Hello, On 07.10.2005 19:14, David Catanach wrote:
Thank you for the advice on filesets from the other day. We are storing file images to two different filesystems. We want all fulls to go onto a USB disk and all differential and incrementals should go onto a partition on the filesystem. I have a 'pool' for each type of backup set up and a 'storage' for each location on the filesystem in both the dir.conf and sd.conf, and I can successfully write to either if I start a job manually and select the storage I want. The scheduler detects which type of backup we're doing (full, incr, diff) and sets the correct pool (even for upgrades to full), but the storage is not picked correctly; it comes up with whatever is listed explicitly in the job resource. How can I get the storage to be based on the pool used, or alternatively, on the type of backup used? If I set up seperate jobs for each, I can direct where the data goes explicitly, but then Incr->Full or Diff->Full upgrades don't work right (wrong storage again), which shouldn't happen often, but it would be nice for this to be detected. Is there an example of this somewhere?
That's not currently possible. You can set the device in the schedule, though,and only have to manually manage stuff when you do backups out-of-order, for example with he initial full backups for new jobs or filesets.
Also, is the 1.37 series ok for production use? It looks like many of you are using it, but I am not.
1.37 is strictly a development version, but many people already use it - it is running quite stable, but you should be aware that the main purpose of running this version is bug hunting. I think Kern likes to get good bug reports to improve bacula, but for you that will probably mean spending more time on getting everything working and observing opertion and filing he occasional report than when you use the stable version 1.36.3.
I'm not sure if what you originally wanted can be addressed by the 1.37 way of manging events. The main idea is to have python code which can influence many aspects of baculas inernal operation, and I hope that there will be an event that allows you to upgrade jobs more completely. Something like a method called when a job is actually started, which is given all job parameters, and which then can change the pool, storage, start time and god knows what else. I can't tell you if it's there already, or if it ever will exist.
Arno
Thanks ------------------------------------------------------- This SF.Net email is sponsored by: Power Architecture Resource Center: Free content, downloads, discussions, and more. http://solutions.newsforge.com/ibmarch.tmpl _______________________________________________ Bacula-users mailing list Bacula-users@lists.sourceforge.net https://lists.sourceforge.net/lists/listinfo/bacula-users
-- IT-Service Lehmann [EMAIL PROTECTED] Arno Lehmann http://www.its-lehmann.de ------------------------------------------------------- This SF.Net email is sponsored by: Power Architecture Resource Center: Free content, downloads, discussions, and more. http://solutions.newsforge.com/ibmarch.tmpl _______________________________________________ Bacula-users mailing list Bacula-users@lists.sourceforge.net https://lists.sourceforge.net/lists/listinfo/bacula-users
