From: Qu Wenruo <quwenruo.btrfs@gmx.com>
To: dsterba@suse.cz, Qu Wenruo <wqu@suse.com>, linux-btrfs@vger.kernel.org
Subject: Re: [PATCH v5.1 00/12] btrfs: Enhancement to tree block validation
Date: Wed, 27 Feb 2019 21:40:12 +0800 [thread overview]
Message-ID: <88d6815c-1e15-4f89-54f4-bcb6cd2177e6@gmx.com> (raw)
In-Reply-To: <20190227122235.GE24609@twin.jikos.cz>
[-- Attachment #1.1: Type: text/plain, Size: 5282 bytes --]
On 2019/2/27 下午8:22, David Sterba wrote:
> On Sat, Feb 23, 2019 at 08:47:09AM +0800, Qu Wenruo wrote:
>>
>>
>> On 2019/2/22 下午11:18, David Sterba wrote:
>>> On Mon, Feb 18, 2019 at 01:27:41PM +0800, Qu Wenruo wrote:
>>>> v5.1:
>>>> - Add "block=%llu " output for write/read time error line.
>>>> - Also output read time error message for fsid/start/level check.
>>>
>>> I see reports from tests btrfs/124 and btrfs/125,
>>
>> That two tests are RAID1 reconstruction.
>> It's known to trigger read error on missing devices.
>>
>> Especially for bytenr mismatch, as all read from missing device will
>> result 0 filled page.
>>
>> I could change the read time error message to skip bytenr/csum mismatch
>> to avoid such report.
>
> Yeah the output needs some tuning. See the example below.
>>
[snip]
>>> [ 2860.244462] BTRFS error (device vdb): bad tree block start, want 31703040 have 0
>>> [ 2860.248013] BTRFS error (device vdb): block=31703040 read time tree block corruption detected
>>> [ 2860.284017] BTRFS error (device vdb): block=31719424 read time tree block corruption detected
>>> [ 2860.315705] BTRFS error (device vdb): block=31735808 read time tree block corruption detected
>>> [ 2860.336001] BTRFS error (device vdb): block=31752192 read time tree block corruption detected
>
> Block 31752192 does not have any previous message about a specific error
> (unlike the 'bad tree block' messages of other blocks). I don't think
> it's good to print just the 'read time' message.
The first is caused by the bytenr mismatch.
But for the remaining 3, they can be caused by read error from missing
device. EXTENT_BUFFER_READ_ERR would cause no detailed error report here.
>
>>> [ 2860.336054] BTRFS error (device vdb): block=31768576 read time tree block corruption detected
>>> [ 2860.339786] BTRFS error (device vdb): block=31784960 read time tree block corruption detected
>>> [ 2860.366693] BTRFS error (device vdb): block=31768576 read time tree block corruption detected
>>> [ 2860.389495] BTRFS error (device vdb): block=31784960 read time tree block corruption detected
>>> [ 2860.418551] BTRFS error (device vdb): block=31866880 read time tree block corruption detected
>>> [ 2860.418713] BTRFS error (device vdb): block=31883264 read time tree block corruption detected
>>> [ 2860.418721] BTRFS error (device vdb): block=31916032 read time tree block corruption detected
>>> [ 2860.418736] BTRFS error (device vdb): block=31932416 read time tree block corruption detected
>>> [ 2860.418744] BTRFS error (device vdb): block=31948800 read time tree block corruption detected
>>> [ 2860.419721] BTRFS error (device vdb): block=31899648 read time tree block corruption detected
>>> [ 2860.462401] BTRFS error (device vdb): block=31866880 read time tree block corruption detected
>>> [ 2860.490821] BTRFS error (device vdb): block=31932416 read time tree block corruption detected
>>> [ 2860.506718] BTRFS error (device vdb): block=31899648 read time tree block corruption detected
>>> [ 2860.535013] BTRFS error (device vdb): block=31916032 read time tree block corruption detected
>>> [ 2860.568602] BTRFS error (device vdb): block=31948800 read time tree block corruption detected
>>> [ 2860.590434] BTRFS error (device vdb): block=31965184 read time tree block corruption detected
>>> [ 2860.590532] BTRFS error (device vdb): block=31981568 read time tree block corruption detected
>>> [ 2860.590677] BTRFS error (device vdb): block=31997952 read time tree block corruption detected
>>> [ 2860.590688] BTRFS error (device vdb): block=32014336 read time tree block corruption detected
>>>
>>> [ 2860.590697] BTRFS error (device vdb): block=32030720 read time tree block corruption detected
>>> [ 2860.590705] BTRFS error (device vdb): block=32047104 read time tree block corruption detected
>>> [ 2860.625834] BTRFS error (device vdb): block=31965184 read time tree block corruption detected
>>> [ 2860.658723] BTRFS error (device vdb): block=31981568 read time tree block corruption detected
>>> [ 2860.696644] BTRFS error (device vdb): block=32014336 read time tree block corruption detected
>>> [ 2860.735014] BTRFS error (device vdb): block=32030720 read time tree block corruption detected
>>> [ 2860.743240] BTRFS error (device vdb): block=32096256 read time tree block corruption detected
>>> [ 2860.780296] BTRFS error (device vdb): block=32047104 read time tree block corruption detected
>>> [ 2860.807073] BTRFS error (device vdb): block=32063488 read time tree block corruption detected
>>> [ 2860.807173] BTRFS error (device vdb): block=32079872 read time tree block corruption detected
>>> [ 2860.841582] BTRFS error (device vdb): block=32079872 read time tree block corruption detected
>
> This long sequence of messages lacks details about what happen, and
> there are duplicates (block 32079872). If this is caused by reads from
> missing device, this can potentially flood the logs.
Not only from missing device, but also from stale old device.
Anyway, since I have reverted the behavior to the original one, which
only output the "read time" error message for tree checker error, it
should no longer flood the kernel message.
Thanks,
Qu
>
[-- Attachment #2: OpenPGP digital signature --]
[-- Type: application/pgp-signature, Size: 488 bytes --]
prev parent reply other threads:[~2019-02-27 13:40 UTC|newest]
Thread overview: 30+ messages / expand[flat|nested] mbox.gz Atom feed top
2019-02-18 5:27 [PATCH v5.1 00/12] btrfs: Enhancement to tree block validation Qu Wenruo
2019-02-18 5:27 ` [PATCH v5.1 01/12] btrfs: Always output error message when key/level verification fails Qu Wenruo
2019-02-18 5:27 ` [PATCH v5.1 02/12] btrfs: extent_io: Kill the forward declaration of flush_write_bio() Qu Wenruo
2019-02-18 5:27 ` [PATCH v5.1 03/12] btrfs: disk-io: Show the timing of corrupted tree block explicitly Qu Wenruo
2019-02-23 4:38 ` [PATCH v5.2 " Qu Wenruo
2019-02-18 5:27 ` [PATCH v5.1 04/12] btrfs: extent_io: Move the BUG_ON() in flush_write_bio() one level up Qu Wenruo
2019-02-18 5:27 ` [PATCH v5.1 05/12] btrfs: extent_io: Handle error better in extent_write_full_page() Qu Wenruo
2019-02-18 5:27 ` [PATCH v5.1 06/12] btrfs: extent_io: Handle error better in btree_write_cache_pages() Qu Wenruo
2019-02-18 5:27 ` [PATCH v5.1 07/12] btrfs: extent_io: Kill the dead branch in extent_write_cache_pages() Qu Wenruo
2019-02-18 5:27 ` [PATCH v5.1 08/12] btrfs: extent_io: Handle error better in extent_write_locked_range() Qu Wenruo
2019-02-18 5:27 ` [PATCH v5.1 09/12] btrfs: extent_io: Kill the BUG_ON() in lock_extent_buffer_for_io() Qu Wenruo
2019-02-18 5:27 ` [PATCH v5.1 10/12] btrfs: extent_io: Kill the BUG_ON() in extent_write_cache_pages() Qu Wenruo
2019-03-12 0:33 ` David Sterba
2019-03-12 0:42 ` Qu Wenruo
2019-03-13 11:31 ` David Sterba
2019-03-13 12:02 ` Qu Wenruo
2019-03-15 6:27 ` [PATCH v5.2 " Qu Wenruo
2019-02-18 5:27 ` [PATCH v5.1 11/12] btrfs: extent_io: Handle error better in extent_writepages() Qu Wenruo
2019-02-18 5:27 ` [PATCH v5.1 12/12] btrfs: Do mandatory tree block check before submitting bio Qu Wenruo
2019-02-18 9:26 ` Nikolay Borisov
2019-02-18 9:32 ` Qu Wenruo
2019-02-20 18:11 ` David Sterba
2019-02-20 18:25 ` [PATCH v5.1 00/12] btrfs: Enhancement to tree block validation David Sterba
2019-02-21 0:37 ` Qu Wenruo
2019-02-22 15:38 ` David Sterba
2019-02-21 4:49 ` Qu Wenruo
2019-02-22 15:18 ` David Sterba
2019-02-23 0:47 ` Qu Wenruo
2019-02-27 12:22 ` David Sterba
2019-02-27 13:40 ` Qu Wenruo [this message]
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=88d6815c-1e15-4f89-54f4-bcb6cd2177e6@gmx.com \
--to=quwenruo.btrfs@gmx.com \
--cc=dsterba@suse.cz \
--cc=linux-btrfs@vger.kernel.org \
--cc=wqu@suse.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).