ACL bug in 2.1rc1?
Hi All! I was just playing with the ACL setting in the new rc1 and I found the following behaviour: I took the tomcat-slide bundle as offsett and enabled the JAAS realm. Also I mapped slide to the root of the server in the server.xml. Using the standard users from the default domain.xml and then I change access rights on the /files collection: /roles/user grant /actions/write-acl /roles/user grant /actions/write-content /roles/user grant /actions/write-propeties /roles/user grant /actions/read /roles/user deny all Then I login with john and try to upload a file to the /files collection and i get an access denied. The put request returns a 403 write access /actions/write denied. Changing the acl info to: /roles/user grant /actions/write /roles/user grant /actions/read /roles/user deny all Now it works fine? Am I misunderstanding something /actions/write and the combination of /actions/write-acl, /actions/write-content and /actions/write-propeties bee the same? Thanks Jacob
Re: AW: version_history table's branch_id column
Sorry, but I've no clue what this branches are for. I've also noticed some
strange things going on in the Slide core regarding versioning (some temporary
hidden 0.0 revisions...)
Maybe it's an option for you to listen on acl events and create the appropriate
folders by calling the MKCOL method triggered by the event?
Cheers,
Daniel
Slide Users Mailing List [EMAIL PROTECTED] schrieb am 07.12.04
01:54:52:
Thank you very much, Daniel. But could you give me a little more explanation
about the
main branch and the backup branch?
I'm using webdavclient in most places. But there are one or two places where
I need to
create folder inside of the other webdav method, for example, I need to
create a folder in
ACLMethod so that when a folder is assigned to some user, accordingly, a
system folder is created at the same time (to server our requirement). This
is why we use the slide api directly in this place.
I need to know a little more of the slide's logic in handling the versioning
and branch, then
I can make a decision of what to do in my application.
Thanks.
regards,
Jun
Daniel Florey [EMAIL PROTECTED] wrote:
Do not use the Slide API directly, but use the webdavclient-library instead.
You'll run into trouble if you want to use versioning. This is handled in
the WebDAV-layer...
Cheers,
Daniel
-Ursprüngliche Nachricht-
Von: [EMAIL PROTECTED]
[mailto:[EMAIL PROTECTED]
Im Auftrag von Gao Jun
Gesendet: Montag, 6. Dezember 2004 08:44
An: [EMAIL PROTECTED]
Betreff: version_history table's branch_id column
Hi,
I've some problem in using slide. When I use propfind on a folder I
created, I was told that folder doesn't exist. The backend log reads
02 Dec 2004 02:27:02 5793672 DEBUG [tcpConnection-80-154] httpclient.wire
- HTTP/1.1 500 Internal Server Error: No revision descriptor found on
object MY_FOLDER_NAME[\r][\n]
I created that folder with my own code listed below:
//create the new folder
SubjectNode collection = new SubjectNode();
structure.create(slideToken, collection,
folderName);//slide's original implement
NodeRevisionDescriptor revisionDescriptor = new
NodeRevisionDescriptor(0);
// Resource type
revisionDescriptor.setResourceType();
// Creation date
revisionDescriptor.setCreationDate(new Date());
// Last modification date
revisionDescriptor.setLastModified(new Date());
// Content length name
revisionDescriptor.setContentLength(0);
// Source
revisionDescriptor.setSource();
// Owner
String creator =
((SubjectNode)security.getPrincipal(slideToken)).getPath().lastSegment();
revisionDescriptor.setCreationUser(creator);
revisionDescriptor.setOwner(creator);
// Added for DeltaV --start--
if( Configuration.useVersionControl() )
{
VersioningHelper vHelp =
VersioningHelper.getVersioningHelper(
slideToken, token, req, resp, getConfig() );
// Workspace
vHelp.setWorkspaceProperty( userLinkPath,
revisionDescriptor );
}
// Added for DeltaV --end--
if (isMsProprietarySupport()) {
NodeProperty property = null;
// Is hidden
property = new NodeProperty(ishidden, 0,
MICROSOFT);
revisionDescriptor.setProperty(property);
// Is collection
property = new NodeProperty(iscollection, 1,
MICROSOFT);
revisionDescriptor.setProperty(property);
// Is read only
property = new NodeProperty(isreadonly, 0,
MICROSOFT);
revisionDescriptor.setProperty(property);
// Last accessed
property = new NodeProperty(lastaccessed,
(new
Date()).toString(), MICROSOFT);
revisionDescriptor.setProperty(property);
}
content.create(slideToken, folderName,
revisionDescriptor, null);
...
I found the new folder name has been created in URI table. The odd thing I
found is
in the table version_history, the row for this new folder's URI has a
branch_id which
is a backup instead of main according to table branch.
I have no idea of what this backup and main mean. Could anybody help
explain
and could you see what's the cause of my problem? Thanks a lot.
regards,
Jun
-
Do you Yahoo!?
Yahoo! Mail - You care about security. So do we.
-
To unsubscribe, e-mail: [EMAIL PROTECTED]
For additional commands, e-mail: [EMAIL PROTECTED]
__
Do You Yahoo!?
Tired of spam? Yahoo! Mail has the best spam protection around
http://mail.yahoo.com
__
Mit WEB.DE FreePhone mit hoechster Qualitaet ab 0 Ct./Min.
weltweit telefonieren! http://freephone.web.de/?mc=021201
-
To unsubscribe, e-mail: [EMAIL PROTECTED]
For additional commands, e-mail: [EMAIL PROTECTED]
history-collection-hack bug
Hi all,
When slide is configured to split the history collection into seperate
stores parameter name=historypath/history/${store}/parameter, then
the history-collection-hack breaks the versioning part of slide.
I have attached an example domain.xml file. When history-collection-hack is
set to false it works fine.
I tested this with 2.1rc1.
/jacob
?xml version=1.0?
slide
namespace name=slide
definition
store name=tx
parameter name=tlock-timeout120/parameter
nodestore classname=org.apache.slide.store.txfile.TxXMLFileDescriptorsStore
parameter name=rootpathstore/metadata/parameter
parameter name=workpathwork/metadata/parameter
parameter name=defer-savingtrue/parameter
parameter name=timeout120/parameter
/nodestore
sequencestore classname=org.apache.slide.store.txfile.FileSequenceStore
parameter name=rootpathstore/sequence/parameter
/sequencestore
securitystore
reference store=nodestore/
/securitystore
lockstore
reference store=nodestore/
/lockstore
revisiondescriptorsstore
reference store=nodestore/
/revisiondescriptorsstore
revisiondescriptorstore
reference store=nodestore/
/revisiondescriptorstore
contentstore classname=org.apache.slide.store.txfile.TxFileContentStore
parameter name=rootpathstore/content/parameter
parameter name=workpathwork/content/parameter
parameter name=defer-savingtrue/parameter
parameter name=timeout120/parameter
/contentstore
/store
scope match=/ store=tx/
store name=tx2
parameter name=tlock-timeout120/parameter
nodestore classname=org.apache.slide.store.txfile.TxXMLFileDescriptorsStore
parameter name=rootpathstoreh/metadata/parameter
parameter name=workpathworkh/metadata/parameter
parameter name=defer-savingtrue/parameter
parameter name=timeout120/parameter
/nodestore
sequencestore classname=org.apache.slide.store.txfile.FileSequenceStore
parameter name=rootpathstoreh/sequence/parameter
/sequencestore
securitystore
reference store=nodestore/
/securitystore
lockstore
reference store=nodestore/
/lockstore
revisiondescriptorsstore
reference store=nodestore/
/revisiondescriptorsstore
revisiondescriptorstore
reference store=nodestore/
/revisiondescriptorstore
contentstore classname=org.apache.slide.store.txfile.TxFileContentStore
parameter name=rootpathstoreh/content/parameter
parameter name=workpathworkh/content/parameter
parameter name=defer-savingtrue/parameter
parameter name=timeout120/parameter
/contentstore
/store
scope match=/history/tx store=tx2/
store name=files
parameter name=tlock-timeout120/parameter
nodestore classname=org.apache.slide.store.txfile.TxXMLFileDescriptorsStore
parameter name=rootpathstoref/metadata/parameter
parameter name=workpathworfk/metadata/parameter
parameter name=defer-savingtrue/parameter
parameter name=timeout120/parameter
/nodestore
sequencestore classname=org.apache.slide.store.txfile.FileSequenceStore
parameter name=rootpathstoref/sequence/parameter
/sequencestore
securitystore
reference store=nodestore/
/securitystore
lockstore
reference store=nodestore/
/lockstore
revisiondescriptorsstore
reference store=nodestore/
/revisiondescriptorsstore
revisiondescriptorstore
reference store=nodestore/
/revisiondescriptorstore
contentstore classname=org.apache.slide.store.txfile.TxFileContentStore
parameter name=rootpathstoref/content/parameter
parameter name=workpathworkf/content/parameter
parameter name=defer-savingtrue/parameter
parameter name=timeout120/parameter
/contentstore
/store
scope match=/files store=files/
store name=files2
parameter name=tlock-timeout120/parameter
nodestore classname=org.apache.slide.store.txfile.TxXMLFileDescriptorsStore
parameter name=rootpathstorefh/metadata/parameter
parameter name=workpathworkfh/metadata/parameter
parameter name=defer-savingtrue/parameter
parameter name=timeout120/parameter
/nodestore
sequencestore classname=org.apache.slide.store.txfile.FileSequenceStore
parameter name=rootpathstorefh/sequence/parameter
/sequencestore
securitystore
reference store=nodestore/
/securitystore
lockstore
reference store=nodestore/
/lockstore
revisiondescriptorsstore
reference store=nodestore/
/revisiondescriptorsstore
revisiondescriptorstore
reference store=nodestore/
/revisiondescriptorstore
contentstore classname=org.apache.slide.store.txfile.TxFileContentStore
parameter name=rootpathstorefh/content/parameter
parameter name=workpathworkfh/content/parameter
parameter name=defer-savingtrue/parameter
parameter name=timeout120/parameter
/contentstore
/store
scope match=/history/files
WebDav LOCK and MOVE question
Hello, I want to upload a file to a web server. Several machines maybe try in parallel to upload the same file. Therefore locking is needed to allow only one upload. Moreover I have to ensure that no other machine downloads the file before the complete upload is done. So the algorithm I have in mind goes like this: 0) Check whether the file already exists on the file server. If yes = done 1) Lock non existing temporary file. 2) Upload the bytes to the locked temporary file. 3) Move the temporary file to the destination name. 4) Unlock the lock. Unfortunately it seems that I'm to dump and/or have not enough knowledge about WebDay and the slide WebDav client library to implement the algorithm. Who is able to help me? Thx! Bernd -- Bernd Wellhöfer VerdiSoft GmbH Phone: +49 40 280054-24 Süderstraße 77 Fax: +49 40 280054-50 20097 Hamburg Email: [EMAIL PROTECTED]Germany - To unsubscribe, e-mail: [EMAIL PROTECTED] For additional commands, e-mail: [EMAIL PROTECTED]
auto-version-values
hi slide-users ok, because my post about versioning ends up in a huge discussion about other things, i make a new one :-) i'm now a step ahead (thanks to warwick burrows). i set the following in domain.xml: parameter name=auto-versioncheckout/parameter parameter name=auto-version-controltrue/parameter when i put a new file in the repository i do a put, and right after that, i'm doing about 20 proppatches. finally, i do a manual checkin. result: 2 (equal) versions in history and a current one in files. i guess that i always have the current version in history too, have i? question: why have i got two equal versions in history? are they created on (automatical) checkout and (manual) checkin? and second question: warwick burrows wrote ... disable auto-versioning altogether and do both checkout and checkin yourself... how can i do this? set auto-version-control to false or auto-version to an empty string? thanks and greetings stefan - To unsubscribe, e-mail: [EMAIL PROTECTED] For additional commands, e-mail: [EMAIL PROTECTED]
Re: history-collection-hack bug
Any idea what the problem with the hack is?
Oliver
On Tue, 7 Dec 2004 12:33:16 +0100, Jacob Lund [EMAIL PROTECTED] wrote:
Hi all,
When slide is configured to split the history collection into seperate
stores parameter name=historypath/history/${store}/parameter, then
the history-collection-hack breaks the versioning part of slide.
I have attached an example domain.xml file. When history-collection-hack is
set to false it works fine.
I tested this with 2.1rc1.
/jacob
-
To unsubscribe, e-mail: [EMAIL PROTECTED]
For additional commands, e-mail: [EMAIL PROTECTED]
-
To unsubscribe, e-mail: [EMAIL PROTECTED]
For additional commands, e-mail: [EMAIL PROTECTED]
Re: auto-version-values
Stefan, If you leave the auto-version parameter blank, empty, no value then NOTHING will happen automatically. I will mention that you should be careful with understanding how these parameters work with dumb webdav clients such as Windows WebFolders. The dumb behaviour is to automatically overwrite whatever is live as there is no checkout/checking issued. I would recommend using the checkout option for that reason, as a safeguard. There is a duplicate of the file created in history to allow the uncheckout command to function. Uncheckout is like undo. You can checkout a file, do 20 changes on it, then decide No, I really don't want to save those changes for this version, so you issue the uncheckout command to rollback to the last good copy from the history. Hope that helps. Tim Stefan Burkard wrote on 07/12/04 09:23 AM: hi slide-users ok, because my post about versioning ends up in a huge discussion about other things, i make a new one :-) i'm now a step ahead (thanks to warwick burrows). i set the following in domain.xml: parameter name=auto-versioncheckout/parameter parameter name=auto-version-controltrue/parameter when i put a new file in the repository i do a put, and right after that, i'm doing about 20 proppatches. finally, i do a manual checkin. result: 2 (equal) versions in history and a current one in files. i guess that i always have the current version in history too, have i? question: why have i got two equal versions in history? are they created on (automatical) checkout and (manual) checkin? and second question: warwick burrows wrote ... disable auto-versioning altogether and do both checkout and checkin yourself... how can i do this? set auto-version-control to false or auto-version to an empty string? thanks and greetings stefan - To unsubscribe, e-mail: [EMAIL PROTECTED] For additional commands, e-mail: [EMAIL PROTECTED] - To unsubscribe, e-mail: [EMAIL PROTECTED] For additional commands, e-mail: [EMAIL PROTECTED]
Re: Exceptions on MySQL 4.1
As a final hint: set isolation level to read committed! Oliver P.S.: Thanks for reporting all this; I am currently working on the deadlock retry thing... On Mon, 06 Dec 2004 17:02:56 -0800, Mirko Froehlich [EMAIL PROTECTED] wrote: In that case, I don't think this will solve my issue. Slide actually started up fine with MySQL 4.1 for the first time, using the MySql41RDBMSAdapter and the new schema. I ran my concurrency test, which again failed after less than a minute. Only subsequent startups failed due to the schema issues, for some reason. -Mirko On Mon, 2004-12-06 at 16:47, Oliver Zeigermann wrote: Nono, use the MySql41RDBMSAdapter. That's what worked for me. Oliver On Mon, 06 Dec 2004 16:37:36 -0800, Mirko Froehlich [EMAIL PROTECTED] wrote: In that case, should I continue to use the MySqlRDBMSAdapter, or should I use the MySql41RDBMSAdapter? -Mirko On Mon, 2004-12-06 at 16:25, Oliver Zeigermann wrote: May be true, I did my tests with the 4.0 schema and it worked just fine. No idea what the original problem is. Anyway, use the 4.0 schema even for 4.1 and it should work. Oliver - To unsubscribe, e-mail: [EMAIL PROTECTED] For additional commands, e-mail: [EMAIL PROTECTED] - To unsubscribe, e-mail: [EMAIL PROTECTED] For additional commands, e-mail: [EMAIL PROTECTED]
Re: history-collection-hack bug
I am looking into it now - i just found out!
It does not throw an exception! I will report back as soon as I have
something.
/jacob
- Original Message -
From: Oliver Zeigermann [EMAIL PROTECTED]
To: Slide Users Mailing List [EMAIL PROTECTED]
Sent: Tuesday, December 07, 2004 3:35 PM
Subject: Re: history-collection-hack bug
Any idea what the problem with the hack is?
Oliver
On Tue, 7 Dec 2004 12:33:16 +0100, Jacob Lund [EMAIL PROTECTED] wrote:
Hi all,
When slide is configured to split the history collection into seperate
stores parameter name=historypath/history/${store}/parameter, then
the history-collection-hack breaks the versioning part of slide.
I have attached an example domain.xml file. When history-collection-hack
is
set to false it works fine.
I tested this with 2.1rc1.
/jacob
-
To unsubscribe, e-mail: [EMAIL PROTECTED]
For additional commands, e-mail: [EMAIL PROTECTED]
-
To unsubscribe, e-mail: [EMAIL PROTECTED]
For additional commands, e-mail: [EMAIL PROTECTED]
-
To unsubscribe, e-mail: [EMAIL PROTECTED]
For additional commands, e-mail: [EMAIL PROTECTED]
Re: WebDav LOCK and MOVE question
What is the WebDAV server you are programming to? Slide? If so, just upload the file. Oliver On Tue, 7 Dec 2004 15:11:42 +0100, Bernd Kühl [EMAIL PROTECTED] wrote: Hello, I want to upload a file to a web server. Several machines maybe try in parallel to upload the same file. Therefore locking is needed to allow only one upload. Moreover I have to ensure that no other machine downloads the file before the complete upload is done. So the algorithm I have in mind goes like this: 0) Check whether the file already exists on the file server. If yes = done 1) Lock non existing temporary file. 2) Upload the bytes to the locked temporary file. 3) Move the temporary file to the destination name. 4) Unlock the lock. Unfortunately it seems that I'm to dump and/or have not enough knowledge about WebDay and the slide WebDav client library to implement the algorithm. Who is able to help me? Thx! Bernd -- Bernd Wellhöfer VerdiSoft GmbH Phone: +49 40 280054-24 Süderstraße 77 Fax: +49 40 280054-50 20097 Hamburg Email: [EMAIL PROTECTED]Germany - To unsubscribe, e-mail: [EMAIL PROTECTED] For additional commands, e-mail: [EMAIL PROTECTED] - To unsubscribe, e-mail: [EMAIL PROTECTED] For additional commands, e-mail: [EMAIL PROTECTED]
Default max file size?
What is the default max file size allowed? And more importantly, where is this setting? Can you also point me to some docs that address limiting the amount of disk space each user may use? thanks, Mike - To unsubscribe, e-mail: [EMAIL PROTECTED] For additional commands, e-mail: [EMAIL PROTECTED]
RE: WebDav LOCK and MOVE question
Hello, It is Apache/2.0.44 (Unix) DAV/2 mod_webapp/1.2.0-dev. Just uploading the file is not good enough since then different server machines can do the upload in parallel. Moreover a client machine can fetch the file while a [slow] server machine is currently doing the upload. This leads into a corrupted file. Thx, Bernd -Original Message- From: Oliver Zeigermann [mailto:[EMAIL PROTECTED] Sent: Tuesday, December 07, 2004 3:47 PM To: Slide Users Mailing List Subject: Re: WebDav LOCK and MOVE question What is the WebDAV server you are programming to? Slide? If so, just upload the file. Oliver On Tue, 7 Dec 2004 15:11:42 +0100, Bernd Kühl [EMAIL PROTECTED] wrote: Hello, I want to upload a file to a web server. Several machines maybe try in parallel to upload the same file. Therefore locking is needed to allow only one upload. Moreover I have to ensure that no other machine downloads the file before the complete upload is done. So the algorithm I have in mind goes like this: 0) Check whether the file already exists on the file server. If yes = done 1) Lock non existing temporary file. 2) Upload the bytes to the locked temporary file. 3) Move the temporary file to the destination name. 4) Unlock the lock. Unfortunately it seems that I'm to dump and/or have not enough knowledge about WebDay and the slide WebDav client library to implement the algorithm. Who is able to help me? Thx! Bernd -- Bernd Wellhöfer VerdiSoft GmbH Phone: +49 40 280054-24 Süderstraße 77 Fax: +49 40 280054-50 20097 Hamburg Email: [EMAIL PROTECTED]Germany - To unsubscribe, e-mail: [EMAIL PROTECTED] For additional commands, e-mail: [EMAIL PROTECTED] - To unsubscribe, e-mail: [EMAIL PROTECTED] For additional commands, e-mail: [EMAIL PROTECTED] - To unsubscribe, e-mail: [EMAIL PROTECTED] For additional commands, e-mail: [EMAIL PROTECTED]
Re: WebDav LOCK and MOVE question
So, you still could has a problem while moving the temporary file. Bottom line: You try to reimplement transactional features modav does not seem to have. Maybe then it is not the right choice in the first place. Oliver On Tue, 7 Dec 2004 16:06:49 +0100, Bernd Kühl [EMAIL PROTECTED] wrote: Hello, It is Apache/2.0.44 (Unix) DAV/2 mod_webapp/1.2.0-dev. Just uploading the file is not good enough since then different server machines can do the upload in parallel. Moreover a client machine can fetch the file while a [slow] server machine is currently doing the upload. This leads into a corrupted file. Thx, Bernd -Original Message- From: Oliver Zeigermann [mailto:[EMAIL PROTECTED] Sent: Tuesday, December 07, 2004 3:47 PM To: Slide Users Mailing List Subject: Re: WebDav LOCK and MOVE question What is the WebDAV server you are programming to? Slide? If so, just upload the file. Oliver On Tue, 7 Dec 2004 15:11:42 +0100, Bernd Kühl [EMAIL PROTECTED] wrote: Hello, I want to upload a file to a web server. Several machines maybe try in parallel to upload the same file. Therefore locking is needed to allow only one upload. Moreover I have to ensure that no other machine downloads the file before the complete upload is done. So the algorithm I have in mind goes like this: 0) Check whether the file already exists on the file server. If yes = done 1) Lock non existing temporary file. 2) Upload the bytes to the locked temporary file. 3) Move the temporary file to the destination name. 4) Unlock the lock. Unfortunately it seems that I'm to dump and/or have not enough knowledge about WebDay and the slide WebDav client library to implement the algorithm. Who is able to help me? Thx! Bernd -- Bernd Wellhöfer VerdiSoft GmbH Phone: +49 40 280054-24 Süderstraße 77 Fax: +49 40 280054-50 20097 Hamburg Email: [EMAIL PROTECTED]Germany - To unsubscribe, e-mail: [EMAIL PROTECTED] For additional commands, e-mail: [EMAIL PROTECTED] - To unsubscribe, e-mail: [EMAIL PROTECTED] For additional commands, e-mail: [EMAIL PROTECTED] - To unsubscribe, e-mail: [EMAIL PROTECTED] For additional commands, e-mail: [EMAIL PROTECTED] - To unsubscribe, e-mail: [EMAIL PROTECTED] For additional commands, e-mail: [EMAIL PROTECTED]
RE: WebDav LOCK and MOVE question
Hello, Yes I have a problem with the move. Please find attached a simple test program. The output of the program is Empty put temp file status is 201, result is true Put temp file status is 201, result is true lock temp file status is 200, result is true Put temp file status is 423, result is false Move status is 201, result is true unlock status is 204, result is true As you see I get a 423 error status code for the move. Apache logged: [Tue Dec 07 16:30:42 2004] [error] [client 172.16.8.1] This resource is locked and an If: header was not supplied to allow access to the resource. [423, #0] I do not understand why no (or the wrong) If: is supplied to the move request. Who can help me? Thx, Bernd -Original Message- From: Oliver Zeigermann [mailto:[EMAIL PROTECTED] Sent: Tuesday, December 07, 2004 4:18 PM To: Slide Users Mailing List Subject: Re: WebDav LOCK and MOVE question So, you still could has a problem while moving the temporary file. Bottom line: You try to reimplement transactional features modav does not seem to have. Maybe then it is not the right choice in the first place. Oliver On Tue, 7 Dec 2004 16:06:49 +0100, Bernd Kühl [EMAIL PROTECTED] wrote: Hello, It is Apache/2.0.44 (Unix) DAV/2 mod_webapp/1.2.0-dev. Just uploading the file is not good enough since then different server machines can do the upload in parallel. Moreover a client machine can fetch the file while a [slow] server machine is currently doing the upload. This leads into a corrupted file. Thx, Bernd -Original Message- From: Oliver Zeigermann [mailto:[EMAIL PROTECTED] Sent: Tuesday, December 07, 2004 3:47 PM To: Slide Users Mailing List Subject: Re: WebDav LOCK and MOVE question What is the WebDAV server you are programming to? Slide? If so, just upload the file. Oliver On Tue, 7 Dec 2004 15:11:42 +0100, Bernd Kühl [EMAIL PROTECTED] wrote: Hello, I want to upload a file to a web server. Several machines maybe try in parallel to upload the same file. Therefore locking is needed to allow only one upload. Moreover I have to ensure that no other machine downloads the file before the complete upload is done. So the algorithm I have in mind goes like this: 0) Check whether the file already exists on the file server. If yes = done 1) Lock non existing temporary file. 2) Upload the bytes to the locked temporary file. 3) Move the temporary file to the destination name. 4) Unlock the lock. Unfortunately it seems that I'm to dump and/or have not enough knowledge about WebDay and the slide WebDav client library to implement the algorithm. Who is able to help me? Thx! Bernd -- Bernd Wellhöfer VerdiSoft GmbH Phone: +49 40 280054-24 Süderstraße 77 Fax: +49 40 280054-50 20097 Hamburg Email: [EMAIL PROTECTED]Germany - To unsubscribe, e-mail: [EMAIL PROTECTED] For additional commands, e-mail: [EMAIL PROTECTED] - To unsubscribe, e-mail: [EMAIL PROTECTED] For additional commands, e-mail: [EMAIL PROTECTED] - To unsubscribe, e-mail: [EMAIL PROTECTED] For additional commands, e-mail: [EMAIL PROTECTED] - To unsubscribe, e-mail: [EMAIL PROTECTED] For additional commands, e-mail: [EMAIL PROTECTED] - To unsubscribe, e-mail: [EMAIL PROTECTED] For additional commands, e-mail: [EMAIL PROTECTED]
RE: auto-version-values
Hello Stefan, Adding to Tim's response, the propatch commands does not change the file content, but just it's metadata, witch is also being versioned but in the associated xml file of versioning, inside the metainfo store. It's pretty easy to confirm this if your metadata store is a txfilestore. Hope this helps, Miguel Figueiredo -Original Message- From: Tim Frank [mailto:[EMAIL PROTECTED] Sent: terça-feira, 7 de Dezembro de 2004 14:38 To: Slide Users Mailing List Subject: Re: auto-version-values Stefan, If you leave the auto-version parameter blank, empty, no value then NOTHING will happen automatically. I will mention that you should be careful with understanding how these parameters work with dumb webdav clients such as Windows WebFolders. The dumb behaviour is to automatically overwrite whatever is live as there is no checkout/checking issued. I would recommend using the checkout option for that reason, as a safeguard. There is a duplicate of the file created in history to allow the uncheckout command to function. Uncheckout is like undo. You can checkout a file, do 20 changes on it, then decide No, I really don't want to save those changes for this version, so you issue the uncheckout command to rollback to the last good copy from the history. Hope that helps. Tim Stefan Burkard wrote on 07/12/04 09:23 AM: hi slide-users ok, because my post about versioning ends up in a huge discussion about other things, i make a new one :-) i'm now a step ahead (thanks to warwick burrows). i set the following in domain.xml: parameter name=auto-versioncheckout/parameter parameter name=auto-version-controltrue/parameter when i put a new file in the repository i do a put, and right after that, i'm doing about 20 proppatches. finally, i do a manual checkin. result: 2 (equal) versions in history and a current one in files. i guess that i always have the current version in history too, have i? question: why have i got two equal versions in history? are they created on (automatical) checkout and (manual) checkin? and second question: warwick burrows wrote ... disable auto-versioning altogether and do both checkout and checkin yourself... how can i do this? set auto-version-control to false or auto-version to an empty string? thanks and greetings stefan - To unsubscribe, e-mail: [EMAIL PROTECTED] For additional commands, e-mail: [EMAIL PROTECTED] - To unsubscribe, e-mail: [EMAIL PROTECTED] For additional commands, e-mail: [EMAIL PROTECTED] - To unsubscribe, e-mail: [EMAIL PROTECTED] For additional commands, e-mail: [EMAIL PROTECTED]
Re: auto-version-values
Tim Frank wrote: If you leave the auto-version parameter blank, empty, no value then NOTHING will happen automatically. I will mention that you should be careful with understanding how these parameters work with dumb webdav clients such as Windows WebFolders. The dumb behaviour is to automatically overwrite whatever is live as there is no checkout/checking issued. I would recommend using the checkout option for that reason, as a safeguard. --- ok, thanks for the advice. additionally i saw that the double-version only occurs when i first put in the document. when i put in a new version of an existing doc, it just creates ONE new version. just two other questions: - what are the consequences for dumb-client-documents that are checked out (automatically), but not checked in (manually) again? - what is the versioncontrol-method for? to activate and disactivate version-controlling on a document? There is a duplicate of the file created in history to allow the uncheckout command to function. Uncheckout is like undo. You can checkout a file, do 20 changes on it, then decide No, I really don't want to save those changes for this version, so you issue the uncheckout command to rollback to the last good copy from the history. --- understand, thanks a lot! stefan - To unsubscribe, e-mail: [EMAIL PROTECTED] For additional commands, e-mail: [EMAIL PROTECTED]
Re: auto-version-values
Stefan, Stefan Burkard wrote on 07/12/04 11:00 AM: --- ok, thanks for the advice. additionally i saw that the double-version only occurs when i first put in the document. when i put in a new version of an existing doc, it just creates ONE new version. Hmm, maybe this wasn't what I thought it was, but I guess that might depend on which client you used to create the document. I think I'd only ever seen 1 version created myself, but I honestly wasn't paying that much attention. just two other questions: - what are the consequences for dumb-client-documents that are checked out (automatically), but not checked in (manually) again? The consequence would be that a dumb client would never be able to actually create a new version of your document. You would have to perform the checking from a smart client. This is normally not a problem if you have some sort of CMS/workflow interface to control the checkin features. You could leverage the ease of use of web folders or the KDE webdav:// protocol for day-to-day updates rather than a clunky web interface. I have had locking issues with dumb clients though, so if something gets locked by a smart client after an auto-checkout it will automatically get checked-in when the lock is removed again from the smart client. This is per the spec, but not always desired program behaviour. Long and the short of it, use a smart client for all checkout/checkin/uncheckout/locking. - what is the versioncontrol-method for? to activate and disactivate version-controlling on a document? If you don't use the auto-version-control parameter, yes, this is the method that would be called to place a resource under version control. We had an earlier discussion about how to take a resource out of version control and I think the consensus was you couldn't. I think you would have to create a new resource that was specifically not under version and fill the contents/metadata from the last version. I have spent quite a bit of time investigating how all of these properties work with or against each other. The gist of everything is that dumb clients just mess everything up or make things more complicated. The only upside is easier use by the end user because you can have drag-drop, remote editing, remote saving, etc. I think you are wading your way through :) Tim - To unsubscribe, e-mail: [EMAIL PROTECTED] For additional commands, e-mail: [EMAIL PROTECTED]
New feature for
Folks, I have added a new option to internally and transparently repeat a transaction if it has failed becasue of concurrency confclits. Right now only tx file and mysql support this experimental feature. There are two new parameters in the configuration section of Domain.xml repeat-upon-conflict which enables this feature and max-retry-repeat which sets the maximum number of repeats, defaults are false and 1. E.g. this parameter name=repeat-upon-conflicttrue/parameter parameter name=max-retry-repeat5/parameter enables the repeast feature and sets the maximum number of repeats to 5. If it is disabled everything should work just as before. Oliver - To unsubscribe, e-mail: [EMAIL PROTECTED] For additional commands, e-mail: [EMAIL PROTECTED]
Re: New feature for internally repeating conflicting requests
Sorry, messed up the subject... On Tue, 7 Dec 2004 18:11:49 +0100, Oliver Zeigermann [EMAIL PROTECTED] wrote: Folks, I have added a new option to internally and transparently repeat a transaction if it has failed becasue of concurrency confclits. Right now only tx file and mysql support this experimental feature. There are two new parameters in the configuration section of Domain.xml repeat-upon-conflict which enables this feature and max-retry-repeat which sets the maximum number of repeats, defaults are false and 1. E.g. this parameter name=repeat-upon-conflicttrue/parameter parameter name=max-retry-repeat5/parameter enables the repeast feature and sets the maximum number of repeats to 5. If it is disabled everything should work just as before. Oliver - To unsubscribe, e-mail: [EMAIL PROTECTED] For additional commands, e-mail: [EMAIL PROTECTED]
Re: Exceptions on MySQL 4.1
OK, this should work for MySQL now. Mirko, do you still have enough patience to try it with the latest CVS head? Oliver On Tue, 7 Dec 2004 15:40:15 +0100, Oliver Zeigermann [EMAIL PROTECTED] wrote: As a final hint: set isolation level to read committed! Oliver P.S.: Thanks for reporting all this; I am currently working on the deadlock retry thing... On Mon, 06 Dec 2004 17:02:56 -0800, Mirko Froehlich [EMAIL PROTECTED] wrote: In that case, I don't think this will solve my issue. Slide actually started up fine with MySQL 4.1 for the first time, using the MySql41RDBMSAdapter and the new schema. I ran my concurrency test, which again failed after less than a minute. Only subsequent startups failed due to the schema issues, for some reason. -Mirko On Mon, 2004-12-06 at 16:47, Oliver Zeigermann wrote: Nono, use the MySql41RDBMSAdapter. That's what worked for me. Oliver On Mon, 06 Dec 2004 16:37:36 -0800, Mirko Froehlich [EMAIL PROTECTED] wrote: In that case, should I continue to use the MySqlRDBMSAdapter, or should I use the MySql41RDBMSAdapter? -Mirko On Mon, 2004-12-06 at 16:25, Oliver Zeigermann wrote: May be true, I did my tests with the 4.0 schema and it worked just fine. No idea what the original problem is. Anyway, use the 4.0 schema even for 4.1 and it should work. Oliver - To unsubscribe, e-mail: [EMAIL PROTECTED] For additional commands, e-mail: [EMAIL PROTECTED] - To unsubscribe, e-mail: [EMAIL PROTECTED] For additional commands, e-mail: [EMAIL PROTECTED]
Re: Exceptions on MySQL 4.1
Absolutely! I'll let you know how it goes. Again, thanks for implementing these fixes. -Mirko On Tue, 2004-12-07 at 09:18, Oliver Zeigermann wrote: OK, this should work for MySQL now. Mirko, do you still have enough patience to try it with the latest CVS head? Oliver On Tue, 7 Dec 2004 15:40:15 +0100, Oliver Zeigermann [EMAIL PROTECTED] wrote: As a final hint: set isolation level to read committed! Oliver P.S.: Thanks for reporting all this; I am currently working on the deadlock retry thing... On Mon, 06 Dec 2004 17:02:56 -0800, Mirko Froehlich [EMAIL PROTECTED] wrote: In that case, I don't think this will solve my issue. Slide actually started up fine with MySQL 4.1 for the first time, using the MySql41RDBMSAdapter and the new schema. I ran my concurrency test, which again failed after less than a minute. Only subsequent startups failed due to the schema issues, for some reason. -Mirko On Mon, 2004-12-06 at 16:47, Oliver Zeigermann wrote: Nono, use the MySql41RDBMSAdapter. That's what worked for me. Oliver On Mon, 06 Dec 2004 16:37:36 -0800, Mirko Froehlich [EMAIL PROTECTED] wrote: In that case, should I continue to use the MySqlRDBMSAdapter, or should I use the MySql41RDBMSAdapter? -Mirko On Mon, 2004-12-06 at 16:25, Oliver Zeigermann wrote: May be true, I did my tests with the 4.0 schema and it worked just fine. No idea what the original problem is. Anyway, use the 4.0 schema even for 4.1 and it should work. Oliver - To unsubscribe, e-mail: [EMAIL PROTECTED] For additional commands, e-mail: [EMAIL PROTECTED] - To unsubscribe, e-mail: [EMAIL PROTECTED] For additional commands, e-mail: [EMAIL PROTECTED]
RE: WebDav LOCK and MOVE question
Bernd, Slide's WebDAV client creates a lock-null resource when you lock a file that doesn't exist and that you want to create. This is how the WebDAV standard deals with two uploads of the same new file at the same time. When the second server tries to lock the resource they will be denied with the 423 error code. So you don't need a temp file (I don't think) as the WebDAV server creates it for you. So to do what you want and assuming that all of your clients that are doing uploads simultaneously will be your DAV clients then you need to: 0) Check whether the file already exists on the file server. If yes = done 1) Call webdav client lockMethod on target file (creates lock-null resource) 2) Call webdav client putMethod to upload bytes to the lock-null file. 4) Unlock the lock with webdav client unlockMethod on target file. Try this approach and you should not see the 423 error you're seeing now -- at least if you're using the Slide 2.1 WebDAV client which I am using and which passes the lock to the server in the request as it is supposed to. Warwick -Original Message- From: Bernd Kühl [mailto:[EMAIL PROTECTED] Sent: Tuesday, December 07, 2004 9:34 AM To: Slide Users Mailing List; [EMAIL PROTECTED] Subject: RE: WebDav LOCK and MOVE question Hello, Yes I have a problem with the move. Please find attached a simple test program. The output of the program is Empty put temp file status is 201, result is true Put temp file status is 201, result is true lock temp file status is 200, result is true Put temp file status is 423, result is false Move status is 201, result is true unlock status is 204, result is true As you see I get a 423 error status code for the move. Apache logged: [Tue Dec 07 16:30:42 2004] [error] [client 172.16.8.1] This resource is locked and an If: header was not supplied to allow access to the resource. [423, #0] I do not understand why no (or the wrong) If: is supplied to the move request. Who can help me? Thx, Bernd -Original Message- From: Oliver Zeigermann [mailto:[EMAIL PROTECTED] Sent: Tuesday, December 07, 2004 4:18 PM To: Slide Users Mailing List Subject: Re: WebDav LOCK and MOVE question So, you still could has a problem while moving the temporary file. Bottom line: You try to reimplement transactional features modav does not seem to have. Maybe then it is not the right choice in the first place. Oliver On Tue, 7 Dec 2004 16:06:49 +0100, Bernd Kühl [EMAIL PROTECTED] wrote: Hello, It is Apache/2.0.44 (Unix) DAV/2 mod_webapp/1.2.0-dev. Just uploading the file is not good enough since then different server machines can do the upload in parallel. Moreover a client machine can fetch the file while a [slow] server machine is currently doing the upload. This leads into a corrupted file. Thx, Bernd -Original Message- From: Oliver Zeigermann [mailto:[EMAIL PROTECTED] Sent: Tuesday, December 07, 2004 3:47 PM To: Slide Users Mailing List Subject: Re: WebDav LOCK and MOVE question What is the WebDAV server you are programming to? Slide? If so, just upload the file. Oliver On Tue, 7 Dec 2004 15:11:42 +0100, Bernd Kühl [EMAIL PROTECTED] wrote: Hello, I want to upload a file to a web server. Several machines maybe try in parallel to upload the same file. Therefore locking is needed to allow only one upload. Moreover I have to ensure that no other machine downloads the file before the complete upload is done. So the algorithm I have in mind goes like this: 0) Check whether the file already exists on the file server. If yes = done 1) Lock non existing temporary file. 2) Upload the bytes to the locked temporary file. 3) Move the temporary file to the destination name. 4) Unlock the lock. Unfortunately it seems that I'm to dump and/or have not enough knowledge about WebDay and the slide WebDav client library to implement the algorithm. Who is able to help me? Thx! Bernd -- Bernd Wellhöfer VerdiSoft GmbH Phone: +49 40 280054-24 Süderstraße 77 Fax: +49 40 280054-50 20097 Hamburg Email: [EMAIL PROTECTED]Germany - To unsubscribe, e-mail: [EMAIL PROTECTED] For additional commands, e-mail: [EMAIL PROTECTED] -- --- To unsubscribe, e-mail: [EMAIL PROTECTED] For additional commands, e-mail: [EMAIL PROTECTED] - To unsubscribe,
Idea on final release of Slide 2.1
Anyone have an idea of when the final release of Slide 2.1 will be available? Regards, Chris - To unsubscribe, e-mail: [EMAIL PROTECTED] For additional commands, e-mail: [EMAIL PROTECTED]
propfind using webdavclient and mod_dav
I ran into a weird issue using the WebDAV client API as well as the command line tool to access an Apache mod_dav based WebDAV repository. When I create a WebdavResource for a particular path and enumerate the child resources using listWebdavResources() or getChildResources(), it only returns the folder itself, not its children. The command line tool displays the same behavior. When I explicitly use a PropfindMethod with a depth of 1, it works fine. Any ideas? -Mirko
Slide 2.1rc1 thrid party libraries info
Hi, Where could I find detailed version info about all the third party libraries which slide 2.1rc1 is depending on? Like jdom, xerces, etc. We need to compare those libraries' version info with those we already have in our application to see whether we should replace them. Thanks a lot. regards, Jun - Do you Yahoo!? Yahoo! Mail - 250MB free storage. Do more. Manage less.
Re: Cluster cache
On Tue, 2004-12-07 at 18:10 -0800, Warwick Burrows wrote: Hi James, I was hoping you could answer some questions about the cluster cache design. We would like to go into production with Slide but doing so requires the cluster cache implementation -- or no caching at all which is not even an option :-) We need to know the cluster cache mechanics so we can cater for them. - Are the updates (notifications) between servers performed synchronously or asynchronously? They should be asynchronous. Daniel Florey wrote the notification mechanism, so he can confirm, but I'm pretty sure on this. When a resource is modified the server sends notifications to all subscribers that there are events waiting for them. The subscribers (other servers) then POLL the server and get a list of changed resources. Each subscriber then removes the resource in the list from their cache. - If an update to a server fails will it be retried? It depends on where the failure is. If a subscriber is unable to POLL a server, it will retry. In fact, each subscriber periodically POLLs the server just to be sure it hasn't missed any notifications. The POLL frequency should be configurable. Any other design points I should be aware of? Not that I can think of. Incidentally I had mentioned that locks could be an issue with the cache design in that two lock requests coming in at the same time may both return with the lock. It isn't an issue for us though as locking is controlled by our DAV applications and the lock information they share is stored in a DB. Only one app is guaranteed to get the lock and so create a Slide lock. One way to mitigate this problem for other clients may be to allow for the lockstore store to be specifically configured so that only lock related metadata isn't cached. ExtendedStore has five (I think) separate caches. You can set the size of each cache by setting parameters on the store. One of the caches is for locks. Take a look at the source for ExtendedStore, since I'm not sure where, if anywhere, the parameters are documented. -James Thanks, Warwick - To unsubscribe, e-mail: [EMAIL PROTECTED] For additional commands, e-mail: [EMAIL PROTECTED]
