for some reasons I didn't notice the replies until just now :( 2007/2/7, Martin Simmons <[EMAIL PROTECTED]>: > >>>>> On Wed, 07 Feb 2007 13:14:19 +0100, Arno Lehmann said: > > > > Hello, > > > > On 2/4/2007 10:34 PM, Florian Heigl wrote: > > > Hi, > > > > > > I'm currently working on clustering bacula (howto follows as all > > > quirks are removed) > > > > That sounds interesting... > > > > > My requirement is running a bacula fd on each of the cluster nodes, > > > and one for the cluster package. If someone thinks, this makes no > > > sense, I'd be also glad to hear why. > > > > > > For the meantime I'm mostly concerned with making bacula listen to a > > > specific ip. > > > # netstat -na | grep LISTE | grep 910 > > > [...] > > > tcp 0 0 0.0.0.0:9101 0.0.0.0:* > > > LISTEN > > > tcp 0 0 0.0.0.0:9102 0.0.0.0:* > > > LISTEN > > > tcp 0 0 0.0.0.0:9103 0.0.0.0:* > > > LISTEN > > > > > > I haven't found out where I could influence that, how do I achieve this? > > > > There's a feature request regarding that... you can influence it with > > the editor of your choice and the build tools :-)
You don't want code written by me. Trust me on that. > I think listening on a particular IP address is already implemented by the > DirAddress directive (and similarly for the other daemons). The feature > request is for outgoing connections to originate on a particular IP. Thanks I will try that out, all I need / want. For clarification: I use a parallel filesystem as cluster storage, but this will be the only part that does parallel work. I'm not trying to do the clustered-fd approach from the feature requests ( saving a big filesystem in parallel), but wanna build a big bacula server. The bacula-dir can't run in a full active/active configuration (Kern wrote about this long ago, I don't remember much details) but I don't see much of an issue about that - yet. i.e. image having 3 nodes: 1. bacula-dir + one sd handling two devices 2. database + one sd handling two devices 3. two sd's handling two devices [...] this setup should mean that You can start with a 1-node cluster and scale to up to 16 using a light solution (ocfs2+heartbeat). if someone runs a clustered (not failover) mysql, I'd like to hear from You too. so long, and thanks to all that replied! Florian ------------------------------------------------------------------------- Take Surveys. Earn Cash. Influence the Future of IT Join SourceForge.net's Techsay panel and you'll get the chance to share your opinions on IT & business topics through brief surveys-and earn cash http://www.techsay.com/default.php?page=join.php&p=sourceforge&CID=DEVDEV _______________________________________________ Bacula-users mailing list Bacula-users@lists.sourceforge.net https://lists.sourceforge.net/lists/listinfo/bacula-users
