Regis_Odeye: > > Will you try exporting and remote-mounting for each branches, to see > > they are all fine and exportable? > > =20 > I try to export all the branches (instead of only one), but I still have=20 > the warning.
Do you mean the warning is produced repeatedly? Is so, that is strange because after the warning aufs sets a value to a global variable per filesystem type, and before the warning aufs checks this variable whether it is set or not. There may be something unknown changes made by RHEL 5. Also the upper layer that aufs (ie. nfsd or fs/expfs.c) may rejected. I may have to check RHEL 5 source files. sigh... > I try also to set the xino out of the exported aufs mount point. Will you try 'noplink' aufs mount option too? And let me make sure. You did set fsid in your export list, didn't you? > Do you think exporting branches is not enough to get rid of the warning=20 > but a remote-mount is really needed ? > If yes, it will not be so easy to implement on my embedded diskless syste= > m. To export and remount-mount is just a test to see your branches are exported corrently. It is unnecessary in usual operation. Junjiro Okajima ------------------------------------------------------------------------- This SF.net email is sponsored by the 2008 JavaOne(SM) Conference Don't miss this year's exciting event. There's still time to save $100. Use priority code J8TL2D2. http://ad.doubleclick.net/clk;198757673;13503038;p?http://java.sun.com/javaone
