Support

Just upgraded to pro version. First problem was an upload/install error message
" -1 - An error has occurred.

It was not possible to copy the selected file."

This message resolved itself when I uninstalled the core product before installing the Pro version.

However, now the backup fails in what appears to be an infinite loop while attempting to complete the Initialization process.The green bar proceeds to about 5%, then resets and starts again, repeatedly. It also appears that at each recursion a new backup file is created. The first seems properly named, but the others have " (no description)" in the description field. The longer I let it sit while trying to initialize, the more of these files are created in the manage backups screen.

This situation does not appear to self terminate. It forces me to navigate away from the screen in order to get out of it (i waited about an hour once, creating a few hundred of the "no description" files). When I navigate away, a red highlighted error screen flashes briefly, but does not remain long enough for me to read everything on it. In part it says:

Ajax loading error:error.

Inspection of the log file did not, to me, provide any clues of errors, the last entries in the log file were:

[100826 20:04:41] ----- Finished operation 2 ------
[100826 20:04:41] ----- Finished operation 3 ------
[100826 20:04:41] Successful Smart algorithm on AECoreDomainInit
[100826 20:04:41] Kettenrad :: Domain 'init' has finished.
[100826 20:04:41] ====== Finished Step number 1 ======
[100826 20:04:41] *** Batching of engine steps finished. I will now return control to the caller.
[100826 20:04:41] Sleeping for 1984.90214348 msec, using usleep()
[100826 20:04:43] Saving Kettenrad instance backend
[100826 20:04:43]

The log file stops abruptly at this point.

Help!

Nicholas, Thank you for your attention to this problem so far. Uninstalling, and then installing the developers release, does not seem to affect the situation at all. All previously reported symptoms remain, and all attempts to backup from the back end have failed.

I do, however, have more information to add. While using the core version, I had set up cron jobs as per the core product cron instructions on the website. These cron jobs did not work for me, which is why I upgraded to pro. Prior to my uninstalling the pro version and installing the developers version, one of these cron jobs hit. This time it successfully created a backup from the front end!

So something is clearly working better now with the pro version, but the backup on the backend problem remains unresolved

Hello again Nicholas,
Upgraded to 3.1.1b3, no change in behavior. Have not tried the portable firefox, but can confirm that the problem does not exist in Google Chrome or Internet Explorer as I have them configured.

I have gone through all of my plugins and disabled them in groups of four, hoping to identify an aberrant plugin. The problem persisted independently of disabled plugins.

Similarly for disabled extensions.

I could send you a list of my plugins and extensions if this is useful. I would also be happy to test any easy to test configurations on my end, if that would be helpful to your efforts.

Although the problem is not yet resolved, I am currently in a "functional" state. My cron jobs are working via the front end, and I know to use IE or Chrome if I want to do a backend backup. Of course, using Firefox is preferred, so let me know if further tests on my part would be of help in sorting out the problem.

Nicholas, I am using Firefox 3.6.8

But I have good news, it works just fine now. I stepped away from the problem for a while, replaced my McAffee antispam software with norton antivirus, and rebooted both machines that I am using. The problem is now gone on both machines. Cannot say for sure what made the difference, the reboot or the McAfee replacement, or just my own previous operator error somehow. But I would strongly suspect the Mcafee package, as it has been giving me troubles in other areas as well. If I had suspected that that might be the problem, I would have only upgraded one of my two machines to do a comparison with that being the only variable. Too late now.

But I am happy to report that the problem is now fully resolved. Thank you very much for your help on this. Great program!

Robert

Drat, I've been posting in the wrong thread. I just posted three posts to 3.1.b.2 before I realized it wasn't 3.1.b.3

I think these posts may be relevant, Nicholas. Hopefully they will help supply a clue:

http://www.akeebabackup.com/forum/25731/topic.html?p=3#p35099
http://www.akeebabackup.com/forum/25731/topic.html?p=3#p35105
http://www.akeebabackup.com/forum/25731/topic.html?p=3#p35107

You only really need the first link as they are three consecutive posts.

Regards,
Ken

Nicholas,

Looks like I spoke too soon on the problem being completely gone. I was working on setting up cron jobs yesterday, and one of them went astray. It wound up creating about 10,000 bad backup records before I flagged the problem. I had to go directly into the database with Navicat in order to delete them. I will not make any attempt to debug this one, though, until the new release is out for me to install. I see from the other thread that one is coming today.

Nicholas,

Yes, front end cron jobs.

I was having the same problem and spent well over an hour quite frustrated with it. This is the last thread I had to look through in the forum and BAM! Virus Software. I myself just updated from AVG Free 9 to 2011, and the LinkScanner feature was apparently blocking important data in my ajax response. Turing LinkScanner off fixed it immediately.

Thanks for the insight guys.