public inbox for linux-btrfs@vger.kernel.org
 help / color / mirror / Atom feed
From: MASAKI Haruka <reasonset@inflaton.co.jp>
To: Andrei Borzenkov <arvidjaar@gmail.com>
Cc: linux-btrfs@vger.kernel.org
Subject: Re: Some file lost when send back snapshot.
Date: Sat, 27 Apr 2019 14:50:53 +0900	[thread overview]
Message-ID: <20190427145053.250bda47@hydrangea> (raw)
In-Reply-To: <8cdd6929-b768-795c-162f-cf41e8b94029@gmail.com>

On Fri, 26 Apr 2019 21:27:52 +0300
Andrei Borzenkov <arvidjaar@gmail.com> wrote:

> 26.04.2019 21:12, MASAKI Haruka пишет:
> >> You have two subvolumes with the same received_uuid. Is the content of
> >> these subvolumes the same? Is "missing" file present in both subvolumes?
> > 
> > No, each subvolume name (190312-010100, 190315-001000) is timestamp.
> > These snapshot volumes were created by job scheduler on original (already lost) volume, so 190315-001000 is modified from 190312-010100.
> > 
> 
> Which is your problem. Two subvolumes with the same received_uuid are
> expected to have exactly the same content (received_uuid can be viewed
> as dataset UUID). You send incremental stream against base uuid
> 9c10b77f-74fb-f94a-9a24-3c6d30c00ca2; on receiving side btrfs receive
> looks for matching base subvolume and happens to pick 190312-010100.
> 
> There were patches to automatically remove received_uuid when subvolume
> is changed to read-write. I do not know whether they were applied.
> 
> btrfs send/receive expects subvolumes to remain read-only. As soon as
> you start fiddling with read-only status you already lost. btrfs should
> have never allowed to change read-only into read-write in the first place.
> 

Thanks it was solved!

I remember change attributes on original volume.

I deleted 190312-010100 and send back1, it's OK.
I learned that shouldn't change subvolume read-only to read-write.

Thank you very much!!!



> > 190315-001000 has lost file,
> > % ls /HySlave/190315-001000/local/pub/media/sound/notmusic_in_CD/
> > Little_My_Maid  SMEE  el_louis  firstkiss  gakuensai  infinity-kid  kid-memories-off  ojyo_tokkyu  samusupi  wizards_harmony  yuukyuu-chara-seala
> > 
> > but 190312-010100 doesn't.
> > % ls /HySlave/190312-010100/local/pub/media/sound/notmusic_in_CD/
> > Little_My_Maid  el_louis  firstkiss  gakuensai  infinity-kid  kid-memories-off  ojyo_tokkyu  samusupi  wizards_harmony  yuukyuu-chara-seala
> > 
> > 
> > 
> > 
> > On Fri, 26 Apr 2019 20:30:07 +0300
> > Andrei Borzenkov <arvidjaar@gmail.com> wrote:
> > 
> >> 26.04.2019 7:50, MASAKI Haruka пишет:
> >>> Thank you for your reply.
> >>>
> >>> # What does
> >>>
> >>> @Host A ("lily" master, new volume)
> >>> aki@lily /mnt % socat TCP-LISTEN:25500 STDOUT | sudo btrfs receive -vv . 
> >>>
> >>> @Host B ("daisy" backup, having volumes)
> >>> aki@daisy /HySlave % sudo btrfs send -v 190315-001000 | socat STDIN TCP:lily.local:25500 
> >>>
> >>> ( ...Transfared 11TB... )
> >>>
> >>> @Host A
> >>> aki@lily /mnt % sudo btrfs subvolume snapshot 190315-001000 world
> >>> Create a snapshot of '190315-001000' in './world'
> >>> aki@lily /mnt % sudo btrfs subvolume snapshot -r world back1
> >>> Create a readonly snapshot of 'world' in './back1'
> >>>
> >>> @Host A
> >>> aki@lily /mnt % sudo btrfs send -vv -p 190315-001000 back1 | socat STDIN tcp-connect:192.168.1.17:25555
> >>> At subvol back1
> >>> BTRFS_IOC_SEND returned 0
> >>> joining genl thread
> >>>
> >>> @Host B
> >>> root@daisy /HySlave # socat TCP-LISTEN:25555 STDOUT | btrfs receive -vv .
> >>> At snapshot back1
> >>> receiving snapshot back1 uuid=32dda1a2-07bd-684f-b773-c5b4cd5f29c1, ctransid=2658 parent_uuid=9c10b77f-74fb-f94a-9a24-3c6d30c00ca2, parent_ctransid=2658
> >>> BTRFS_IOC_SET_RECEIVED_SUBVOL uuid=32dda1a2-07bd-684f-b773-c5b4cd5f29c1, stransid=2658
> >>>
> >>> # Host A current
> >>>
> >>> aki@lily /mnt % ls
> >>> 190315-001000  back1  back2  world
> >>> aki@lily /mnt % sudo btrfs sub li -Rqu .
> >>> ID 258 gen 2698 top level 5 parent_uuid -                                    received_uuid 9c10b77f-74fb-f94a-9a24-3c6d30c00ca2 uuid bc90d05c-7430-7b42-b683-702d906bb993 path 190315-001000
> >>> ID 2139 gen 2696 top level 5 parent_uuid bc90d05c-7430-7b42-b683-702d906bb993 received_uuid -                                    uuid 23c5fede-a338-d14a-afcd-677db2e97637 path world
> >>> ID 2140 gen 2660 top level 5 parent_uuid 23c5fede-a338-d14a-afcd-677db2e97637 received_uuid -                                    uuid 32dda1a2-07bd-684f-b773-c5b4cd5f29c1 path back1
> >>> ID 2142 gen 2672 top level 5 parent_uuid 23c5fede-a338-d14a-afcd-677db2e97637 received_uuid -                                    uuid a8e3e041-c1fc-e443-9baf-ab481ae32b8b path back2
> >>>
> >>>
> >>> # Host B current
> >>>
> >>> aki@daisy /HySlave % pwd
> >>> /HySlave
> >>> aki@daisy /HySlave % ls
> >>> 190312-010100  190315-001000  back1  back2
> >>> [aki@daisy HySlave]$ sudo btrfs sub li -Rqu .
> >>> ID 256 gen 36238 top level 5 parent_uuid -                                    received_uuid -                                    uuid 4e42555d-a715-0e47-bd0e-85d6eb8911e0 path slave
> >>> ID 259 gen 35116 top level 256 parent_uuid -                                    received_uuid 9c10b77f-74fb-f94a-9a24-3c6d30c00ca2 uuid 4a91d585-ab25-0c49-b183-ae951b5377e1 path 190312-010100
> >>> ID 3059 gen 7273 top level 256 parent_uuid 4a91d585-ab25-0c49-b183-ae951b5377e1 received_uuid 9c10b77f-74fb-f94a-9a24-3c6d30c00ca2 uuid 114cd74f-80d8-af44-a9d0-fd8ce3c991b4 path 190315-001000
> >>
> >> You have two subvolumes with the same received_uuid. Is the content of
> >> these subvolumes the same? Is "missing" file present in both subvolumes?
> >>
> >>> ID 3072 gen 35112 top level 256 parent_uuid 4a91d585-ab25-0c49-b183-ae951b5377e1 received_uuid 32dda1a2-07bd-684f-b773-c5b4cd5f29c1 uuid fe7281d1-8b3e-8f46-a0b4-cdc52a1bb7aa path back1
> >>> ID 3073 gen 36236 top level 256 parent_uuid 4a91d585-ab25-0c49-b183-ae951b5377e1 received_uuid a8e3e041-c1fc-e443-9baf-ab481ae32b8b uuid 65206183-f20d-4648-83ab-e0485b16f7af path back2
> >>>
> >>> # Files missing
> >>>
> >>> aki@daisy /HySlave % ls /HySlave/190315-001000/local/pub/media/sound/notmusic_in_CD 
> >>> Little_My_Maid  SMEE  el_louis  firstkiss  gakuensai  infinity-kid  kid-memories-off  ojyo_tokkyu  samusupi  wizards_harmony  yuukyuu-chara-seala
> >>>
> >>> aki@daisy /HySlave % ls /HySlave/back1/local/pub/media/sound/notmusic_in_CD
> >>> Little_My_Maid  el_louis  firstkiss  gakuensai  infinity-kid  kid-memories-off  ojyo_tokkyu  samusupi  wizards_harmony  yuukyuu-chara-seala
> >>>
> >>> aki@lily /mnt % ls /mnt/190315-001000/local/pub/media/sound/notmusic_in_CD
> >>> Little_My_Maid  SMEE  el_louis  firstkiss  gakuensai  infinity-kid  kid-memories-off  ojyo_tokkyu  samusupi  wizards_harmony  yuukyuu-chara-seala
> >>>
> >>> ki@lily /mnt % ls /mnt/back1/local/pub/media/sound/notmusic_in_CD        
> >>> Little_My_Maid  SMEE  el_louis  firstkiss  gakuensai  infinity-kid  kid-memories-off  ojyo_tokkyu  samusupi  wizards_harmony  yuukyuu-chara-seala
> >>>
> >>> aki@lily /mnt % ls /mnt/world/local/pub/media/sound/notmusic_in_CD 
> >>> Little_My_Maid  SMEE  el_louis  firstkiss  gakuensai  infinity-kid  kid-memories-off  ojyo_tokkyu  samusupi  wizards_harmony  yuukyuu-chara-seala
> >>>
> >>>
> >>> SMEE is missing.
> >>>
> >>>
> >>>
> >>>
> >>> Thanks.
> >>>
> >>>
> >>> On Thu, 25 Apr 2019 20:55:59 +0300
> >>> Andrei Borzenkov <arvidjaar@gmail.com> wrote:
> >>>
> >>>> 25.04.2019 16:34, MASAKI Haruka пишет:
> >>>>  [...]  
> >>>>
> >>>> Please show
> >>>>
> >>>> 1. Full protocol of commands you executed (commands and their output),
> >>>> e.g. captured using script.
> >>>> 2. Commands used to determine missing files on both source and receive
> >>>> sides.
> >>>> 3. Listing of subvolumes on both source and receive sides with
> >>>>
> >>>> btrfs sub li -Rqu /mount/point
> >>>>
> >>>>  [...]  
> >>>>
> >>
> 

  reply	other threads:[~2019-04-27  5:50 UTC|newest]

Thread overview: 8+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2019-04-25 13:34 Some file lost when send back snapshot MASAKI Haruka
2019-04-25 17:55 ` Andrei Borzenkov
2019-04-26  4:50   ` MASAKI Haruka
2019-04-26 17:30     ` Andrei Borzenkov
2019-04-26 18:12       ` MASAKI Haruka
2019-04-26 18:27         ` Andrei Borzenkov
2019-04-27  5:50           ` MASAKI Haruka [this message]
2019-04-27 20:54           ` Chris Murphy

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=20190427145053.250bda47@hydrangea \
    --to=reasonset@inflaton.co.jp \
    --cc=arvidjaar@gmail.com \
    --cc=linux-btrfs@vger.kernel.org \
    /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