You are viewing a plain text version of this content. The canonical link for it is here.
Posted to dev@subversion.apache.org by Neels J Hofmeyr <ne...@elego.de> on 2012/07/06 00:58:53 UTC

why does tortoise, wcng and cygwin break from icon caching?

Today Random Person on #svn reported a problem and later its apparent
solution, and it struck me as rather peculiar:

[[[
<yates`> when trying to do an svn cleanup, i get a strange message:
svn: E200030: disk I/O error, executing statement 'COMMIT TRANSACTION;'

this is under win7 using the cygwin svn client
and also tortoisesvn is installed

...

<yates`> neels: i found the issue

strangely, there is an interaction between tortoisesvn and the cli svn under
cygwin

you must disable tortoisesvn's icon caching, then the problem goes away.
http://www.mail-archive.com/cygwin@cygwin.com/msg123745.html
]]]

and

"I've confirmed what others have stated that it's only a problem with
v3.7.12.1 and NOT a problem with v3.7.3."

Hmm, disable icon caching to not break wc-ng? Why!?

~Neels

Re: why does tortoise, wcng and cygwin break from icon caching?

Posted by Stefan Küng <to...@gmail.com>.
On 06.07.2012 14:41, Johan Corveleyn wrote:

> Ok, but I suppose that's mainly because of eol-style=native issues:
> your cygwin client may have a different idea about what the native
> eol-style is ... I don't think locking of the working copy (or working
> copy database) has been considered when this faq entry was written ...
>
> BTW: I have also recently recommended my users to make sure icon
> caching is disabled if they have TSVN installed. My users are mainly
> using IntelliJ IDEA with svnkit, and they also sometimes experience an
> E200030 error ("svn: E200030: BUSY" [1] in this case). Though I'm
> really shooting in the dark with this recommendation: I don't know if
> TSVNCache is involved at all (it's just a guess: trying to reduce the
> components on the system that may potentially be messing with the
> working copy at the same time) -- it may just as well be an internal
> svnkit or IntelliJ IDEA problem (some kind of multithreaded access
> inside IDEA that's not properly synchronized ... I don't know at this
> point).
>
> Anyway, just wanted to point that out. I have actually no idea what
> TSVNCache does with the working copy: does it query the database, what
> kind of locking effect it has, what if querying the wc.db takes a long
> time ... ? Can TSVNCache block out other clients from accessing or
> modifying the working copy?
>
> [1] http://youtrack.jetbrains.com/issue/IDEA-85496

Here's a quick overview of what the TSVNCache.exe process does:

It answers the shell extension of TSVN when it asks for the status of 
individual files to show the overlay icons.
So when the shell asks for a status, the cache process first checks if 
it already knows that status and if it does, that status is returned.
If it doesn't know the status yet, it returns 'unversioned' immediately 
so the shell isn't blocked. But the path for which the status isn't 
known yet is then passed to another thread that calls 
'svn_client_status5' for the parent directory of that path. While doing 
that, it also finds out where the working copy root is of that path: the 
wc root paths are then added to a monitor list. The cache process 
monitors all wc roots it finds for changes. If the system reports a 
change in a file below a wc root, those paths are then added to the same 
crawling thread that calls 'svn_client_status5' to fetch/refresh the status.

The status cache process uses only a few svn API's, mainly 
'svn_client_status5' and 'svn_client_proplist' and 'svn_client_propget'. 
Property reading is required since 'svn_client_status5' does not return 
information about the svn:needs-lock property.

In 1.6, all these API's didn't lock the working copy. But in 1.7 I think 
svn_client_status5 does for a short time.

Stefan

-- 
        ___
   oo  // \\      "De Chelonian Mobile"
  (_,\/ \_/ \     TortoiseSVN
    \ \_/_\_/>    The coolest Interface to (Sub)Version Control
    /_/   \_\     http://tortoisesvn.net

Re: why does tortoise, wcng and cygwin break from icon caching?

Posted by Johan Corveleyn <jc...@gmail.com>.
On Fri, Jul 6, 2012 at 2:20 PM, Simon Large <si...@gmail.com> wrote:
> On 6 July 2012 09:04, Stefan Fuhrmann <st...@wandisco.com> wrote:
>> On Fri, Jul 6, 2012 at 12:58 AM, Neels J Hofmeyr <ne...@elego.de> wrote:
>>>
>>> Today Random Person on #svn reported a problem and later its apparent
>>> solution, and it struck me as rather peculiar:
>>>
>>> [[[
>>> <yates`> when trying to do an svn cleanup, i get a strange message:
>>> svn: E200030: disk I/O error, executing statement 'COMMIT TRANSACTION;'
>>>
>> I suspect a reader / writer locking issue with Sqlite.
>> The I/O error as such might be related to the win7 /
>> cygwin combination requiring more retries or such.
>>
>>>
>>> this is under win7 using the cygwin svn client
>>>
>>> and also tortoisesvn is installed
>>>
>>> ...
>>>
>>> <yates`> neels: i found the issue
>>>
>>> strangely, there is an interaction between tortoisesvn and the cli svn
>>> under
>>> cygwin
>>>
>>> you must disable tortoisesvn's icon caching, then the problem goes away.
>>> http://www.mail-archive.com/cygwin@cygwin.com/msg123745.html
>>> ]]]
>>
>>
>> I.e. they disabled the background process that will
>> open the working copy soon after the OS detected
>> some file change. The scan itself may take a very
>> long time. Depending on OS and wc details, the db
>> may get opened at some very inconvenient time
>> while the commit is still in progress / in some critical
>> stage.
>>
>>>
>>> and
>>>
>>> "I've confirmed what others have stated that it's only a problem with
>>> v3.7.12.1 and NOT a problem with v3.7.3."
>>>
>>> Hmm, disable icon caching to not break wc-ng? Why!?
>>>
>> I think to remember that access to WCNG is much
>> more exclusive than it used to be in 1.6. But I don't
>> know whether that is actually true not the details.
>
> In our FAQ we advise against using TortoiseSVN with CygWin because
> Subversion was never designed to share a working copy between Windows
> and *nix clients. I thought this was in the SVN FAQ too, but can't
> find it right now.
>
> http://tortoisesvn.net/faq.html#multiclients

Ok, but I suppose that's mainly because of eol-style=native issues:
your cygwin client may have a different idea about what the native
eol-style is ... I don't think locking of the working copy (or working
copy database) has been considered when this faq entry was written ...

BTW: I have also recently recommended my users to make sure icon
caching is disabled if they have TSVN installed. My users are mainly
using IntelliJ IDEA with svnkit, and they also sometimes experience an
E200030 error ("svn: E200030: BUSY" [1] in this case). Though I'm
really shooting in the dark with this recommendation: I don't know if
TSVNCache is involved at all (it's just a guess: trying to reduce the
components on the system that may potentially be messing with the
working copy at the same time) -- it may just as well be an internal
svnkit or IntelliJ IDEA problem (some kind of multithreaded access
inside IDEA that's not properly synchronized ... I don't know at this
point).

Anyway, just wanted to point that out. I have actually no idea what
TSVNCache does with the working copy: does it query the database, what
kind of locking effect it has, what if querying the wc.db takes a long
time ... ? Can TSVNCache block out other clients from accessing or
modifying the working copy?

[1] http://youtrack.jetbrains.com/issue/IDEA-85496

-- 
Johan

Re: why does tortoise, wcng and cygwin break from icon caching?

Posted by Simon Large <si...@gmail.com>.
On 6 July 2012 09:04, Stefan Fuhrmann <st...@wandisco.com> wrote:
> On Fri, Jul 6, 2012 at 12:58 AM, Neels J Hofmeyr <ne...@elego.de> wrote:
>>
>> Today Random Person on #svn reported a problem and later its apparent
>> solution, and it struck me as rather peculiar:
>>
>> [[[
>> <yates`> when trying to do an svn cleanup, i get a strange message:
>> svn: E200030: disk I/O error, executing statement 'COMMIT TRANSACTION;'
>>
> I suspect a reader / writer locking issue with Sqlite.
> The I/O error as such might be related to the win7 /
> cygwin combination requiring more retries or such.
>
>>
>> this is under win7 using the cygwin svn client
>>
>> and also tortoisesvn is installed
>>
>> ...
>>
>> <yates`> neels: i found the issue
>>
>> strangely, there is an interaction between tortoisesvn and the cli svn
>> under
>> cygwin
>>
>> you must disable tortoisesvn's icon caching, then the problem goes away.
>> http://www.mail-archive.com/cygwin@cygwin.com/msg123745.html
>> ]]]
>
>
> I.e. they disabled the background process that will
> open the working copy soon after the OS detected
> some file change. The scan itself may take a very
> long time. Depending on OS and wc details, the db
> may get opened at some very inconvenient time
> while the commit is still in progress / in some critical
> stage.
>
>>
>> and
>>
>> "I've confirmed what others have stated that it's only a problem with
>> v3.7.12.1 and NOT a problem with v3.7.3."
>>
>> Hmm, disable icon caching to not break wc-ng? Why!?
>>
> I think to remember that access to WCNG is much
> more exclusive than it used to be in 1.6. But I don't
> know whether that is actually true not the details.

In our FAQ we advise against using TortoiseSVN with CygWin because
Subversion was never designed to share a working copy between Windows
and *nix clients. I thought this was in the SVN FAQ too, but can't
find it right now.

http://tortoisesvn.net/faq.html#multiclients

Simon

-- 
:       ___
:  oo  // \\      "De Chelonian Mobile"
: (_,\/ \_/ \     TortoiseSVN
:   \ \_/_\_/>    The coolest Interface to (Sub)Version Control
:   /_/   \_\     http://tortoisesvn.net

Re: why does tortoise, wcng and cygwin break from icon caching?

Posted by Stefan Fuhrmann <st...@wandisco.com>.
On Fri, Jul 6, 2012 at 12:58 AM, Neels J Hofmeyr <ne...@elego.de> wrote:

> Today Random Person on #svn reported a problem and later its apparent
> solution, and it struck me as rather peculiar:
>
> [[[
> <yates`> when trying to do an svn cleanup, i get a strange message:
> svn: E200030: disk I/O error, executing statement 'COMMIT TRANSACTION;'
>
> I suspect a reader / writer locking issue with Sqlite.
The I/O error as such might be related to the win7 /
cygwin combination requiring more retries or such.


> this is under win7 using the cygwin svn client
> and also tortoisesvn is installed
>
> ...
>
> <yates`> neels: i found the issue
>
> strangely, there is an interaction between tortoisesvn and the cli svn
> under
> cygwin
>
> you must disable tortoisesvn's icon caching, then the problem goes away.
> http://www.mail-archive.com/cygwin@cygwin.com/msg123745.html
> ]]]
>

I.e. they disabled the background process that will
open the working copy soon after the OS detected
some file change. The scan itself may take a very
long time. Depending on OS and wc details, the db
may get opened at some very inconvenient time
while the commit is still in progress / in some critical
stage.


> and
>
> "I've confirmed what others have stated that it's only a problem with
> v3.7.12.1 and NOT a problem with v3.7.3."
>
> Hmm, disable icon caching to not break wc-ng? Why!?
>
> I think to remember that access to WCNG is much
more exclusive than it used to be in 1.6. But I don't
know whether that is actually true not the details.

-- Stefan^2.

-- 
Certified & Supported Apache Subversion Downloads:
http://www.wandisco.com/subversion/download