|
|
|
|
| Index | Recent Threads | Unanswered Threads | Who's Online | Help |
DataNucleus Forum
DataNucleus News
Announcements
Thread: JPOX Forum Postings
|
| No member browsing this thread |
|
Thread Status: Active Total posts in this thread: 7
|
|
| Author |
|
andy
Expert UK Joined: Mar 13, 2004 Post Count: 5686 Status: Offline |
There has been a corruption of data in the database used by the Forum software that JPOX uses (MVNForum). Unfortunately this database was not backed up and the most recent dump was taken in August 2005. the content of postings from that data is not currently in a recoverable form. The thread listing is correct, but when you see a thread it will show "this topic has been viewed XXX Times and has -1 replies". One solution to see the content of a thread would be to get the URL of the thread and put it into Google and use the Google cache. That is your only workaround. In the future this data will be backed up but this doesnt help older postings. Apologies for the inconvenience, but this project is not heavily resourced and not heavily donated to meaning that we cannot do everything we would like. ---------------------------------------- -Andy   |
||
|
|
jfrantzius
Advanced Member Germany Joined: Jun 18, 2004 Post Count: 385 Status: Offline |
That's really a pity  At least the SF CVS seems to be up again. |
||
|
|
chrisco
Expert Australia Joined: Nov 24, 2005 Post Count: 706 Status: Offline |
It would be great if someone wrote a webcrawler app that extracted all the posts from Google via the google cache (just using an incremented post number at each iteration) and resubmits them to the forum! It looks like the [Link] link each each post contains the thread and post numbers in the form thread=threadNum#postNumso maybe this could help. ---------------------------------------- Chris Colman http://visualclassworks.com Javelin: Visual modeling & coding with automatic JDO 2/DataNucleus meta data generation. http://expojo.com expojo: Simple dependency injection and exposed POJO domain model pattern |
||
|
|
curacao
Newbie Joined: Apr 14, 2006 Post Count: 1 Status: Offline |
I am assuming you are runing on Linux. You can use this script to daily make backup archives on your system and after that FTP those archives to another server on your network. http://www.faqs.org/docs/securing/chap29sec306.html Let's say your create a backup directory /backups and a scripts directory /backups/scripts. After you make your changes in the declaration part of the script, you should run it by placing this in the /etc/crontab file: 00 22 * * * root /backups/scripts/backup.cron 00 23 * * * root /backups/scripts/ftp.cron It will backup at 22:00 h. and FTP the archives at 23:00 h. Make sure backup.cron and ftp.cron are executable (chmod 755 <file>) A typical FTP script (ftp.cron) to do the job would have as content: #!/bin/sh HOST='yourftpserverhostname' USER='ftpusername' PASSWD='ftppassword' /usr/bin/ftp -n $HOST <<END_SCRIPT quote USER $USER quote PASS $PASSWD bin cd "/" lcd /backups prompt off mput *.tar quit END_SCRIPT exit 0 You can also .netrc to FTP the file automatically. Remember to run this before you run the first backup: date +%d%b > /backups/last-full/myserverhostname-full-date I do this in production systems and it works well. You can enhance the script to encrypt and password protect your archive if you want. With this, to restore, you only need to FTP get your archive back to the server and run: cd / tar xpf <tar archive name> [selected specification of files to restore separated by a comma] Bye, curacao |
||
|
|
andy
Expert UK Joined: Mar 13, 2004 Post Count: 5686 Status: Offline |
It would be great if someone wrote a webcrawler app that extracted all the posts from Google via the google cache Our friends at Nightlabs are already attempting this. ---------------------------------------- -Andy |
||
|
|
fleque
Member Germany Joined: Nov 28, 2004 Post Count: 89 Status: Offline |
Our friends at Nightlabs are already attempting this. Hi all, we managed to restore approximately half of the lost post (1675/3352) out of google's cache. Some posts might look a bit ugly as I had to parse the html the forum produces and rewrite it to its syntax. All initial post in a thread could be recovered, too, as mvn stores this data in the mvnforumThread table additionally. There should be no thread with 0 posts in the forum any more. There are still issues with the search for the forum. I think it is due to backup used (from August 2005). Some posts now link to forums that don't exist any more. I could find at least two 27 Requested Features 26 Confirmed Bugs Where should these post be mapped to? Additionally we've set up a backup for the database which runs every sunday. ---------------------------------------- Regards Alex  jfire: free erp, crm, scm and more |
||
|
|
andy
Expert UK Joined: Mar 13, 2004 Post Count: 5686 Status: Offline |
we managed to restore approximately half of the lost post (1675/3352) out of google's cache. Thanks Alex!!! There are still issues with the search for the forum. I think it is due to backup used (from August 2005). Some posts now link to forums that don't exist any more. I could find at least two 27 Requested Features 26 Confirmed Bugs Where should these post be mapped to? I only installed the one table from that backup so it still referred to these deprecated forums. I mapped the 2 deprecated forums into forum 24 ("JDO Objects"). I also just rebuilt the Lucene indexes for searching. Search should be ok now we've set up a backup for the database which runs every sunday. Excellent! ---------------------------------------- -Andy |
||
|
|
|
|
Current timezone is GMT May 17, 2012 11:10:07 AM |
