Re: [OpenAFS] using afs for unix config files

[Garance A Drosihn]

At 2:11 PM -0700 7/27/06, David Bear wrote:

I am wondering how wise it is to use afs to store config
files like

/etc/hosts
/etc/resolv.conf
...

can anyone comment on their experience using afs to store
these files and which config files seem to be less well
suited to putting in afs?


It depends on what you mean by "putting them in afs"...

We put plenty of our config files in afs, which is useful
because we make sure afs is well-backed up.  But we do
not run those machines with them reading config files
directly from afs.  Instead, we have a crontab-controlled
job which run once a day, and compares the versions of
the files in afs with the versions on the local hard
disk.  If there is a difference, we copy the files from
afs to the local hard disk, and possibly reboot the
machine.  Thus, we could change something like /etc/hosts
on a lot of machines by just changing one file in afs.

but local hard disks are certainly big enough now that
the files all fit quite well on the local machine, and
there's no good reason to have every DNS lookup running
off into /afs space just to read /etc/resolv.conf.

--
Garance Alistair Drosehn=   [EMAIL PROTECTED]
Senior Systems Programmer   or  [EMAIL PROTECTED]
Rensselaer Polytechnic Instituteor  [EMAIL PROTECTED]
___
OpenAFS-info mailing list
OpenAFS-info@openafs.org
https://lists.openafs.org/mailman/listinfo/openafs-info

Re: [OpenAFS] using afs for unix config files

[John Rudd]

We have been doing that (having master copies in AFS and copying them 
down to local systems when they get changed) for certain files (passwd 
file, aliases file, etc.), but we're moving away from using AFS for 
this, and toward using cfengine.  Partially because several of the 
sub-groups in our large group aren't using AFS, and we want one 
solution that goes across the entire large group.


But, it does work.  It's a good strategy if you're widely using AFS.  
It's a much better strategy, in my opinion, than actually having 
critical resources exist _in_ AFS.  When I first got here, there were 
machines that either couldn't boot without having access to AFS, or 
were unusable without access to AFS.  This meant lots of extra hoops 
for simple things like "booting to single user mode".  So my peers and 
I started moving away from "/bin/* and much of /etc/* are symlinks into 
AFS" and toward "master copies are in AFS, and regularly copied down to 
local locations if a change is detected".  Made things MUCH better in 
our environment.


You can also use CVS with auto-updating, and a few other mechanisms.  
But if you've got AFS, might as well leverage it.  And, you might be 
able to "store the master copies in AFS, but use cfengine for managing 
how it gets used locally" ... but I'm not as familiar with cfengine's 
internals so I'm not sure about the details there.



On Jul 27, 2006, at 4:41 PM, Brian Sebby wrote:


We use AFS to store the "master" copies of various files like that that
should be the same on all systems, but the systems don't actually look 
at
the AFS copies for normal operation.  We have scripts that copy the 
master
versions to the local /etc directory and other local directories.  To 
us, it's
a good compromise, because you really want the critical files to be 
local to
the machine, but you can use AFS to only have to modify the files in 
one
place, and then just push out the changed files to each of the 
machines.


We do this for files like /etc/pam.conf, sendmail.cf, and ntp.conf.


Brian

On Thu, Jul 27, 2006 at 02:11:44PM -0700, David Bear wrote:

I am wondering how wise it is to use afs to store config files like

/etc/hosts
/etc/resolv.conf
...

can anyone comment on their experience using afs to store these files
and which config files seem to be less well suited to putting in afs?

--
David Bear
phone:  602-496-0424
fax:602-496-0955
College of Public Programs/ASU
University Center Rm 622
411 N Central
Phoenix, AZ 85007-0685
 "Beware the IP portfolio, everyone will be suspect of trespassing"
___
OpenAFS-info mailing list
OpenAFS-info@openafs.org
https://lists.openafs.org/mailman/listinfo/openafs-info


--
Brian Sebby  ([EMAIL PROTECTED])  |  Unix and Operation Services
Phone: +1 630.252.9935|  Computing and Information Systems
Fax:   +1 630.252.4601|  Argonne National Laboratory
___
OpenAFS-info mailing list
OpenAFS-info@openafs.org
https://lists.openafs.org/mailman/listinfo/openafs-info


___
OpenAFS-info mailing list
OpenAFS-info@openafs.org
https://lists.openafs.org/mailman/listinfo/openafs-info

Re: [OpenAFS] using afs for unix config files

[Brian Sebby]

We use AFS to store the "master" copies of various files like that that
should be the same on all systems, but the systems don't actually look at
the AFS copies for normal operation.  We have scripts that copy the master 
versions to the local /etc directory and other local directories.  To us, it's
a good compromise, because you really want the critical files to be local to 
the machine, but you can use AFS to only have to modify the files in one 
place, and then just push out the changed files to each of the machines.

We do this for files like /etc/pam.conf, sendmail.cf, and ntp.conf.


Brian

On Thu, Jul 27, 2006 at 02:11:44PM -0700, David Bear wrote:
> I am wondering how wise it is to use afs to store config files like
> 
> /etc/hosts
> /etc/resolv.conf
> ...
> 
> can anyone comment on their experience using afs to store these files
> and which config files seem to be less well suited to putting in afs?
> 
> -- 
> David Bear
> phone:602-496-0424
> fax:  602-496-0955
> College of Public Programs/ASU
> University Center Rm 622
> 411 N Central
> Phoenix, AZ 85007-0685
>  "Beware the IP portfolio, everyone will be suspect of trespassing"
> ___
> OpenAFS-info mailing list
> OpenAFS-info@openafs.org
> https://lists.openafs.org/mailman/listinfo/openafs-info

-- 
Brian Sebby  ([EMAIL PROTECTED])  |  Unix and Operation Services
Phone: +1 630.252.9935|  Computing and Information Systems
Fax:   +1 630.252.4601|  Argonne National Laboratory
___
OpenAFS-info mailing list
OpenAFS-info@openafs.org
https://lists.openafs.org/mailman/listinfo/openafs-info

Re: [OpenAFS] using afs for unix config files

[Daniel O Winkless]

YMMV, but,
Personally, I'd like to be able to comment
out the AFS start up and still have my system run after a reboot.  Since
/etc/hosts is crucial (at least in Solaris) for the system to learn about
itself, I would not make that info dependent on any other systems.  
I have a similar feeling about /etc/resolv.conf.
I could imagine /etc/hosts.equiv being
in AFS.  I no longer use r commands, so don't have this file on my
systems..
I'm not sure what else might be suitable.
         
      Dan





David Bear <[EMAIL PROTECTED]>

Sent by: [EMAIL PROTECTED]
07/27/2006 03:11 PM



Please respond to
[EMAIL PROTECTED]





To
openafs 


cc



Subject
[OpenAFS] using afs for unix config
files








I am wondering how wise it is to use afs to store
config files like

/etc/hosts
/etc/resolv.conf
...

can anyone comment on their experience using afs to store these files
and which config files seem to be less well suited to putting in afs?

-- 
David Bear
phone:                
 602-496-0424
fax:                  602-496-0955
College of Public Programs/ASU
University Center Rm 622
411 N Central
Phoenix, AZ 85007-0685
 "Beware the IP portfolio, everyone will be suspect of trespassing"
___
OpenAFS-info mailing list
OpenAFS-info@openafs.org
https://lists.openafs.org/mailman/listinfo/openafs-info

Re: [OpenAFS] using afs for unix config files

[Jim Rees]

At one time we had a list of the 11 files that must be present on the local
disk to bring up afs on at RT.  Our "minimal" install is now 1700 files.

I would not want to put hosts or resolv.conf in afs.  I think you could, if
you're not using afsdb, but I still wouldn't want to.
___
OpenAFS-info mailing list
OpenAFS-info@openafs.org
https://lists.openafs.org/mailman/listinfo/openafs-info

[OpenAFS] using afs for unix config files

[David Bear]

I am wondering how wise it is to use afs to store config files like

/etc/hosts
/etc/resolv.conf
...

can anyone comment on their experience using afs to store these files
and which config files seem to be less well suited to putting in afs?

-- 
David Bear
phone:  602-496-0424
fax:602-496-0955
College of Public Programs/ASU
University Center Rm 622
411 N Central
Phoenix, AZ 85007-0685
 "Beware the IP portfolio, everyone will be suspect of trespassing"
___
OpenAFS-info mailing list
OpenAFS-info@openafs.org
https://lists.openafs.org/mailman/listinfo/openafs-info

Re: [OpenAFS] "connection timed out" after salvage completes

[Russ Allbery]

Adam Megacz <[EMAIL PROTECTED]> writes:

> My fileserver seems to want to salvage every time the machine boots, but
> that's another story...

Make sure your system shutdown process is cleanly shutting down the file
server.

> It seems that if I access a volume being salvaged (OpenAFS 1.4.1 Linux
> client), I get the usual "connection timed out" error... but once the
> volume finishes salvaging and comes on-line (and other clients can
> access it), the client that got the error continues getting the error
> for several minutes.

> Is this the expected behavior, or should I narrow down the problem
> further and file a bug report?

It's expected; when a file server is down, the cache manager will mark the
host as down and won't retry for some interval (five minutes sticks in my
head).  You can force an immediate check with fs checkservers.

-- 
Russ Allbery ([EMAIL PROTECTED]) 
___
OpenAFS-info mailing list
OpenAFS-info@openafs.org
https://lists.openafs.org/mailman/listinfo/openafs-info

[OpenAFS] "connection timed out" after salvage completes

[Adam Megacz]

My fileserver seems to want to salvage every time the machine boots,
but that's another story...

It seems that if I access a volume being salvaged (OpenAFS 1.4.1 Linux
client), I get the usual "connection timed out" error... but once the
volume finishes salvaging and comes on-line (and other clients can
access it), the client that got the error continues getting the error
for several minutes.

Is this the expected behavior, or should I narrow down the problem
further and file a bug report?

  - a

-- 
PGP/GPG: 5C9F F366 C9CF 2145 E770  B1B8 EFB1 462D A146 C380

___
OpenAFS-info mailing list
OpenAFS-info@openafs.org
https://lists.openafs.org/mailman/listinfo/openafs-info

[OpenAFS] uss bulk wont work

[Rainer Laatsch]

The following error happens persistently (single RW volume) on all user
homedirs:
#  uss delete zzabic1 /afs/.rrz/usr/smail/33/zzabic1  -verbose \
-admin admin
No cell specified; assuming 'rrz.uni-koeln.de'.
Password for 'admin':
Translating user 'zzabic1' via the Protection DB
uss: Volume 'user.smail.zzabic1' (ID 537292558) exists on multiple
servers!!

 Our sync site runs RedHat Linux 2.4.21-27.0.2.ELsmp with
OpenAFS 1.2.13. The same holds for our other servers and the client where
the query was done.

Querying from a Fedora host (Kernel 2.6.17-1.2139_FC5) with
OpenAFS 1.4.1 yields a different message:

# uss delete zzabic1 /afs/.rrz/usr/smail/33/zzabic1 -dryrun -verbose \
-admin admin
No cell specified; assuming 'rrz.uni-koeln.de'.
Password for 'admin':
Translating user 'zzabic1' via the Protection DB
uss: Can't get volume information from mountpoint \
'/afs/.rrz/usr/smail/33/zzabic1'

Doing it under 'uss bulk' with input lines like this
delete zzabic1:/afs/.rrz/usr/smail/33/zzabic1:delvolume
I get the same behaviour.
(Please correct the missing '%' in the printf statement for
'Volume exists on multiple servers' around lines 1121-1130
in src/uss/uss_vol.c)

Somebody some help or hint? (Return to Solaris?)
Best regards
Rainer Laatsch
__
E-mail: [EMAIL PROTECTED]   Universitaet zu Koeln
Reg. Rechenzentrum (ZAIK/RRZK)
Fax   : (0221) 478-5590 Robert-Koch-Str. 10
Tel   : (0221) 478-5582 D-50931 Koeln

___
OpenAFS-info mailing list
OpenAFS-info@openafs.org
https://lists.openafs.org/mailman/listinfo/openafs-info