bug report

[Fred Batty]

I have upgraded to hg 1.3.1 (in cygwin) and still having the same 
problem. It would appear this happens when the directory is mounted from 
a different machine (using Windows drive mount). I tried to clone the 
workspace fresh and got a different error:

destination directory: msg_next
requesting all changes
adding changesets
adding manifests
adding file changes
transaction abort!
rollback completed
** unknown exception encountered, details follow
** report bug details to http://mercurial.selenic.com/bts/
** or mercurial at selenic.com
** Mercurial Distributed SCM (version 1.3.1)
** Extensions loaded: graphlog
Traceback (most recent call last):
  File "/usr/bin/hg", line 27, in <module>
    mercurial.dispatch.run()
  File "dispatch.py", line 16, in run
  File "dispatch.py", line 27, in dispatch
  File "dispatch.py", line 43, in _runcatch
  File "dispatch.py", line 449, in _dispatch
  File "dispatch.py", line 317, in runcommand
  File "dispatch.py", line 501, in _runcommand
  File "dispatch.py", line 454, in checkargs
  File "dispatch.py", line 448, in <lambda>
  File "util.py", line 402, in check
  File "/usr/lib/python2.5/commands.py", line 636, in clone

  File "hg.py", line 286, in clone
  File "localrepo.py", line 2177, in clone
  File "localrepo.py", line 1466, in pull
  File "localrepo.py", line 2067, in addchangegroup
  File "revlog.py", line 1246, in addgroup
  File "revlog.py", line 998, in revision
  File "revlog.py", line 958, in _chunk
  File "revlog.py", line 108, in decompress
zlib.error: Error -5 while decompressing data

Adrian Buehlmann wrote:
> On 30.11.2009 20:18, Fred Batty wrote:
>   
>> I seem to be getting this type of error frequently:
>>
>> % hg pull -u
>> pulling from ssh://mercurial.sfbay.sun.com//m/comms/msg_next
>> searching for changes
>> adding changesets
>> adding manifests
>> adding file changes
>> added 4 changesets with 20 changes to 20 files
>> merging msg/lib/cyrus/getwrap.c
>> ** unknown exception encountered, details follow
>> ** report bug details to http://www.selenic.com/mercurial/bts
>> ** or mercurial at selenic.com
>> ** Mercurial Distributed SCM (version 1.1)
>>     
>
> Not sure if this related at all with your problem here, but
> I have to mention that 1.1 was quickly bugfixed for 1.1.1
> because of http://mercurial.selenic.com/bts/issue1417
> (which is mentioned on the page [1] in the wiki).
>
> So I have to recommend against 1.1 for that reason alone.
>
> 1.1 was the first version to introduce the fncache
> repository layout (it isn't really a new format, only
> filename encoding changes inside '.hg/store/' were done)
>
> [1] http://mercurial.selenic.com/wiki/fncacheRepoFormat
>