG_WARNING,
@@ -136,6 +140,8 @@ out_record:
 		ctx->runtime_errors++;
 	else if (level == S_CORRUPT)
 		ctx->corruptions_found++;
+	else if (level == S_UNFIXABLE)
+		ctx->unfixable_errors++;
 	else if (level == S_WARN)
 		ctx->warnings_found++;
 	else if (level == S_REPAIR)
diff --git a/scrub/common.h b/scrub/common.h
index b1f2ea2cba..cfd9f186cd 100644
--- a/scrub/common.h
+++ b/scrub/common.h
@@ -18,6 +18,7 @@ bool xfs_scrub_excessive_errors(struct scrub_ctx *ctx);
 enum error_level {
 	S_ERROR	= 0,
 	S_CORRUPT,
+	S_UNFIXABLE,
 	S_WARN,
 	S_INFO,
 	S_REPAIR,
@@ -43,6 +44,8 @@ void __str_out(struct scrub_ctx *ctx, const char *descr, enum error_level level,
 	__str_out(ctx, str, S_REPAIR,	0,	__FILE__, __LINE__, __VA_ARGS__)
 #define record_preen(ctx, str, ...) \
 	__str_out(ctx, str, S_PREEN,	0,	__FILE__, __LINE__, __VA_ARGS__)
+#define str_unfixable_error(ctx, str, ...) \
+	__str_out(ctx, str, S_UNFIXABLE, 0,	__FILE__, __LINE__, __VA_ARGS__)
 
 #define dbg_printf(fmt, ...) \
 	do {if (debug > 1) {printf(fmt, __VA_ARGS__);}} while (0)
diff --git a/scrub/phase4.c b/scrub/phase4.c
index 1cf3f6b7be..a276bc329e 100644
--- a/scrub/phase4.c
+++ b/scrub/phase4.c
@@ -99,7 +99,10 @@ xfs_process_action_items(
 	workqueue_destroy(&wq);
 
 	pthread_mutex_lock(&ctx->lock);
-	if (moveon && ctx->corruptions_found == 0 && want_fstrim) {
+	if (moveon &&
+	    ctx->corruptions_found == 0 &&
+	    ctx->unfixable_errors == 0 &&
+	    want_fstrim) {
 		fstrim(ctx);
 		progress_add(1);
 	}
diff --git a/scrub/phase5.c b/scrub/phase5.c
index dc0ee5e884..e0c4a3dfaf 100644
--- a/scrub/phase5.c
+++ b/scrub/phase5.c
@@ -336,7 +336,7 @@ xfs_scan_connections(
 	bool			moveon = true;
 	bool			ret;
 
-	if (ctx->corruptions_found) {
+	if (ctx->corruptions_found || ctx->unfixable_errors) {
 		str_info(ctx, ctx->mntpoint,
 _("Filesystem has errors, skipping connectivity checks."));
 		return true;
diff --git a/scrub/phase6.c b/scrub/phase6.c
index bb159641aa..aae6b7d80f 100644
--- a/scrub/phase6.c
+++ b/scrub/phase6.c
@@ -161,7 +161,7 @@ report_badfile(
 	bad_length = min(start + length,
 			 br->bmap->bm_physical + br->bmap->bm_length) - start;
 
-	str_error(br->ctx, br->descr,
+	str_unfixable_error(br->ctx, br->descr,
 _("media error at data offset %llu length %llu."),
 			br->bmap->bm_offset + bad_offset, bad_length);
 	return 0;
diff --git a/scrub/xfs_scrub.c b/scrub/xfs_scrub.c
index 222daae15f..963d0d7044 100644
--- a/scrub/xfs_scrub.c
+++ b/scrub/xfs_scrub.c
@@ -511,15 +511,24 @@ static void
 report_outcome(
 	struct scrub_ctx	*ctx)
 {
-	unsigned long long	total_errors;
+	unsigned long long	actionable_errors;
 
-	total_errors = ctx->corruptions_found + ctx->runtime_errors;
+	actionable_errors = ctx->corruptions_found + ctx->runtime_errors;
 
-	if (total_errors == 0 && ctx->warnings_found == 0) {
+	if (actionable_errors == 0 &&
+	    ctx->unfixable_errors == 0 &&
+	    ctx->warnings_found == 0) {
 		log_info(ctx, _("No problems found."));
 		return;
 	}
 
+	if (ctx->unfixable_errors) {
+		fprintf(stderr, _("%s: unfixable errors found: %llu\n"),
+				ctx->mntpoint, ctx->unfixable_errors);
+		log_err(ctx, _("unfixable errors found: %llu"),
+				ctx->unfixable_errors);
+	}
+
 	if (ctx->corruptions_found > 0) {
 		fprintf(stderr, _("%s: corruptions found: %llu\n"),
 				ctx->mntpoint, ctx->corruptions_found);
@@ -545,7 +554,7 @@ report_outcome(
 	 * setting up the scrub and we actually saw corruptions.  Warnings
 	 * are not corruptions.
 	 */
-	if (ctx->scrub_setup_succeeded && total_errors > 0) {
+	if (ctx->scrub_setup_succeeded && actionable_errors > 0) {
 		char		*msg;
 
 		if (ctx->mode == SCRUB_MODE_DRY_RUN)
diff --git a/scrub/xfs_scrub.h b/scrub/xfs_scrub.h
index 5abc41fd60..61831c92fa 100644
--- a/scrub/xfs_scrub.h
+++ b/scrub/xfs_scrub.h
@@ -74,6 +74,7 @@ struct scrub_ctx {
 	unsigned long long	max_errors;
 	unsigned long long	runtime_errors;
 	unsigned long long	corruptions_found;
+	unsigned long long	unfixable_errors;
 	unsigned long long	warnings_found;
 	unsigned long long	inodes_checked;
 	unsigned long long	bytes_checked;