corrupt local repository

classic Classic list List threaded Threaded
10 messages Options
Reply | Threaded
Open this post in threaded view
|

corrupt local repository

Uwe Brauer

Hi

I started this morning a new repository and just found out that it is
corrupt, the first time since a long long time[1]

 hg verify gives
,----
| Agrup4/zip4/Calificaciones-19-221172-Entrega_Examen_4_1130-1200-4748998.csv@0: manifest refers to unknown revision d2eca500b6a6
| checked 13 changesets with 71 changes to 102 files
| 39 warnings encountered!
| hint: run "hg debugrebuildfncache" to recover from corrupt fncache
| 81 integrity errors encountered!
| (first damaged changeset appears to be 0)
`----

(first damaged changeset appears to be 0) sounds really bad
hg debugrebuildfncache

Does not really help and I cannot commit any changes so, I am not sure
what to do, any advice would be welcome.

 Hg clone --pull repo repo-repair

abort: empty or missing file data for Agrup1/Calificaciones-Agrup1.org


As a method of last resort

hg convert --config convert.hg.ignoreerrors=True Corr/ Repair

Worked not sure how much data I lost

Regards

Uwe Brauer

Footnotes:
[1]  in the past it happened when I used rscync to copy a repository to
     an USB stick


_______________________________________________
Mercurial mailing list
[hidden email]
https://www.mercurial-scm.org/mailman/listinfo/mercurial
Reply | Threaded
Open this post in threaded view
|

Re: corrupt local repository

Pierre-Yves David-2
Sounds like you have at least two different files whose data are
missing. How have you been using your repository ? is the repository
server side corrupted ?

On 6/8/20 5:20 PM, Uwe Brauer wrote:

>
> Hi
>
> I started this morning a new repository and just found out that it is
> corrupt, the first time since a long long time[1]
>
>   hg verify gives
> ,----
> | Agrup4/zip4/Calificaciones-19-221172-Entrega_Examen_4_1130-1200-4748998.csv@0: manifest refers to unknown revision d2eca500b6a6
> | checked 13 changesets with 71 changes to 102 files
> | 39 warnings encountered!
> | hint: run "hg debugrebuildfncache" to recover from corrupt fncache
> | 81 integrity errors encountered!
> | (first damaged changeset appears to be 0)
> `----
>
> (first damaged changeset appears to be 0) sounds really bad
> hg debugrebuildfncache
>
> Does not really help and I cannot commit any changes so, I am not sure
> what to do, any advice would be welcome.
>
>   Hg clone --pull repo repo-repair
>
> abort: empty or missing file data for Agrup1/Calificaciones-Agrup1.org
>
>
> As a method of last resort
>
> hg convert --config convert.hg.ignoreerrors=True Corr/ Repair
>
> Worked not sure how much data I lost
>
> Regards
>
> Uwe Brauer
>
> Footnotes:
> [1]  in the past it happened when I used rscync to copy a repository to
>       an USB stick
>
>
> _______________________________________________
> Mercurial mailing list
> [hidden email]
> https://www.mercurial-scm.org/mailman/listinfo/mercurial
>

--
Pierre-Yves David
_______________________________________________
Mercurial mailing list
[hidden email]
https://www.mercurial-scm.org/mailman/listinfo/mercurial
Reply | Threaded
Open this post in threaded view
|

Re: corrupt local repository

Uwe Brauer
>>> "PD" == Pierre-Yves David <[hidden email]> writes:

   > Sounds like you have at least two different files whose data are
   > missing. How have you been using your repository ? is the repository
   > server side corrupted ?

I create this repository this morning and just committed around 10
times.
However I did a lot of renaming  of directories and sometimes I
committed without having run hg addremove before, may that is the
reason.

I also found out that
hg convert --config convert.hg.ignoreerrors=True Corr/ Repair

Is basically worseless, the resulting repo contains only half of my
commits.

I am sort of desperate.


_______________________________________________
Mercurial mailing list
[hidden email]
https://www.mercurial-scm.org/mailman/listinfo/mercurial

smime.p7s (7K) Download Attachment
Reply | Threaded
Open this post in threaded view
|

Re: corrupt local repository

Pierre-Yves David-2
Is this as new repostiory that you never pushed anywhere ?
What is the underlying file system ?
Do you have any build/cleanup script running as part of your development
? possibly deleting file automatically ?

On 6/8/20 8:42 PM, Uwe Brauer wrote:

>>>> "PD" == Pierre-Yves David <[hidden email]> writes:
>
>     > Sounds like you have at least two different files whose data are
>     > missing. How have you been using your repository ? is the repository
>     > server side corrupted ?
>
> I create this repository this morning and just committed around 10
> times.
> However I did a lot of renaming  of directories and sometimes I
> committed without having run hg addremove before, may that is the
> reason.

That should not matter. Do you use symlink in the mix ?


> I also found out that
> hg convert --config convert.hg.ignoreerrors=True Corr/ Repair
>
> Is basically worseless, the resulting repo contains only half of my
> commits.

Well, it cannot "invent" the missing content, so I guess it simply
dropped the associated commit


--
Pierre-Yves David
_______________________________________________
Mercurial mailing list
[hidden email]
https://www.mercurial-scm.org/mailman/listinfo/mercurial
Reply | Threaded
Open this post in threaded view
|

Re: corrupt local repository

Uwe Brauer
>>> "PD" == Pierre-Yves David <[hidden email]> writes:

   > Is this as new repostiory that you never pushed anywhere ?

Never ever.

   > What is the underlying file system ?

Ext4 (Ubuntu 16.04)

   > Do you have any build/cleanup script running as part of your
   > development ? possibly deleting file automatically ?

Nope,
   > On 6/8/20 8:42 PM, Uwe Brauer wrote:
   >>>>> "PD" == Pierre-Yves David <[hidden email]> writes:
   >> > Sounds like you have at least two different files whose data
   >> are
   >> > missing. How have you been using your repository ? is the repository
   >> > server side corrupted ?
   >> I create this repository this morning and just committed around 10
   >> times.
   >> However I did a lot of renaming  of directories and sometimes I
   >> committed without having run hg addremove before, may that is the
   >> reason.

   > That should not matter. Do you use symlink in the mix ?

No


   >> I also found out that
   >> hg convert --config convert.hg.ignoreerrors=True Corr/ Repair
   >> Is basically worseless, the resulting repo contains only half of my
   >> commits.

   > Well, it cannot "invent" the missing content, so I guess it simply
   > dropped the associated commit

So it seems that this repository cannot be recovered.

I am running
hg version 5.2rc0+20200125

And the last time I suffered a repository corruption is I think with
3.8.

But it seems difficult to track down and to know how to avoid this in
the feature.  I still think it has to do with the amount of renaming I
did, but I have no proof and no recipe.

_______________________________________________
Mercurial mailing list
[hidden email]
https://www.mercurial-scm.org/mailman/listinfo/mercurial

smime.p7s (7K) Download Attachment
Reply | Threaded
Open this post in threaded view
|

Re: corrupt local repository

Pierre-Yves David-2
That's very odd, it seems like a fresh corruption of a very new
repository I have head of that in a very long time.

I see you are using a rc version, can you move the latest official version ?

Is there anything else strange that happened like a power failure on the
machine or something similar ? Any process being abrutly killed ? (not
that this point should matters ?)

On 6/10/20 5:55 PM, Uwe Brauer wrote:

>>>> "PD" == Pierre-Yves David <[hidden email]> writes:
>
>     > Is this as new repostiory that you never pushed anywhere ?
>
> Never ever.
>
>     > What is the underlying file system ?
>
> Ext4 (Ubuntu 16.04)
>
>     > Do you have any build/cleanup script running as part of your
>     > development ? possibly deleting file automatically ?
>
> Nope,
>     > On 6/8/20 8:42 PM, Uwe Brauer wrote:
>     >>>>> "PD" == Pierre-Yves David <[hidden email]> writes:
>     >> > Sounds like you have at least two different files whose data
>     >> are
>     >> > missing. How have you been using your repository ? is the repository
>     >> > server side corrupted ?
>     >> I create this repository this morning and just committed around 10
>     >> times.
>     >> However I did a lot of renaming  of directories and sometimes I
>     >> committed without having run hg addremove before, may that is the
>     >> reason.
>
>     > That should not matter. Do you use symlink in the mix ?
>
> No
>
>
>     >> I also found out that
>     >> hg convert --config convert.hg.ignoreerrors=True Corr/ Repair
>     >> Is basically worseless, the resulting repo contains only half of my
>     >> commits.
>
>     > Well, it cannot "invent" the missing content, so I guess it simply
>     > dropped the associated commit
>
> So it seems that this repository cannot be recovered.
>
> I am running
> hg version 5.2rc0+20200125
>
> And the last time I suffered a repository corruption is I think with
> 3.8.
>
> But it seems difficult to track down and to know how to avoid this in
> the feature.  I still think it has to do with the amount of renaming I
> did, but I have no proof and no recipe.
>
>
> _______________________________________________
> Mercurial mailing list
> [hidden email]
> https://www.mercurial-scm.org/mailman/listinfo/mercurial
>

--
Pierre-Yves David
_______________________________________________
Mercurial mailing list
[hidden email]
https://www.mercurial-scm.org/mailman/listinfo/mercurial
Reply | Threaded
Open this post in threaded view
|

Re: corrupt local repository

Uwe Brauer
>>> "PD" == Pierre-Yves David <[hidden email]> writes:

   > That's very odd, it seems like a fresh corruption of a very new
   > repository I have head of that in a very long time.

   > I see you are using a rc version, can you move the latest official version ?

Ok, please remind me which would that be?


   > Is there anything else strange that happened like a power failure on
   > the machine or something similar ? Any process being abrutly killed ?
   > (not that this point should matters ?)

No, nothing I can think off. (As I said the only thing I can think of is
that I did quite a lot of directory renaming and, as I must say a lot of
interactive commits (record) using the nurses interface. (Shouldn't
influence).

_______________________________________________
Mercurial mailing list
[hidden email]
https://www.mercurial-scm.org/mailman/listinfo/mercurial

smime.p7s (7K) Download Attachment
Reply | Threaded
Open this post in threaded view
|

Re: corrupt local repository

Pierre-Yves David-2


On 6/10/20 9:35 PM, Uwe Brauer wrote:
>>>> "PD" == Pierre-Yves David <[hidden email]> writes:
>
>     > That's very odd, it seems like a fresh corruption of a very new
>     > repository I have head of that in a very long time.
>
>     > I see you are using a rc version, can you move the latest official version ?
>
> Ok, please remind me which would that be?

Latest version is 5.4.1 Being at least on 5.1 would be better too.

Which version of python is used by you 5.2rc0 install ?

>     > Is there anything else strange that happened like a power failure on
>     > the machine or something similar ? Any process being abrutly killed ?
>     > (not that this point should matters ?)
>
> No, nothing I can think off. (As I said the only thing I can think of is
> that I did quite a lot of directory renaming and, as I must say a lot of
> interactive commits (record) using the nurses interface. (Shouldn't
> influence).

The nurses interface should have help mending your repository. Joke
aside (no offense intended), I agree that I don't see a way for these to
corrupt your repository.


--
Pierre-Yves David
_______________________________________________
Mercurial mailing list
[hidden email]
https://www.mercurial-scm.org/mailman/listinfo/mercurial
Reply | Threaded
Open this post in threaded view
|

Re: corrupt local repository

Uwe Brauer
>>> "PD" == Pierre-Yves David <[hidden email]> writes:

   > On 6/10/20 9:35 PM, Uwe Brauer wrote:
   >>>>> "PD" == Pierre-Yves David <[hidden email]> writes:
   >> > That's very odd, it seems like a fresh corruption of a very
   >> new
   >> > repository I have head of that in a very long time.
   >> > I see you are using a rc version, can you move the latest
   >> official version ?
   >> Ok, please remind me which would that be?

   > Latest version is 5.4.1 Being at least on 5.1 would be better too.

Ok, I do this. But that would only help me to avoid future corruption.
That corrupted one is beyond help I presume.


   > Which version of python is used by you 5.2rc0 install ?

Well python 2.7 (I am on Ubuntu 16.04 and the python 3 version provided is
not supported by mercurial. This is one of the reasons I have to upgrade
Ubuntu, and most likely I will do in August, right now I cannot afford
any problems)
so I do

In contrib/packing/debian/rules

# export HGPYTHON3=1
export PYTHON=python


   >> > Is there anything else strange that happened like a power failure on
   >> > the machine or something similar ? Any process being abrutly killed ?
   >> > (not that this point should matters ?)
   >> No, nothing I can think off. (As I said the only thing I can think
   >> of is
   >> that I did quite a lot of directory renaming and, as I must say a lot of
   >> interactive commits (record) using the nurses interface. (Shouldn't
   >> influence).

   > The nurses interface should have help mending your repository. Joke
   > aside (no offense intended), I agree that I don't see a way for these
   > to corrupt your repository.

Honestly nurses is a huge help for commit -I
it is maybe not that useful for histedit (I am faster scrolling with emacs).

_______________________________________________
Mercurial mailing list
[hidden email]
https://www.mercurial-scm.org/mailman/listinfo/mercurial

smime.p7s (7K) Download Attachment
Reply | Threaded
Open this post in threaded view
|

Re: corrupt local repository

Uwe Brauer
In reply to this post by Pierre-Yves David-2
>>> "PD" == Pierre-Yves David <[hidden email]> writes:

   > On 6/10/20 9:35 PM, Uwe Brauer wrote:
   >>>>> "PD" == Pierre-Yves David <[hidden email]> writes:
   >> > That's very odd, it seems like a fresh corruption of a very
   >> new
   >> > repository I have head of that in a very long time.
   >> > I see you are using a rc version, can you move the latest
   >> official version ?
   >> Ok, please remind me which would that be?

   > Latest version is 5.4.1 Being at least on 5.1 would be better too.

I just upgraded to 5.4.1 and run into problems

 hg verify tells me *** failed to import extension artemis from
/home/oub/ALLES/src/Artemis/artemis: No module named 'artemis' ***
failed to import extension evolve: No module named 'evolve' *** failed
to import extension hggit from
/home/oub/ALLES/src/hg-git-heptapod/hggit: No module named 'compat' ***
failed to import extension mercurial_all_dirs from
~/ALLES/src/mercurial-all_dirs/mercurial_all_dirs.py: module
'mercurial.cmdutil' has no attribute 'command' *** failed to import
extension mercurial_all_paths: No module named 'mercurial_all_paths' ***
failed to import extension topic: No module named 'topic' Traceback
(most recent call last): File
"/usr/lib/python3/dist-packages/mercurial/extensions.py", line 251, in
_runuisetup uisetup(ui) File
"/home/oub/ALLES/src/hgchangelog/hgchangelog.py", line 95, in uisetup
extensions.wrapcommand(commands.table, 'commit', new_commit) File
"/usr/lib/python3/dist-packages/mercurial/extensions.py", line 557, in
wrapcommand aliases, entry = cmdutil.findcmd(command, table) File
"/usr/lib/python3/dist-packages/mercurial/cmdutil.py", line 961, in
findcmd raise error.UnknownCommand(cmd, allcmds)
mercurial.error.UnknownCommand: ('commit', [b'debugpathcomplete',
b'debugrevspec', b'shelve', b'recover', b'identify', b'id',
b'debugrebuildfncache', b'update', b'up', b'checkout', b'co', b'help',
b'debugsuccessorssets', b'debugdiscovery', b'tip', b'debugcomplete',
b'bisect', b'debugsub', b'pull', b'commit', b'ci', b'branch', b'bundle',
b'paths', b'phase', b'debugpushkey', b'tags', b'diff', b'debugpeer',
b'serve', b'debugobsolete', b'push', b'debugindex', b'add',
b'debugbundle', b'debugserve', b'archive', b'bookmarks', b'bookmark',
b'unbundle', b'incoming', b'in', b'debugnodemap', b'debugdirstate',
b'debugstate', b'debugdownload', b'branches', b'debuginstall',
b'manifest', b'debugmanifestfulltextcache', b'debuguiprompt',
b'debuguigetpass', b'log', b'history', b'outgoing', b'out', b'copy',
b'cp', b'debugp1copies', b'debugrename', b'debugupdatecaches', b'init',
b'debugcheckstate', b'debugdag', b'cat', b'debugbuilddag',
b'debuglocks', b'debugindexdot', b'continue', b'debugsidedata',
b'addremove', b'debugextensions', b'debugwireargs', b'debugignore',
b'debugindexstats', b'grep', b'debugmergestate', b'debugancestor',
b'graft', b'status', b'st', b'heads', b'root', b'rename', b'move',
b'mv', b'verify', b'annotate', b'blame', b'debugrevlog', b'debugcolor',
b'debugknown', b'debugapplystreamclonebundle', b'revert',
b'debugformat', b'remove', b'rm', b'debugdeltachain', b'export',
b'config', b'showconfig', b'debugconfig', b'debugpathcopies',
b'rollback', b'debugdate', b'unshelve', b'debugsetparents', b'summary',
b'sum', b'debugnamecomplete', b'locate', b'debugfsinfo',
b'debugp2copies', b'import', b'patch', b'debugrevlogindex',
b'debugtemplate', b'files', b'abort', b'debugssl', b'debugwhyunstable',
b'resolve', b'tag', b'forget', b'backout', b'debugpickmergetool',
b'debugcommands', b'debugpvec', b'debugupgraderepo',
b'debugcapabilities', b'clone', b'debugwalk', b'debugdata',
b'debugwireproto', b'debuglabelcomplete', b'debugfileset',
b'debugcreatestreamclonebundle', b'version', b'parents',
b'debugbackupbundle', b'debuggetbundle', b'merge',
b'debugrebuilddirstate', b'debugrebuildstate', b'debugtagscache']) ***
failed to set up extension hgchangelog: ('commit',
'[b'debugpathcomplete', b'debugrevspec', b'shelve', b'recover',
b'identify', b'id', b'debugrebuildfncache', b'update', b'up',
b'checkout', b'co', b'help', b'debugsuccessorssets', b'debugdiscovery',
b'tip', b'debugcomplete', b'bisect', b'debugsub', b'pull', b'commit',
b'ci', b'branch', b'bundle', b'paths', b'phase', b'debugpushkey',
b'tags', b'diff', b'debugpeer', b'serve', b'debugobsolete', b'push',
b'debugindex', b'add', b'debugbundle', b'debugserve', b'archive',
b'bookmarks', b'bookmark', b'unbundle', b'incoming', b'in',
b'debugnodemap', b'debugdirstate', b'debugstate', b'debugdownload',
b'branches', b'debuginstall', b'manifest',
b'debugmanifestfulltextcache', b'debuguiprompt', b'debuguigetpass',
b'log', b'history', b'outgoing', b'out', b'copy', b'cp',
b'debugp1copies', b'debugrename', b'debugupdatecaches', b'init',
b'debugcheckstate', b'debugdag', b'cat', b'debugbuilddag',
b'debuglocks', b'debugindexdot', b'continue', b'debugsidedata',
b'addremove', b'debugextensions', b'debugwireargs', b'debugignore',
b'debugindexstats', b'grep', b'debugmergestate', b'debugancestor',
b'graft', b'status', b'st', b'heads', b'root', b'rename', b'move',
b'mv', b'verify', b'annotate', b'blame', b'debugrevlog', b'debugcolor',
b'debugknown', b'debugapplystreamclonebundle', b'revert',
b'debugformat', b'remove', b'rm', b'debugdeltachain', b'export',
b'config', b'showconfig', b'debugconfig', b'debugpathcopies',
b'rollback', b'debugdate', b'unshelve', b'debugsetparents', b'summary',
b'sum', b'debugnamecomplete', b'locate', b'debugfsinfo',
b'debugp2copies', b'import', b'patch', b'debugrevlogindex',
b'debugtemplate', b'files', b'abort', b'debugssl', b'debugwhyunstable',
b'resolve', b'tag', b'forget', b'backout', b'debugpickmergetool',
b'debugcommands', b'debugpvec', b'debugupgraderepo',
b'debugcapabilities', b'clone', b'debugwalk', b'debugdata',
b'debugwireproto', b'debuglabelcomplete', b'debugfileset',
b'debugcreatestreamclonebundle', b'version', b'parents',
b'debugbackupbundle', b'debuggetbundle', b'merge',
b'debugrebuilddirstate', b'debugrebuildstate', b'debugtagscache']')
obsolete feature not enabled but 17 markers found!


I have to go back for the moment.

_______________________________________________
Mercurial mailing list
[hidden email]
https://www.mercurial-scm.org/mailman/listinfo/mercurial

smime.p7s (7K) Download Attachment