linux-f2fs-devel.lists.sourceforge.net archive mirror
 help / color / mirror / Atom feed
From: Sahitya Tummala <stummala@codeaurora.org>
To: Jaegeuk Kim <jaegeuk@kernel.org>, Chao Yu <yuchao0@huawei.com>,
	linux-f2fs-devel@lists.sourceforge.net
Subject: [f2fs-dev] [PATCH] fsck: Fix data loss issue observed during SPO
Date: Fri,  5 Jul 2019 10:31:03 +0530	[thread overview]
Message-ID: <1562302863-14418-1-git-send-email-stummala@codeaurora.org> (raw)

With the commit <979b25727800> ("fsck.f2fs: check validity of nat journal"),
there is a serious data loss issue observed during sudden power off tests.

The reason is due to incorrect update of journal->n_nats in that patch.
When fsck encounters a nat journal entry with invalid blkaddr/nid, it
updates journal->n_nats with the index of such first entry and then continues
the for loop which scans all nats in journal in f2fs_init_nid_bitmap().
But that loop doesn't continue and doesn't update nm_i->nid_bitmap as well,
as it fails the for loop condition i < nats_in_cursum(journal).

There could be several other valid entries in the journal after this
invalid entry. Due to this invalid journal->n_nats, build_nat_area_bitmap()
could not be build properly for valid nat journal entries. It further results
into these fsck asserts/errors and causes data loss due to incorrect fsck fix.

[ASSERT] (sanity_check_nid: 372)  --> nid[0xfb01] ino is 0\x0a
[FIX] (__chk_dentries:1438)  --> Unlink [0xfb01] - _JeNcl242yB3Apz2MW,VOh6WmjsVht1W len[0x1c], type[0x1]\x0a
[ASSERT] (sanity_check_nid: 372)  --> nid[0xfa2d] ino is 0\x0a
[FIX] (__chk_dentries:1438)  --> Unlink [0xfa2d] - _++vN7axccjjGNI,TQEVmcOemY1EAAAA len[0x14], type[0x1]\x0a
...
NID[0xba5] is unreachable\x0a
NID[0xfa9] is unreachable\x0a
...
[FIX] (nullify_nat_entry:2089)  --> Remove nid [0xba5] in NAT\x0a
[FIX] (nullify_nat_entry:2089)  --> Remove nid [0xfa9] in NAT\x0a
...

Fixes: 979b25727800 ("fsck.f2fs: check validity of nat journal")
Signed-off-by: Sahitya Tummala <stummala@codeaurora.org>
---
 fsck/mount.c | 2 --
 1 file changed, 2 deletions(-)

diff --git a/fsck/mount.c b/fsck/mount.c
index 60e0e4a..1b93f49 100644
--- a/fsck/mount.c
+++ b/fsck/mount.c
@@ -1160,7 +1160,6 @@ static int f2fs_init_nid_bitmap(struct f2fs_sb_info *sbi)
 		addr = le32_to_cpu(nat_in_journal(journal, i).block_addr);
 		if (!IS_VALID_BLK_ADDR(sbi, addr)) {
 			MSG(0, "\tError: f2fs_init_nid_bitmap: addr(%u) is invalid!!!\n", addr);
-			journal->n_nats = cpu_to_le16(i);
 			c.fix_on = 1;
 			continue;
 		}
@@ -1168,7 +1167,6 @@ static int f2fs_init_nid_bitmap(struct f2fs_sb_info *sbi)
 		nid = le32_to_cpu(nid_in_journal(journal, i));
 		if (!IS_VALID_NID(sbi, nid)) {
 			MSG(0, "\tError: f2fs_init_nid_bitmap: nid(%u) is invalid!!!\n", nid);
-			journal->n_nats = cpu_to_le16(i);
 			c.fix_on = 1;
 			continue;
 		}
-- 
Qualcomm India Private Limited, on behalf of Qualcomm Innovation Center, Inc.
Qualcomm Innovation Center, Inc. is a member of Code Aurora Forum, a Linux Foundation Collaborative Project.



_______________________________________________
Linux-f2fs-devel mailing list
Linux-f2fs-devel@lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/linux-f2fs-devel

             reply	other threads:[~2019-07-05  5:02 UTC|newest]

Thread overview: 14+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2019-07-05  5:01 Sahitya Tummala [this message]
2019-07-05 10:11 ` [f2fs-dev] [PATCH] fsck: Fix data loss issue observed during SPO Chao Yu
2019-07-05 11:03   ` Sahitya Tummala
2019-07-05 13:48     ` Sahitya Tummala
2019-07-06  1:14       ` Chao Yu
2019-07-06  1:31         ` Sahitya Tummala
2019-07-06  1:37           ` Sahitya Tummala
2019-07-06  2:10             ` Chao Yu
2019-07-06  2:10             ` Chao Yu
2019-07-06  2:17               ` Chao Yu
2019-07-06  4:50                 ` Jaegeuk Kim
2019-07-07  5:25                   ` Chao Yu
2019-07-08 23:43                     ` Jaegeuk Kim
2019-07-06  1:12     ` Chao Yu

Reply instructions:

You may reply publicly to this message via plain-text email
using any one of the following methods:

* Save the following mbox file, import it into your mail client,
  and reply-to-all from there: mbox

  Avoid top-posting and favor interleaved quoting:
  https://en.wikipedia.org/wiki/Posting_style#Interleaved_style

* Reply using the --to, --cc, and --in-reply-to
  switches of git-send-email(1):

  git send-email \
    --in-reply-to=1562302863-14418-1-git-send-email-stummala@codeaurora.org \
    --to=stummala@codeaurora.org \
    --cc=jaegeuk@kernel.org \
    --cc=linux-f2fs-devel@lists.sourceforge.net \
    --cc=yuchao0@huawei.com \
    /path/to/YOUR_REPLY

  https://kernel.org/pub/software/scm/git/docs/git-send-email.html

* If your mail client supports setting the In-Reply-To header
  via mailto: links, try the mailto: link
Be sure your reply has a Subject: header at the top and a blank line before the message body.
This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox;
as well as URLs for NNTP newsgroup(s).