Registered Member
|
I recently upgraded from KDE 4.12.4 to 4.13.0.
I followed the instructions to migrate from Nepomuk to Baloo given in the thread Transitioning to KDE's next generation semantic search but it does not seem to have worked correctly, as Nepomuk is still running. nepomukbaloomigrator crashes if I run it. Before I ran nepomukbaloomigrator I noticed that the file nepomukserverrc already contained "migrated=true". Was nepomukserverrc run automatically the first time I logged-in to KDE 4.13.0? Anyway, here is the current situation:
I would be grateful for advice on what to do in order to complete the migration successfully. Thanks in advance. |
Administrator
|
Did you try removing the [Baloo] key and associated bits entirely?
"Violence is the last refuge of the incompetent."
Plasma FAQ maintainer - Plasma programming with Python |
KDE Developer
|
Please do as Ben suggested. Remove the [Baloo] migrated = true from the config file, and then run the nepomukbaloomigrator. You may want to run it under gdb as follows -
$ gdb nepomukbaloomigrator > run This way if it crashes, you can get a backtrace by typing 'bt'. You should file a bug if it crashes again. |
Registered Member
|
Unfortunately that did not work:
I also tried with gdb as suggested by vHanda, but got the following:
|
Registered Member
|
Further to my previous post, I decided to reboot even though nepomukbaloomigrator continued to crash, just to see if anything would change after a complete restart. There is now a difference:
I also see that nepomukserverrc was again altered to include [Baloo] migrated=true, but this time both [Basic Settings] Start Nepomuk and [Service-nepomukfileindexer] autostart have been changed to =false as you can see below:
Does this mean that, despite nepomukbaloomigrator crashing as shown in my previous post, this time the migration completed successfully? How can I check that Nepomuk has been properly disabled and that Baloo is running and indexing correctly? |
KDE Developer
|
Previously, when you couldn't get a backtrace off gdb. That was cause the migrator was still running in the background (I think). If you want to get a proper backtrace, you need to kill it and then start it via gdb.
Anyway, it seems like the migration is done. A backtrace would have been nice though. Check if baloo_file is running. If it is, then you are good to go. |
Registered Member
|
Does the following look correct?
So just akonadi_baloo_indexer and baloo_file take care of indexing everything? How can I be sure that everything in the Nepomuk database has been moved to Baloo's database? Is there some way for me to check that? Is there a way of making Baloo re-index all files if necessary? I'd also be grateful if you could point me to some good on-line documentation on Baloo, so that I can learn how to use it properly, including any commands I can issue from the command line. Thanks again for your help. |
Administrator
|
Yes, it looks fine from a functionality perspective. While I am not well versed in the new system yet, I can tell you that "baloosearch" is a command-line search client, and "balooshow" shows metadata information about a file. The rest behaves like with Nepomuk, IOW you can use Dolphin to search, or KRunner (the "Run Command" interface), or the new "Milou" plasmoid.
"Violence is the last refuge of the incompetent."
Plasma FAQ maintainer - Plasma programming with Python |
Registered users: Bing [Bot], Evergrowing, Google [Bot], rblackwell