Thus spake Amar P. Patel ([EMAIL PROTECTED]):
> I have a multiprocessor machine, however the module was created
> automatically during compilation using the kernel-sources. Do I need
> to rename the module or create another one specifically for
> multiprocessing machine? If so how do I create ano
Thus spake Amar P. Patel ([EMAIL PROTECTED]):
> The module that I see in /usr/vice/etc is:
> libafs-2.6.15-1.2054_FC5.ko. Should I create a symbolic link to it
> named libafs-2.6.15-1.2054_FC5smp.mp.ko?
Uhm ... do you have a single- or multi-processor machine/kernel?
--
Fashion is a form of ug
Thus spake Konstantin Boyanov ([EMAIL PROTECTED]):
> [EMAIL PROTECTED] fs sysname
> bash: fs: command not found
>
> I can't understand what's wrong. If someone has ideas on what may be
> wrong I'll appreciate it.
The fs binary is not in your $PATH.
--
Fashion is a form of ugliness so intolerabl
Hi,
Thus spake KELEMEN Peter ([EMAIL PROTECTED]):
> > I've seen too many software raids (x86, any linux 2.4; never
> > tried 2.6 anymore) corrupt the files under high load, so I trust
> > them less than I trust a single USB harddisk. [...]
>
> That's very interesting, what kind of load do you exp
Hi,
Thus spake Jeffrey Altman ([EMAIL PROTECTED]):
> What about software RAID? At the very least you can implement that
> for the same $200.
I've seen too many software raids (x86, any linux 2.4; never tried 2.6
anymore) corrupt the files under high load, so I trust them less than I
trust a sing
Hi,
Thus spake Padiyath Sreekumaran ([EMAIL PROTECTED]):
> I get the following message on the client amchine:
>
> /afscache: write failed, file system is full
you have to specify the size of the cache in /usr/vice/etc/cacheinfo.
The client believes you and will use this amount of diskspace, so it
Thus spake ted creedon ([EMAIL PROTECTED]):
> http://www.pmw.org/afsbpw05/workshop.html look at the NW Air ppt
> slides. AFS is accepted by the FAA.
actually it is United, not NW. The title is "AFS Cell Profile" and the
talk was held by Dexter Kimball. Now you should be able to find it ;-)
--
F
Hi Lars,
Thus spake Lars Schimmer ([EMAIL PROTECTED]):
> But: Where is the difference between RO copies and a backup volume?
a backup volume can only exist on the same partition as the RW volume,
since they use 'copy on write' to save the differences between the RW at
the time you updated (vos ba
Thus spake Derrick J Brashear ([EMAIL PROTECTED]):
> >What's the reason for this behaviour? What has changed between before
> >and after the first attempt to start the client? Just the kernel
> >error I pasted in my previous mail or is there something else forcing
> >me to reboot?
>
> no, the kern
Hi,
Thus spake Derrick J Brashear ([EMAIL PROTECTED]):
> you'll need to reboot. otherwise, it won't work.
Thanks!
What's the reason for this behaviour? What has changed between before
and after the first attempt to start the client? Just the kernel error I
pasted in my previous mail or is there
Hi,
Thus spake Derrick J Brashear ([EMAIL PROTECTED]):
> there's a bug when mount fails that makes shutting down fail. No one
> has worked out a fix yet.
Is it save to start the client again (after stopping all instances of
afsd and removing the module) without risking a kernel panic or
similarly
Hi,
I'm running a debian stable on i686, kernel 2.6.14.2. Downloaded the
debian sources for openafs-1.4.0-3 from "testing" and compiled it with
debuild. Installed the client, build and installed the kernel module.
When I started the client, it ran into a firewall between the client and
the db-ser
Hi,
Thus spake Noel Yap ([EMAIL PROTECTED]):
> From what I've heard, NFSv4 provides the security you mention above.
> Is this not true?
Well, next year.
Independent of _when_ you ask.
> Also, I'm not sure what you mean by 'flexible ways to manage storage'.
> Could you elaborate, please?
vos m
Thus spake Joe Buehler ([EMAIL PROTECTED]):
> Jim Rees wrote:
> > You need "StrictModes no" in sshd_config.
>
> This seems like a bad idea for security reasons...
Well ... erm ... since afs doesn't care about these permissions anyhow
you're talking about the security-by-obscurity concept, without
Thus spake Joe Buehler ([EMAIL PROTECTED]):
> My question is, where does the mode 777 come from?
Well, who created the directory?
> Is there any real reason for it to be 777 given that it's the AFS
> mount point? Wouldn't 755 be a better mode?
[10:38] [EMAIL PROTECTED]:~ $ ls -dl /afs
drwxr-xr-
Thus spake Horst Birthelmer ([EMAIL PROTECTED]):
> >ext3 is fine, as is ext2. reiserfs ist evil ...
> If Hendrik checked that, maybe it's fine.
> I had trouble with that on Linux.
What kind of trouble? I'm running my cache on ext3 for a year or so,
without problems (linux on x86). So I'd be really
Thus spake Horst Birthelmer ([EMAIL PROTECTED]):
> Your cache is on an ext3 filesystem, which is not supposed to be that
> way, and can cause trouble.
ext3 is fine, as is ext2. reiserfs ist evil ...
--
Consistency: Every time you release an apple over Sir Isaac Newton,
it will drop on his head.
Hi,
Thus spake Stefaan ([EMAIL PROTECTED]):
> I'm asking mainly because I'm packaging this for gentoo, and putting a
> disk cache on /var/cache/openafs is the default.
Uuuh. While it _is_ possible to have the cache on a partition which is
also used for other stuff (I'm currently doing that on my
Hi Stefaan,
Thus spake Stefaan ([EMAIL PROTECTED]):
> I have the feeling I will be shot for asking this, but here goes :)
don't you think we have more elegant methods? ;)
> I have my cache on my root filesystem, in /var/cache/openafs. It
> works, but is it supported? It'd be nice if it was, b
Thus spake Madhusudan Singh ([EMAIL PROTECTED]):
> Can the cache be located on a reiserfs partition as well ?
No, the cache can't.
--
Consistency: Every time you release an apple over Sir Isaac Newton,
it will drop on his head. That's good.
___
OpenAFS
Thus spake Nathan Neulinger ([EMAIL PROTECTED]):
> If you mean namei, there's no reason it shouldn't work.
Well ... with 2.4 I observed corrupt data on linux software raids under
high load. I burned my fingers, wouldn't do it again. Never tried 2.6.
Hendrik
--
More than gold, lead changed
Thus spake Derrick J Brashear ([EMAIL PROTECTED]):
> Incidentally, Linux isn't Unix(tm), and my SunOS 4 system (which
> strictly speaking is BSD and not Unix(tm) either) didn't do it. And
> IIRC neither does Solaris, which is probably the most Unix(tm) of the
> bunch I routinely use.
And neither
Thus spake Jeffrey B. Woodward ([EMAIL PROTECTED]):
> svnadmin create *svn* -fs-type *fsfs
>
> rather than the default berkeley-db backend.
Ah, thanks!! I missed that one -- was looking at a too old version of
the svnbook. :-)
--
Die gesellschaftliche Konversation waere ein ausgezeichnetes
Sc
Hi,
I'd like to use subversion, but it uses berkeley-db for the repository.
Since berkeley-db uses mmap for locking (right?), in the svn
documentation they write
"As opposed to CVS, subversion is not based on RCS, but rather on the
Berkeley Database. Make sure not to install a repository on
Thus spake ajpearce ([EMAIL PROTECTED]):
> Some answers I was hoping for:
>
> "Oh yeah, I heard about this guy who hacked together an NFS server
> with a ext3 IFS, downloadable at foo.com"
> "OpenAFS will do exactly what you want and easily" - nearly got this
Uhm, I still don't even know what e
Thus spake Jason McCormick ([EMAIL PROTECTED]):
> In my experience, vim has had other problems with AFS as well. If
> you're klog-ed as another user other than the UNIX "owner" of the
> file, vim barfs warnings and errors all over the place.
I've never seen this behaviour, and vim is the one a
Hi,
I didn't have the time to follow all the discussions recently. Anybody
managed to run the openafs client on x86 Linux 2.6.10?
Cheers,
Hendrik
--
Wisdoms from "Illinois Rules of the Road" (fifth item of 'left turns'):
'Point the wheels straight ahead until you actually start to turn.'
Thus spake Derrick J Brashear ([EMAIL PROTECTED]):
> >Never heard of such a file ... Anyway what does $HOME have to do with fs?
>
> that's where it will look for the .AFSSERVER file... which should only
> matter if you're using the afs-nfs translator.
^^^
Thus spake Garry Glendown ([EMAIL PROTECTED]):
> OK; seems like I either missed something in the docs, or they're not
> accurate ... I saw the the fs command tried to read .AFSSERVER ... after
> creating the file in $HOME, with the local servername in it, the fs
> commands I tried before succee
Thus spake Garry Glendown ([EMAIL PROTECTED]):
> like e.g. the init.d/afs script ... not in the "dest" hierarchy,
> either ... anyway, got that one from the RPM ...
/usr/src/openafs-1.3.73/i386_linux26/dest/root.client/usr/vice/etc/afs.rc
--
Wisdoms from "Illinois Rules of the Road" (fifth item
Thus spake Steve Wray ([EMAIL PROTECTED]):
> >You have to define "support" more clearly. The 1.3.7x line works with
> >2.6, with a short list of caveats.
>
> Does that short list still include read-only and no authentication?
Hm???
You're not confusing the kernel-built-in-want-to-be-afs-cli
Thus spake Derrick J Brashear ([EMAIL PROTECTED]):
> So Solaris, Linux, Darwin are all one platform now? Is that platform
> "Unix"? ;-)
I'd love to see netbsd on alpha and on sparc64 as well, but I have no
time contributing to it, sorry ... :-/
--
"Ich habe es ja auch nur wegen svn getan."
Thus spake Jeffrey Altman ([EMAIL PROTECTED]):
> The thing which is preventing the release of 1.3.7x as a stable 1.4
> tree is lack of deployment and testing by users. There has been very
> little feedback both positive or negative on the existing releases.
I'm using 1.3.73 as client on i386 Lin
Thus spake Jeffrey Hutzelman ([EMAIL PROTECTED]):
> >Same here. Kernel 2.6.9 Gentoo Linux, OpenAFS 1.3.73 Client. Reading
> >works fine, writing a few bytes and the cache manager crashes.
>
> Neither of you have said what architecture you're talking about, so
> I'll assume it's probably i386.
Ye
Thus spake Tom Fischer ([EMAIL PROTECTED]):
> My Kernel is also vanilla. Are there some Kernel Options which are
> related to afs?
nothing I know.
--
It's not the people who vote that count.
It's the people who count the votes.
(Joseph Stalin)
__
Thus spake Helmut Jarausch ([EMAIL PROTECTED]):
> I try to build the cvs version of OpenAFS (from today) unter Linux
> 2.6.10-rc2 The compilation fails with the following error message.
>
> /OBJ/OpenAFS/openafs-cvs-041202/src/libafs/MODLOAD-2.6.10-rc2-MP/afs_osi.c:
> In function `afs_osi_Travers
Thus spake Tom Fischer ([EMAIL PROTECTED]):
> Same here. Kernel 2.6.9 Gentoo Linux, OpenAFS 1.3.73 Client. Reading
> works fine, writing a few bytes and the cache manager crashes.
No problem here. Kernel 2.6.9 vanilla, openafs 1.3.73 client. Not a
single crash yet, both reading and writing.
--
Thus spake Chris Huebsch ([EMAIL PROTECTED]):
> >This raises the question how many group/users can be listed in an ACL?
>
> It seems that this limit is 20.
That's the number I also have in memory.
--
It's not the people who vote that count.
It's the people who count the votes.
Thus spake Derrick J Brashear ([EMAIL PROTECTED]):
> On Wed, 24 Nov 2004, Marc Schmitt wrote:
>
> >This problem still exists in 1.2.13 under RHEL 3. I'm successfully using
> >the patch posted at http://grand.central.org/rt/index.html?q=3812 by Eric
> >Sturdivant and I guess the one of Karl E. K
Thus spake Lars Schimmer ([EMAIL PROTECTED]):
> HihO!
HiHo!
> Breach is: for homedirs are RO pretty useless in kind of loadbalacing,
> or?
Yup.
> Because the users has to use the RW volume, the RO copies are
> senseless.
Right.
> In my eyes this is a breach in kind of loadbalacing the traffi
Thus spake Lars Schimmer ([EMAIL PROTECTED]):
> The root.cell is a volume on the first database/fileserver, and ro
> copys of it on every fileserver we've got. Yes, it's independet of the
> databse servers, but it describes the cell and should be mounted RW on
> at least one server.
It's mounted
another update:
Thus spake Hendrik Hoeth ([EMAIL PROTECTED]):
> Apparently they removed TASK_ZOMBIE from the kernel between 2.6.9 and
> 2.6.10-rc1. In 2.6.9 it's still available, though I didn't try to build
> openafs with 2.6.9 yet -- that's my next step.
With 2.6.9 i
Thus spake Hendrik Hoeth ([EMAIL PROTECTED]):
> /usr/src/openafs-1.3.73/src/libafs/MODLOAD-2.6.10-rc1-MP/afs_osi.c:861:
> error: `TASK_ZOMBIE' undeclared (first use in this function)
Apparently they removed TASK_ZOMBIE from the kernel between 2.6.9 and
2.6.10-rc1. In
Hi,
I tried to compile openafs with a 2.6 linux kernel for the first time
(new laptop, and I need some 2.6 features). Using openafs-1.3.73 I did
./configure --enable-transarc-paths
make
and got the following error:
CC [M] /usr/src/openafs-1.3.
Thus spake Rodney M Dyer ([EMAIL PROTECTED]):
> I don't know anyone who runs a cell on one server machine.
Well, I do. At home on my private PC. Usefull for connecting the laptop
or doing experiments I don't want to do in the production environment at
university ... ;-)
Hendrik
--
It's not t
Hi,
has anybody built the openafs client (I don't need the server) for
netbsd on sparc64 and alpha? Any hints on how to do this?
Cheers,
Hendrik
--
"Lord grant me the serenity to accept the things I cannot change,
the courage to change the things I can, and the wisdom to hide the
bodi
Hi Martin,
> Just read this comment, then think and write if you don't think it's
> futile.
Just read the archive and see how many times Derek gives highly valuable
hints and solutions for all kind of problems. Then think if introducing
oneself to a new community by offending people is a wise ide
Thus spake Rodney M Dyer ([EMAIL PROTECTED]):
> True. But that problem only occurs because the kernel code allows all
> memory to be read by "root". It would be nice if all OS's has a
> "protected store" memory area who's sections could only be mapped to
> each authenticated user.
And who compi
Thus spake Timothee Besset ([EMAIL PROTECTED]):
> I don't want to sound like I'm flamming or anything, but it's sad to
> see a project have it's technical decisions governed by the lowest
> skills set of it's user base (aka I can't compile my own kernels).
It's not that we can't compile our own k
Thus spake hanasaki ([EMAIL PROTECTED]):
> :( No support on Reiser?
No support on reiser.
--
Z decays have a tendency to be neutral.
___
OpenAFS-info mailing list
[EMAIL PROTECTED]
https://lists.openafs.org/mailman/listinfo/openafs-info
Thus spake John Koyle ([EMAIL PROTECTED]):
> Last I checked, Amanda suffers the same problem Thomas is trying to
> avoid - single volumes cannot span multiple tapes.
We use split for cutting large dumps into chunks of max. 2G.
--
Je mehr ich plane, desto haerter trifft mich die Wirklichkeit.
(
Hi,
> The good news is, it is running.
congratulations!
> I spent hours trying to understand uss.
:-)
> The bad news is that, in spite of using the pam modules, tokens are
> not being issued at login time. I have created a user with the same
> UID, password and login name, configured pam.d/log
Hi,
I've got a small but annoying problem. My configuration is:
- openafs-client (plain afs, no third-party kerberos)
- openssh-3.7.1
- pam
When I login via ssh, I won't get a new token (though I can login). If
I then use klog to obtain a token, logout (no unlog), ssh again, I have
the token wh
53 matches
Mail list logo