Transaction commit: none (default)

Transaction commit: none (default)

[Marc MERLIN]

Taking snapshots is now output this in addition of the snapshot
operation.

Transaction commit: none (default)

1) Is that expected/normal? It looks kind of spamming/useless to me?
2) If it's useful, what's the use I'm not getting?

Thanks,
Marc
-- 
"A mouse is a device used to point at the xterm you want to type in" - A.S.R.
Microsoft is to operating systems ....
                                      .... what McDonalds is to gourmet cooking
Home page: http://marc.merlins.org/                         | PGP 1024R/763BE901

Re: Transaction commit: none (default)

[Wang Shilong]

On 06/15/2014 12:05 AM, Marc MERLIN wrote:
> Taking snapshots is now output this in addition of the snapshot
> operation.
>
> Transaction commit: none (default)
>
> 1) Is that expected/normal? It looks kind of spamming/useless to me?
> 2) If it's useful, what's the use I'm not getting?
Addressed a issue that subvolumes reappear after deletetion
if poweroff happen.

http://www.spinics.net/lists/linux-btrfs/msg29547.html
>
> Thanks,
> Marc

Re: Transaction commit: none (default)

[Marc MERLIN]

On Mon, Jun 16, 2014 at 09:35:23AM +0800, Wang Shilong wrote:
> On 06/15/2014 12:05 AM, Marc MERLIN wrote:
> >Taking snapshots is now output this in addition of the snapshot
> >operation.
> >
> >Transaction commit: none (default)
> >
> >1) Is that expected/normal? It looks kind of spamming/useless to me?
> >2) If it's useful, what's the use I'm not getting?
> Addressed a issue that subvolumes reappear after deletetion
> if poweroff happen.
> 
> http://www.spinics.net/lists/linux-btrfs/msg29547.html

I see, thank you for the pointer.
In the case of my script, it's console spam that I wish I could make go
away, but grep -v will do that for me :)

If it annoys other people, maybe having a -q (quiet) that removes this
message would be useful?

Thanks,
Marc
-- 
"A mouse is a device used to point at the xterm you want to type in" - A.S.R.
Microsoft is to operating systems ....
                                      .... what McDonalds is to gourmet cooking
Home page: http://marc.merlins.org/  

Re: Transaction commit: none (default)

[Holger Hoffstätte]

On Sun, 15 Jun 2014 18:47:39 -0700, Marc MERLIN wrote:

> On Mon, Jun 16, 2014 at 09:35:23AM +0800, Wang Shilong wrote:
>> On 06/15/2014 12:05 AM, Marc MERLIN wrote:
>> >Taking snapshots is now output this in addition of the snapshot
>> >operation.
>> >
>> >Transaction commit: none (default)
>> >
>> >1) Is that expected/normal? It looks kind of spamming/useless to me?
>> >2) If it's useful, what's the use I'm not getting?
>> Addressed a issue that subvolumes reappear after deletetion if poweroff
>> happen.
>> 
>> http://www.spinics.net/lists/linux-btrfs/msg29547.html
> 
> I see, thank you for the pointer.
> In the case of my script, it's console spam that I wish I could make go
> away, but grep -v will do that for me :)
> 
> If it annoys other people, maybe having a -q (quiet) that removes this
> message would be useful?

IMHO it should be the reverse: be quiet by default and require -v, just 
as more explicit behaviour also requires additional options.
With default options I expect default behaviour. Telling me that the 
default behaviour happened is just noise.

cheers
Holger

Re: Transaction commit: none (default)

[Wang Shilong]

On 06/16/2014 03:06 PM, Holger Hoffstätte wrote:
> On Sun, 15 Jun 2014 18:47:39 -0700, Marc MERLIN wrote:
>
>> On Mon, Jun 16, 2014 at 09:35:23AM +0800, Wang Shilong wrote:
>>> On 06/15/2014 12:05 AM, Marc MERLIN wrote:
>>>> Taking snapshots is now output this in addition of the snapshot
>>>> operation.
>>>>
>>>> Transaction commit: none (default)
>>>>
>>>> 1) Is that expected/normal? It looks kind of spamming/useless to me?
>>>> 2) If it's useful, what's the use I'm not getting?
>>> Addressed a issue that subvolumes reappear after deletetion if poweroff
>>> happen.
>>>
>>> http://www.spinics.net/lists/linux-btrfs/msg29547.html
>> I see, thank you for the pointer.
>> In the case of my script, it's console spam that I wish I could make go
>> away, but grep -v will do that for me :)
>>
>> If it annoys other people, maybe having a -q (quiet) that removes this
>> message would be useful?
> IMHO it should be the reverse: be quiet by default and require -v, just
> as more explicit behaviour also requires additional options.
> With default options I expect default behaviour. Telling me that the
> default behaviour happened is just noise.
Fair point, Another question is that current default output has been there.
Let's Cc David, and see what is his opinion.

>
> cheers
> Holger
>
> --
> To unsubscribe from this list: send the line "unsubscribe linux-btrfs" in
> the body of a message to majordomo@vger.kernel.org
> More majordomo info at  http://vger.kernel.org/majordomo-info.html
>

Re: Transaction commit: none (default)

[Marc MERLIN]

On Mon, Jun 16, 2014 at 07:06:08AM +0000, Holger Hoffstätte wrote:
> > If it annoys other people, maybe having a -q (quiet) that removes this
> > message would be useful?
> 
> IMHO it should be the reverse: be quiet by default and require -v, just 
> as more explicit behaviour also requires additional options.
> With default options I expect default behaviour. Telling me that the 
> default behaviour happened is just noise.

So that's my general thought by default too :) but 
1) I wanted to be nice :)
2) More generally I would have missed this new feature without the output
and I thought it might not be unreasonable to show it to draw attention to
it. But at the same time, I wouldn't complain if it were quiet by default.

On Mon, Jun 16, 2014 at 03:09:53PM +0800, Wang Shilong wrote:
> >as more explicit behaviour also requires additional options.
> >With default options I expect default behaviour. Telling me that the
> >default behaviour happened is just noise.
>
> Fair point, Another question is that current default output has been there.
> Let's Cc David, and see what is his opinion.

Thanks for taking the feedback.

Marc
-- 
"A mouse is a device used to point at the xterm you want to type in" - A.S.R.
Microsoft is to operating systems ....
                                      .... what McDonalds is to gourmet cooking
Home page: http://marc.merlins.org/  

Re: Transaction commit: none (default)

[David Sterba]

On Mon, Jun 16, 2014 at 03:09:53PM +0800, Wang Shilong wrote:
> >>>>1) Is that expected/normal? It looks kind of spamming/useless to me?
> >>>>2) If it's useful, what's the use I'm not getting?
> >>>Addressed a issue that subvolumes reappear after deletetion if poweroff
> >>>happen.
> >>>
> >>>http://www.spinics.net/lists/linux-btrfs/msg29547.html
> >>I see, thank you for the pointer.
> >>In the case of my script, it's console spam that I wish I could make go
> >>away, but grep -v will do that for me :)
> >>
> >>If it annoys other people, maybe having a -q (quiet) that removes this
> >>message would be useful?
> >IMHO it should be the reverse: be quiet by default and require -v, just
> >as more explicit behaviour also requires additional options.
> >With default options I expect default behaviour. Telling me that the
> >default behaviour happened is just noise.
> Fair point, Another question is that current default output has been there.
> Let's Cc David, and see what is his opinion.

I agree the way it's now (2 lines of output per 1 subvolume) is not
good, but I'd like to keep the information about the commit mode, even
for the default one. So would something like this work?

current:
Delete subvolume '/mnt/subvol'

new:
Delete subvolume (no commit): '/mnt/subvol'

or
Delete subvolume (commit): '/mnt/subvol'

Where the modes of commit (none, each) would stick to each line, and the
'commit after last' would say 'commit' for the last one. Ie. no separate
line that mentions the commit mode globally.

Re: Transaction commit: none (default)

[Marc MERLIN]

On Wed, Jun 18, 2014 at 04:19:32PM +0200, David Sterba wrote:
> On Mon, Jun 16, 2014 at 03:09:53PM +0800, Wang Shilong wrote:
> > >>>>1) Is that expected/normal? It looks kind of spamming/useless to me?
> > >>>>2) If it's useful, what's the use I'm not getting?
> > >>>Addressed a issue that subvolumes reappear after deletetion if poweroff
> > >>>happen.
> > >>>
> > >>>http://www.spinics.net/lists/linux-btrfs/msg29547.html
> > >>I see, thank you for the pointer.
> > >>In the case of my script, it's console spam that I wish I could make go
> > >>away, but grep -v will do that for me :)
> > >>
> > >>If it annoys other people, maybe having a -q (quiet) that removes this
> > >>message would be useful?
> > >IMHO it should be the reverse: be quiet by default and require -v, just
> > >as more explicit behaviour also requires additional options.
> > >With default options I expect default behaviour. Telling me that the
> > >default behaviour happened is just noise.
> > Fair point, Another question is that current default output has been there.
> > Let's Cc David, and see what is his opinion.
> 
> I agree the way it's now (2 lines of output per 1 subvolume) is not
> good, but I'd like to keep the information about the commit mode, even
> for the default one. So would something like this work?
> 
> current:
> Delete subvolume '/mnt/subvol'
> 
> new:
> Delete subvolume (no commit): '/mnt/subvol'
> 
> or
> Delete subvolume (commit): '/mnt/subvol'

I like this. The old line telling me it did what I just asked it to was kind
of redundant (rm doesn't tell me "I just deleted the file you just asked me
to" unless I use -v, and neither does rmdir).

So making the single line of output more useful would alleviate that a bit.
Although in general and in the future a command should indeed not be verbose
and reply "I just did what yo asked me to" :)
(just for the record, 

Note that my crontabs already have stuff like this. I really wish I didn't
need them, and hope I don't have to extend the egrep line:
btrfs-snaps weekly 2 varlocalspace | egrep -v '(Create a snapshot of|Delete subvolume)'

Thanks,
Marc
-- 
"A mouse is a device used to point at the xterm you want to type in" - A.S.R.
Microsoft is to operating systems ....
                                      .... what McDonalds is to gourmet cooking
Home page: http://marc.merlins.org/