Marcus Watts wrote:
I've built a cache manager with WORDS_BIGENDIAN defined.
Works fine.
There's some really screwey logic in src/config/afsconfig.h that
is supposed to set WORDS_BIGENDIAN . For aix 4.3, this starts
simply enough:
#define AUTOCONF_FOUND_BIGENDIAN 1
#define ENDI
Marcus Watts wrote:
It would be helpful (to me at least) if Steve Roseman could
verify that "pts" and other tools work. That would be useful
confirmation that it's the same problem I see.
-Marcus
pts seems to nominally work. I klog as ad
Jeffrey Altman wrote:
Steve Roseman wrote:
Derrick J Brashear wrote:
On Fri, 1 Dec 2006, Steve Roseman wrote:
I can get a token, but it errors when trying to access AFS filespace:
$ klog
Password:
$ tokens
Try the patch I committed yesterday to src/rxkad/private_data.h
I assume the one
Derrick J Brashear wrote:
On Fri, 1 Dec 2006, Steve Roseman wrote:
I can get a token, but it errors when trying to access AFS filespace:
$ klog
Password:
$ tokens
Try the patch I committed yesterday to src/rxkad/private_data.h
I assume the one with:
! afs_int32 ticketLen
I can get a token, but it errors when trying to access AFS filespace:
$ klog
Password:
$ tokens
Tokens held by the Cache Manager:
User's (AFS ID 258) tokens for [EMAIL PROTECTED] [Expires Dec 5 17:54]
--End of list--
$ cd /afs/cc/home/lusgr
$ pwd
afs: Tokens for user of AFS id 258 for cell
Frederic Gilbert wrote:
Hi,
We have 3 DB servers to upgrade, hardware + OS + OpenAFS (1.2.13 to 1.4.1).
These upgrades imply moving the data volumes out of the old server,
doing hardware and software install, and moving the data back into the
new server.
I estimate that upgrading one server co
Juha Jäykkä wrote:
No, actually, it's "if you want to run old servers, have fun"
Ok. I am now moving to 1.4.1 clients, but the servers need a little more
care. Is the procedure as simple as one would wish:
1) shut down *one* server (file or db)
2) install new version on it
3) restart
4) repea
Forgive the paranoia, but, I have 2 quick questions, requiring Yes or No
answers. (3 servers, Linux 2.4.22, Fedora Core 1)
1) Can I go from 1.2.11 File and DB servers to 1.4.0 by replacing the
/usr/afs binaries and restarting?
2) Can I do one system at a time (one a day)?
Thanks,
Steve
---
Derrick J Brashear wrote:
On Sun, 5 Mar 2006, Steve Roseman wrote:
Jason McCormick wrote:
After installing the new Fedora 4 kernel (2.6.14-1.1644_FC4), when you
insert the AFS module is now reports that it cannot find the syscall
table.
System call hooks will not be installed; proceeding
Jason McCormick wrote:
After installing the new Fedora 4 kernel (2.6.14-1.1644_FC4), when you
insert the AFS module is now reports that it cannot find the syscall
table.
System call hooks will not be installed; proceeding anyway
Starting AFS cache scan...found 1396 non-empty cache files (13%).
Joe Buehler wrote:
Hans-Gunther Borrmann wrote:
the client itself works very well. No problem to compile and run.
The AIX loadable authentication module afs_dynamic_auth is not usable. If a
user logs in and is authenticated by this module all commands he submits
segfault.
I believe I subm
Derrick J Brashear wrote:
On Fri, 15 Apr 2005, Hans-Gunther Borrmann wrote:
"make dest" terminates with an error message:
"terminates". The error is:
make: 1254-005 Ignored error code 1 from last command.
ignored.
afs_dynamic_auth and afs_dynamic_kerbauth are completely missing. The
1.3.80
versi
More data points:
AIX 5.1 ML7 and 1.3.78.
CC=cc ./configure --enable-transarc-paths --disable-pam --host=rs6000
--with-afs-sysname=rs_aix51
(compiling on a fast chrp machine, testing on an old MCA system, thus
the --host and --with-afs-sysname)
I can start AFS, cd to system:anyuser directories,
Bill Hays wrote:
We have three older afs db servers running on suns and are migrating to
three new servers running redhat linux. What we'd like to know is whether
we can take down the old servers, and renumber and rename the new servers
to match the three old servers so that we do not have to touch
Todd M. Lewis wrote:
Jim Rees wrote:
'/afs/isis' is a symbolic link, leading to a mount point for
volume 'root.cell'.
So you broke one of the most important features of afs, the global name
space. Why?
'Cause we're stupid? 'Cause I didn't want to make an already too long
message even long
Derrick J Brashear wrote:
looking at what mutexes can be in play here, i'm going to have to bet
on peer->peer_lock. the rest are init'd in rx_InitHost (rx_Init),
which seeing as we have a kernel rx listener, well, it must have run.
likewise, only shutdown_rx kills said mutexes.
Ok, this obvious
Horst Birthelmer wrote:
I don't know about AIX 4.x but that code works on AIX 5.2.
No reason why the lock wouldn't be initialized (well, at first glance)...
I assume something weird happened to the peer but I'm still looking.
Horst
Note that I'm running AIX 5.1. (I have no test systems capable of
Derrick J Brashear wrote:
The real problem is that I'm 100% unfamiliar with kernel debugging tools
on AIX.
But, I'll make a guess. If this patch finds it, I'll be surprised. If it
doesn't, I'll guess there's an uninitialized mutex, but I'd have to look
harder to see if that's detectable.
I can
18 matches
Mail list logo
