[OpenAFS-devel] namei vs. inode on AIX

[Hartmut Reuter]

Matt Bacchi schrieb:
> Ok, I'm confused.  I cannot compile with --disable-namei-fileserver for 
> some reason.  At one point the inode fileserver was the default on AIX, 
> but that doesn't appear to be the case any longer.  I have a partition 
> on my test server that previously housed volumes with the inode 
> fileserver.  When I run configure with the --disable-namei-fileserver 
> option, then check src/config/afsconfig.h, AFS_NAMEI_ENV is still 
> defined as before.
> 
> So my questions are:
> 
> 1.  Can I actually disable AFS_NAMEI_ENV with configure?

the src/config/param.rs_aix51.h is the version I used. I hard coded the 
#define AFS_NAMEI_ENV here because I thought nobody should start on a 
new server with the inode interface, anymore.

If you want just remove these lines.

> 
> 2.  If I uncomment line 8 in src/config/afsconfig.h should I get an 
> inode fileserver?

see above

> 
> 3.  How can I determine if the binaries in /usr/afs/bin are inode or 
> namei capable?  (symbols?)

do a "strings fileserver |grep namei" this shows lots of routine names 
if it is build with AFS_NAMEI_ENV.

> 
> 4.  How can I determine if the volumes on the partition are ok, and I'm 
> not just getting an error because they are corrupt?  (I don't really 
> care about the volumes because it's only a test server, I just want to 
> determine if the error is valid)

If you have volumes which were created with the inode-interface you 
cannot use the partition with a namei-fileserver. You should start your 
old fileserver and move them else where, then best remake your 
filesystem, start the namei-fileserver and move them back.

> 
> 5.  What is actually creating the log /usr/afs/logs/FileLog? 
> Fileserver?  That is where the error messages are output, example:

Yes, the following message I introduced into the fileserver to prevent 
you from destroying your exsiting inode-volumes!

-Hartmut
> 
> Tue Jul  8 13:51:22 2003 This program is compiled with AFS_NAMEI_ENV, 
> but partition /vicepa seems to contain
> volumes which don't use the namei-interface; aborting
> 
> Thanks in advance,
> -Matt
> 
> _______________________________________________
> OpenAFS-devel mailing list
> OpenAFS-devel@openafs.org
> https://lists.openafs.org/mailman/listinfo/openafs-devel


-- 
-----------------------------------------------------------------
Hartmut Reuter                           e-mail reuter@rzg.mpg.de
					   phone +49-89-3299-1328
RZG (Rechenzentrum Garching)               fax   +49-89-3299-1301
Computing Center of the Max-Planck-Gesellschaft (MPG) and the
Institut fuer Plasmaphysik (IPP)
-----------------------------------------------------------------