All of lore.kernel.org
 help / color / mirror / Atom feed
* [U-Boot] corrupted tags in git://git.denx.de/u-boot ?
@ 2015-06-23 12:13 Igor Grinberg
  2015-06-23 13:31 ` Wolfgang Denk
  2015-06-26  2:06 ` Joe Hershberger
  0 siblings, 2 replies; 7+ messages in thread
From: Igor Grinberg @ 2015-06-23 12:13 UTC (permalink / raw)
  To: u-boot

Hi guys,

While running git fsck --full, I've noticed the following:
-------------------------------cut-------------------------------------
$ git fsck --full
Checking object directories: 100% (256/256), done.
error in tag 9bf86baaa3b35b25baa2d664e2f7f6cafad689ee: unterminated header
error in tag c7071e6d645a8e13adb0d4cea2caad27213fa62f: unterminated header
error in tag eb394f56db3e05d00891d6dc36a00df0025cf255: unterminated header
Checking objects: 100% (356677/356677), done.
Checking connectivity: 356680, done.
-------------------------------cut-------------------------------------

So, I did git describe on those:
-------------------------------cut-------------------------------------
$ git describe 9bf86baaa3b35b25baa2d664e2f7f6cafad689ee c7071e6d645a8e13adb0d4cea2caad27213fa62f eb394f56db3e05d00891d6dc36a00df0025cf255
LABEL_2006_04_18_1106
LABEL_2006_05_19_1133
LABEL_2006_03_12_0025
-------------------------------cut-------------------------------------

Having no apparent need of those tags, I just git tag -d them and git gc:
-------------------------------cut-------------------------------------
$ git tag -d LABEL_2006_03_12_0025 LABEL_2006_04_18_1106 LABEL_2006_05_19_1133
Deleted tag 'LABEL_2006_03_12_0025' (was eb394f5)
Deleted tag 'LABEL_2006_04_18_1106' (was 9bf86ba)
Deleted tag 'LABEL_2006_05_19_1133' (was c7071e6)
$ git gc --prune=now
Counting objects: 356677, done.
Delta compression using up to 8 threads.
Compressing objects: 100% (64947/64947), done.
Writing objects: 100% (356677/356677), done.
Total 356677 (delta 285384), reused 356677 (delta 285384)
Checking connectivity: 356677, done.
-------------------------------cut-------------------------------------

After the above git fsck --full got much happier:
-------------------------------cut-------------------------------------
$ git fsck --full 
Checking object directories: 100% (256/256), done.
Checking objects: 100% (356677/356677), done.
Checking connectivity: 356677, done.
-------------------------------cut-------------------------------------

Now, next time I git remote update, I get them back:
-------------------------------cut-------------------------------------
$ git remote update 
Fetching origin
remote: Counting objects: 3, done.
remote: Compressing objects: 100% (3/3), done.
remote: Total 3 (delta 0), reused 3 (delta 0)
Unpacking objects: 100% (3/3), done.
From git://git.denx.de/u-boot
 * [new tag]         LABEL_2006_03_12_0025 -> LABEL_2006_03_12_0025
 * [new tag]         LABEL_2006_04_18_1106 -> LABEL_2006_04_18_1106
 * [new tag]         LABEL_2006_05_19_1133 -> LABEL_2006_05_19_1133
-------------------------------cut-------------------------------------

and once again git fsck --full complains:
-------------------------------cut-------------------------------------
$ git fsck --full
Checking object directories: 100% (256/256), done.
error in tag 9bf86baaa3b35b25baa2d664e2f7f6cafad689ee: unterminated header
error in tag c7071e6d645a8e13adb0d4cea2caad27213fa62f: unterminated header
error in tag eb394f56db3e05d00891d6dc36a00df0025cf255: unterminated header
Checking objects: 100% (356677/356677), done.
Checking connectivity: 356680, done.
-------------------------------cut-------------------------------------

Is it only me suffering from some kind of corruption on remote update
or do we all have the same problem and I'm the first to notice (or just
missed a thread about this)?

Thanks for help!

-- 
Regards,
Igor.

^ permalink raw reply	[flat|nested] 7+ messages in thread

* [U-Boot] corrupted tags in git://git.denx.de/u-boot ?
  2015-06-23 12:13 [U-Boot] corrupted tags in git://git.denx.de/u-boot ? Igor Grinberg
@ 2015-06-23 13:31 ` Wolfgang Denk
  2015-06-23 15:23   ` Igor Grinberg
  2015-06-26  2:06 ` Joe Hershberger
  1 sibling, 1 reply; 7+ messages in thread
From: Wolfgang Denk @ 2015-06-23 13:31 UTC (permalink / raw)
  To: u-boot

Dear Igor,

In message <55894D63.508@compulab.co.il> you wrote:
> 
> While running git fsck --full, I've noticed the following:
> -------------------------------cut-------------------------------------
> $ git fsck --full
> Checking object directories: 100% (256/256), done.
> error in tag 9bf86baaa3b35b25baa2d664e2f7f6cafad689ee: unterminated header
> error in tag c7071e6d645a8e13adb0d4cea2caad27213fa62f: unterminated header
> error in tag eb394f56db3e05d00891d6dc36a00df0025cf255: unterminated header
> Checking objects: 100% (356677/356677), done.
> Checking connectivity: 356680, done.
> -------------------------------cut-------------------------------------

Running this on the master repo gives:

-> git fsck --full
Checking object directories: 100% (256/256), done.
Checking objects: 100% (328644/328644), done.
Checking connectivity: 325719, done.

> Is it only me suffering from some kind of corruption on remote update
> or do we all have the same problem and I'm the first to notice (or just
> missed a thread about this)?

Doing a fresh clone from the public repo gives this:

-> git clone git://git.denx.de/u-boot
Cloning into 'u-boot'...
remote: Counting objects: 325719, done.
remote: Compressing objects: 100% (62600/62600), done.
remote: Total 325719 (delta 260566), reused 322218 (delta 257187)
Receiving objects: 100% (325719/325719), 73.56 MiB | 1.55 MiB/s, done.
Resolving deltas: 100% (260566/260566), done.
Checking connectivity... done.
-> cd u-boot/
/tmp/u-boot
-> git fsck --full
Checking object directories: 100% (256/256), done.
Checking objects: 100% (325719/325719), done.
Checking connectivity: 325719, done.

In other words, I cannot reproduce your issue here.

Best regards,

Wolfgang Denk

-- 
DENX Software Engineering GmbH,      Managing Director: Wolfgang Denk
HRB 165235 Munich, Office: Kirchenstr.5, D-82194 Groebenzell, Germany
Phone: (+49)-8142-66989-10 Fax: (+49)-8142-66989-80 Email: wd at denx.de
The years of peak mental activity are undoubtedly between the ages of
four and eighteen. At four we know all the questions, at eighteen all
the answers.

^ permalink raw reply	[flat|nested] 7+ messages in thread

* [U-Boot] corrupted tags in git://git.denx.de/u-boot ?
  2015-06-23 13:31 ` Wolfgang Denk
@ 2015-06-23 15:23   ` Igor Grinberg
  2015-06-25 13:22     ` Wolfgang Denk
  0 siblings, 1 reply; 7+ messages in thread
From: Igor Grinberg @ 2015-06-23 15:23 UTC (permalink / raw)
  To: u-boot

Hi Wolfgang,

Thanks for the quick reply!

On 06/23/15 16:31, Wolfgang Denk wrote:
> Dear Igor,
> 
> In message <55894D63.508@compulab.co.il> you wrote:
>>
>> While running git fsck --full, I've noticed the following:
>> -------------------------------cut-------------------------------------
>> $ git fsck --full
>> Checking object directories: 100% (256/256), done.
>> error in tag 9bf86baaa3b35b25baa2d664e2f7f6cafad689ee: unterminated header
>> error in tag c7071e6d645a8e13adb0d4cea2caad27213fa62f: unterminated header
>> error in tag eb394f56db3e05d00891d6dc36a00df0025cf255: unterminated header
>> Checking objects: 100% (356677/356677), done.
>> Checking connectivity: 356680, done.
>> -------------------------------cut-------------------------------------
> 
> Running this on the master repo gives:
> 
> -> git fsck --full
> Checking object directories: 100% (256/256), done.
> Checking objects: 100% (328644/328644), done.
> Checking connectivity: 325719, done.
> 
>> Is it only me suffering from some kind of corruption on remote update
>> or do we all have the same problem and I'm the first to notice (or just
>> missed a thread about this)?
> 
> Doing a fresh clone from the public repo gives this:
> 
> -> git clone git://git.denx.de/u-boot
> Cloning into 'u-boot'...
> remote: Counting objects: 325719, done.
> remote: Compressing objects: 100% (62600/62600), done.
> remote: Total 325719 (delta 260566), reused 322218 (delta 257187)
> Receiving objects: 100% (325719/325719), 73.56 MiB | 1.55 MiB/s, done.
> Resolving deltas: 100% (260566/260566), done.
> Checking connectivity... done.
> -> cd u-boot/
> /tmp/u-boot
> -> git fsck --full
> Checking object directories: 100% (256/256), done.
> Checking objects: 100% (325719/325719), done.
> Checking connectivity: 325719, done.
> 
> In other words, I cannot reproduce your issue here.

May be it has something to do with my git version?
$ git --version
git version 2.3.6

I performed the clone test too and it gets the same error:
$ git clone git://git.denx.de/u-boot test
Cloning into 'test'...
remote: Counting objects: 325719, done.
remote: Compressing objects: 100% (62600/62600), done.
remote: Total 325719 (delta 260566), reused 322218 (delta 257187)
Receiving objects: 100% (325719/325719), 73.56 MiB | 2.48 MiB/s, done.
Resolving deltas: 100% (260566/260566), done.
Checking connectivity... done.
$ cd test/
$ git fsck --full
Checking object directories: 100% (256/256), done.
error in tag eb394f56db3e05d00891d6dc36a00df0025cf255: unterminated header
error in tag 9bf86baaa3b35b25baa2d664e2f7f6cafad689ee: unterminated header
error in tag c7071e6d645a8e13adb0d4cea2caad27213fa62f: unterminated header
Checking objects: 100% (325719/325719), done.
Checking connectivity: 325719, done.

It seems that the check for termination with '\n' in tag headers
started to be a requirement from git version 2.2.0?
This [1] seems to be the patch that triggers the message I see.

[1] https://github.com/git/git/commit/4d0d89755e82c40df88cf94d84031978f8eac827

-- 
Regards,
Igor.

^ permalink raw reply	[flat|nested] 7+ messages in thread

* [U-Boot] corrupted tags in git://git.denx.de/u-boot ?
  2015-06-23 15:23   ` Igor Grinberg
@ 2015-06-25 13:22     ` Wolfgang Denk
  2015-06-25 13:27       ` Tom Rini
  0 siblings, 1 reply; 7+ messages in thread
From: Wolfgang Denk @ 2015-06-25 13:22 UTC (permalink / raw)
  To: u-boot

Dear Igor,

In message <558979E4.9030501@compulab.co.il> you wrote:
> 
> > In other words, I cannot reproduce your issue here.
> 
> May be it has something to do with my git version?
> $ git --version
> git version 2.3.6

I did my tests with:

-> git --version
git version 2.1.0

[This is what Fedora 21 uses.]


In Fedora 22 I get

-> git --version
git version 2.4.2

which then shows:

-> git fsck --full
Checking object directories: 100% (256/256), done.
error in tag eb394f56db3e05d00891d6dc36a00df0025cf255: unterminated header
error in tag 9bf86baaa3b35b25baa2d664e2f7f6cafad689ee: unterminated header
error in tag c7071e6d645a8e13adb0d4cea2caad27213fa62f: unterminated header
Checking objects: 100% (328644/328644), done.
Checking connectivity: 325719, done.


> $ git fsck --full
> Checking object directories: 100% (256/256), done.
> error in tag eb394f56db3e05d00891d6dc36a00df0025cf255: unterminated header
> error in tag 9bf86baaa3b35b25baa2d664e2f7f6cafad689ee: unterminated header
> error in tag c7071e6d645a8e13adb0d4cea2caad27213fa62f: unterminated header
> Checking objects: 100% (325719/325719), done.
> Checking connectivity: 325719, done.

Confirmed.

> It seems that the check for termination with '\n' in tag headers
> started to be a requirement from git version 2.2.0?

I think you are right.  The fact that v2.1.0 does not show anything
but v2.3.6 and v2.4.2 do also supports your analysis.

> This [1] seems to be the patch that triggers the message I see.
> 
> [1] https://github.com/git/git/commit/4d0d89755e82c40df88cf94d84031978f8eac827

Yes, I agree.

So apparently the tags  LABEL_2006_03_12_0025, LABEL_2006_04_18_1106,
and LABEL_2006_05_19_1133 are affected.

IIUC it should be a harmless thing to just recreate the same tags
(referring to the same commits)?  Or can anybody see any problems tha
might be caused by such a tag re-creation?  I mean, it is not like a
rebase of the repository or something like that?  Right?


Best regards,

Wolfgang Denk

-- 
DENX Software Engineering GmbH,      Managing Director: Wolfgang Denk
HRB 165235 Munich, Office: Kirchenstr.5, D-82194 Groebenzell, Germany
Phone: (+49)-8142-66989-10 Fax: (+49)-8142-66989-80 Email: wd at denx.de
C++ was an interesting and valuable experiment, but we've learned its
lessons and it's time to move on.
                            - Peter Curran in <DCqM4z.BxB@isgtec.com>

^ permalink raw reply	[flat|nested] 7+ messages in thread

* [U-Boot] corrupted tags in git://git.denx.de/u-boot ?
  2015-06-25 13:22     ` Wolfgang Denk
@ 2015-06-25 13:27       ` Tom Rini
  2015-06-26  6:12         ` Wolfgang Denk
  0 siblings, 1 reply; 7+ messages in thread
From: Tom Rini @ 2015-06-25 13:27 UTC (permalink / raw)
  To: u-boot

On Thu, Jun 25, 2015 at 03:22:21PM +0200, Wolfgang Denk wrote:
> Dear Igor,
> 
> In message <558979E4.9030501@compulab.co.il> you wrote:
> > 
> > > In other words, I cannot reproduce your issue here.
> > 
> > May be it has something to do with my git version?
> > $ git --version
> > git version 2.3.6
> 
> I did my tests with:
> 
> -> git --version
> git version 2.1.0
> 
> [This is what Fedora 21 uses.]
> 
> 
> In Fedora 22 I get
> 
> -> git --version
> git version 2.4.2
> 
> which then shows:
> 
> -> git fsck --full
> Checking object directories: 100% (256/256), done.
> error in tag eb394f56db3e05d00891d6dc36a00df0025cf255: unterminated header
> error in tag 9bf86baaa3b35b25baa2d664e2f7f6cafad689ee: unterminated header
> error in tag c7071e6d645a8e13adb0d4cea2caad27213fa62f: unterminated header
> Checking objects: 100% (328644/328644), done.
> Checking connectivity: 325719, done.
> 
> 
> > $ git fsck --full
> > Checking object directories: 100% (256/256), done.
> > error in tag eb394f56db3e05d00891d6dc36a00df0025cf255: unterminated header
> > error in tag 9bf86baaa3b35b25baa2d664e2f7f6cafad689ee: unterminated header
> > error in tag c7071e6d645a8e13adb0d4cea2caad27213fa62f: unterminated header
> > Checking objects: 100% (325719/325719), done.
> > Checking connectivity: 325719, done.
> 
> Confirmed.
> 
> > It seems that the check for termination with '\n' in tag headers
> > started to be a requirement from git version 2.2.0?
> 
> I think you are right.  The fact that v2.1.0 does not show anything
> but v2.3.6 and v2.4.2 do also supports your analysis.
> 
> > This [1] seems to be the patch that triggers the message I see.
> > 
> > [1] https://github.com/git/git/commit/4d0d89755e82c40df88cf94d84031978f8eac827
> 
> Yes, I agree.
> 
> So apparently the tags  LABEL_2006_03_12_0025, LABEL_2006_04_18_1106,
> and LABEL_2006_05_19_1133 are affected.
> 
> IIUC it should be a harmless thing to just recreate the same tags
> (referring to the same commits)?  Or can anybody see any problems tha
> might be caused by such a tag re-creation?  I mean, it is not like a
> rebase of the repository or something like that?  Right?

Lets take it up on the git list and ask for help/advice?

-- 
Tom
-------------- next part --------------
A non-text attachment was scrubbed...
Name: signature.asc
Type: application/pgp-signature
Size: 836 bytes
Desc: Digital signature
URL: <http://lists.denx.de/pipermail/u-boot/attachments/20150625/05f65b57/attachment.sig>

^ permalink raw reply	[flat|nested] 7+ messages in thread

* [U-Boot] corrupted tags in git://git.denx.de/u-boot ?
  2015-06-23 12:13 [U-Boot] corrupted tags in git://git.denx.de/u-boot ? Igor Grinberg
  2015-06-23 13:31 ` Wolfgang Denk
@ 2015-06-26  2:06 ` Joe Hershberger
  1 sibling, 0 replies; 7+ messages in thread
From: Joe Hershberger @ 2015-06-26  2:06 UTC (permalink / raw)
  To: u-boot

Hi Igor,

On Tue, Jun 23, 2015 at 7:13 AM, Igor Grinberg <grinberg@compulab.co.il> wrote:
> Hi guys,
>
> While running git fsck --full, I've noticed the following:
> -------------------------------cut-------------------------------------
> $ git fsck --full
> Checking object directories: 100% (256/256), done.
> error in tag 9bf86baaa3b35b25baa2d664e2f7f6cafad689ee: unterminated header
> error in tag c7071e6d645a8e13adb0d4cea2caad27213fa62f: unterminated header
> error in tag eb394f56db3e05d00891d6dc36a00df0025cf255: unterminated header
> Checking objects: 100% (356677/356677), done.
> Checking connectivity: 356680, done.
> -------------------------------cut-------------------------------------
>
> So, I did git describe on those:
> -------------------------------cut-------------------------------------
> $ git describe 9bf86baaa3b35b25baa2d664e2f7f6cafad689ee c7071e6d645a8e13adb0d4cea2caad27213fa62f eb394f56db3e05d00891d6dc36a00df0025cf255
> LABEL_2006_04_18_1106
> LABEL_2006_05_19_1133
> LABEL_2006_03_12_0025
> -------------------------------cut-------------------------------------
>
> Having no apparent need of those tags, I just git tag -d them and git gc:
> -------------------------------cut-------------------------------------
> $ git tag -d LABEL_2006_03_12_0025 LABEL_2006_04_18_1106 LABEL_2006_05_19_1133
> Deleted tag 'LABEL_2006_03_12_0025' (was eb394f5)
> Deleted tag 'LABEL_2006_04_18_1106' (was 9bf86ba)
> Deleted tag 'LABEL_2006_05_19_1133' (was c7071e6)
> $ git gc --prune=now
> Counting objects: 356677, done.
> Delta compression using up to 8 threads.
> Compressing objects: 100% (64947/64947), done.
> Writing objects: 100% (356677/356677), done.
> Total 356677 (delta 285384), reused 356677 (delta 285384)
> Checking connectivity: 356677, done.
> -------------------------------cut-------------------------------------
>
> After the above git fsck --full got much happier:
> -------------------------------cut-------------------------------------
> $ git fsck --full
> Checking object directories: 100% (256/256), done.
> Checking objects: 100% (356677/356677), done.
> Checking connectivity: 356677, done.
> -------------------------------cut-------------------------------------
>
> Now, next time I git remote update, I get them back:
> -------------------------------cut-------------------------------------
> $ git remote update
> Fetching origin
> remote: Counting objects: 3, done.
> remote: Compressing objects: 100% (3/3), done.
> remote: Total 3 (delta 0), reused 3 (delta 0)
> Unpacking objects: 100% (3/3), done.
> From git://git.denx.de/u-boot
>  * [new tag]         LABEL_2006_03_12_0025 -> LABEL_2006_03_12_0025
>  * [new tag]         LABEL_2006_04_18_1106 -> LABEL_2006_04_18_1106
>  * [new tag]         LABEL_2006_05_19_1133 -> LABEL_2006_05_19_1133
> -------------------------------cut-------------------------------------
>
> and once again git fsck --full complains:
> -------------------------------cut-------------------------------------
> $ git fsck --full
> Checking object directories: 100% (256/256), done.
> error in tag 9bf86baaa3b35b25baa2d664e2f7f6cafad689ee: unterminated header
> error in tag c7071e6d645a8e13adb0d4cea2caad27213fa62f: unterminated header
> error in tag eb394f56db3e05d00891d6dc36a00df0025cf255: unterminated header
> Checking objects: 100% (356677/356677), done.
> Checking connectivity: 356680, done.
> -------------------------------cut-------------------------------------
>
> Is it only me suffering from some kind of corruption on remote update
> or do we all have the same problem and I'm the first to notice (or just
> missed a thread about this)?

I don't see this either.

Have you tried a fresh clone?

Thanks,
-Joe

^ permalink raw reply	[flat|nested] 7+ messages in thread

* [U-Boot] corrupted tags in git://git.denx.de/u-boot ?
  2015-06-25 13:27       ` Tom Rini
@ 2015-06-26  6:12         ` Wolfgang Denk
  0 siblings, 0 replies; 7+ messages in thread
From: Wolfgang Denk @ 2015-06-26  6:12 UTC (permalink / raw)
  To: u-boot

Dear Tom,
dear Igor,

In message <20150625132731.GC28577@bill-the-cat> you wrote:
> 
> > > It seems that the check for termination with '\n' in tag headers
> > > started to be a requirement from git version 2.2.0?
...
> > So apparently the tags  LABEL_2006_03_12_0025, LABEL_2006_04_18_1106,
> > and LABEL_2006_05_19_1133 are affected.
...
> > IIUC it should be a harmless thing to just recreate the same tags
> > (referring to the same commits)?  Or can anybody see any problems tha
> > might be caused by such a tag re-creation?  I mean, it is not like a
> > rebase of the repository or something like that?  Right?
> 
> Lets take it up on the git list and ask for help/advice?

Well, the summary of the comments on the git list is like this:

        I am inclined to say that these messages are that recent Git
        barking at a non-errors, and hopefully there is nothing other
        than ignoring them you have to do until it gets fixed.

See [1] for more details.

At the moment I tend to just wait for the results of the ongoing
discussion,  It appears to me that we might need no changes to the
existing repo.

[1] http://thread.gmane.org/gmane.comp.version-control.git/272689/focus=272704

Best regards,

Wolfgang Denk

-- 
DENX Software Engineering GmbH,      Managing Director: Wolfgang Denk
HRB 165235 Munich, Office: Kirchenstr.5, D-82194 Groebenzell, Germany
Phone: (+49)-8142-66989-10 Fax: (+49)-8142-66989-80 Email: wd at denx.de
What is mind?  No matter.  What is matter?  Never mind.
                                      -- Thomas Hewitt Key, 1799-1875

^ permalink raw reply	[flat|nested] 7+ messages in thread

end of thread, other threads:[~2015-06-26  6:12 UTC | newest]

Thread overview: 7+ messages (download: mbox.gz / follow: Atom feed)
-- links below jump to the message on this page --
2015-06-23 12:13 [U-Boot] corrupted tags in git://git.denx.de/u-boot ? Igor Grinberg
2015-06-23 13:31 ` Wolfgang Denk
2015-06-23 15:23   ` Igor Grinberg
2015-06-25 13:22     ` Wolfgang Denk
2015-06-25 13:27       ` Tom Rini
2015-06-26  6:12         ` Wolfgang Denk
2015-06-26  2:06 ` Joe Hershberger

This is an external index of several public inboxes,
see mirroring instructions on how to clone and mirror
all data and code used by this external index.