S.J.Chun pisze:
Are you sure on disabling crypt at debian side? For me, it seems that
you turned off crypt at centos(which is turned off by default), and
debian, you did not(which might be turned on by default?)
Crypt in server settings or in client settings? Where can I check this
setting?
I
15:12:25
Subject: Re: [OpenAFS] Speed difference between OpenAFS 1.4.x on Debian and
CentOS
S.J.Chun pisze:
Are you sure on disabling crypt at debian side? For me, it seems that
you turned off crypt at centos(which is turned off by default), and
debian, you did not(which might be turned
S.J.Chun pisze:
For debian, /etc/openafs/afs.conf.client in case you installed with package.
There you can find AFS_CRYPT and which should be false to make crypt off
On CentOS only options for AFS Client (located in
/etc/sysconfig/openafs) are:
AFSD_ARGS=-afsdb -fakestat
On Debian in
S.J.Chun pisze:
For debian, /etc/openafs/afs.conf.client in case you installed with package.
There you can find AFS_CRYPT and which should be false to make crypt off
With crypt disabled I get major speedup (25-31MiB/s) and this is very
similiar to the CentOS results.
So this mistery is
: [OpenAFS] Speed difference between OpenAFS 1.4.x on Debian and
CentOS
S.J.Chun pisze:
For debian, /etc/openafs/afs.conf.client in case you installed with package.
There you can find AFS_CRYPT and which should be false to make crypt off
With crypt disabled I get major speedup (25-31MiB/s
Michał Droździewicz wrote:
Is AFS_CRYPT really that needed that debian is turning this _ON_ by
default?
One of the benefits that AFS provides over other file systems
is privacy. For that you need crypt to be on.
The Windows client defaults to use of encrypted sessions as well.
smime.p7s
-BEGIN PGP SIGNED MESSAGE-
Hash: SHA1
Michał Droździewicz wrote:
Jeffrey Altman pisze:
Is AFS_CRYPT really that needed that debian is turning this _ON_ by
default?
One of the benefits that AFS provides over other file systems
is privacy. For that you need crypt to be on.
The
Jeffrey Altman [EMAIL PROTECTED] wrote:
MichaÅ, Droździewicz wrote:
Is AFS_CRYPT really that needed that debian is turning this _ON_ by
default?
One of the benefits that AFS provides over other file systems
is privacy. For that you need crypt to be on.
The Windows client defaults to use
Christopher D. Clausen wrote:
I think the better question is why CentOS has it _OFF_ by default.
Packages should fail safe by being in the safest operating mode by
default.
Agreed but then you get the folks who install AFS and perform
some tests and say NFS is 20 times faster, AFS sucks.
Jeffrey Altman [EMAIL PROTECTED] wrote:
Christopher D. Clausen wrote:
I think the better question is why CentOS has it _OFF_ by default.
Packages should fail safe by being in the safest operating mode by
default.
Agreed but then you get the folks who install AFS and perform
some tests and
Jeffrey Altman pisze:
Is AFS_CRYPT really that needed that debian is turning this _ON_ by
default?
One of the benefits that AFS provides over other file systems
is privacy. For that you need crypt to be on.
The Windows client defaults to use of encrypted sessions as well.
Ok, but if I'll
Jeffrey Altman wrote:
Christopher D. Clausen wrote:
I think the better question is why CentOS has it _OFF_ by default.
Packages should fail safe by being in the safest operating mode by
default.
Agreed but then you get the folks who install AFS and perform
some tests and say NFS is 20
What if OpenSSH left encryption turned off by default so people
could benchmark it against FTP?
According to http://www.globus.org/security/overview.html that's
exactly what the globus versions of the ssh stuff does: As default
turn encryption off in the gsi-ssh so it does not get in the way
Does turning crypt off mean data in transit can be read *and* tampered
with? Or read, but still safe from tampering?
Also, does this imply that a server participating in the public
directory is trusting that all clients are using encryption to connect
to it? Is there a way for a server to
Wesley Chow [EMAIL PROTECTED] wrote:
Does turning crypt off mean data in transit can be read *and*
tampered with? Or read, but still safe from tampering?
Also, does this imply that a server participating in the public
directory is trusting that all clients are using encryption to connect
to
Harald Barth wrote:
But has anyone here on this list experimented with HW-acceleration for
encryption? It might be a good investment for a server (I hope that
my clients should cope).
I doubt you will find a hw engine engine for fcrypt.
smime.p7s
Description: S/MIME Cryptographic Signature
In message [EMAIL PROTECTED],Christopher D. Clausen
writes:
setcrypt off command before running benchmarks. What if OpenSSH left
encryption turned off by default so people could benchmark it against
FTP?
openssh sucks.
___
OpenAFS-info mailing list
Wesley Chow wrote:
Does turning crypt off mean data in transit can be read *and* tampered
with? Or read, but still safe from tampering?
Also, does this imply that a server participating in the public
directory is trusting that all clients are using encryption to connect
to it? Is there a way
Jeffrey Altman wrote:
Harald Barth wrote:
But has anyone here on this list experimented with HW-acceleration for
encryption? It might be a good investment for a server (I hope that
my clients should cope).
I doubt you will find a hw engine engine for fcrypt.
I agree, but what kind of
Derrick Brashear pisze:
I care about kernel, not OS. What kernel version on those machines?
Default distribution kernel:
on Debian: 2.6.18-6-686 i686
on CentOS: 2.6.18-53.1.14.el5 i686
--
xmpp/email: [EMAIL PROTECTED]
xmpp/email: [EMAIL PROTECTED]
On Mon, Apr 7, 2008 at 2:42 AM, Michał Droździewicz
[EMAIL PROTECTED] wrote:
Derrick Brashear pisze:
I care about kernel, not OS. What kernel version on those machines?
Default distribution kernel:
on Debian: 2.6.18-6-686 i686
on CentOS: 2.6.18-53.1.14.el5 i686
Not what I expected.
Derrick Brashear pisze:
Not what I expected. When you self-compiled 1.4.6 on Debian, I assume
you downloaded a tarfile from OpenAFS and did ./configure; make, yes?
What options, if any, to configure?
I've build a debian package using default debian options (1.4.6) and
I've compiled from source
Michał Droździewicz wrote:
Derrick Brashear pisze:
Not what I expected. When you self-compiled 1.4.6 on Debian, I assume
you downloaded a tarfile from OpenAFS and did ./configure; make, yes?
What options, if any, to configure?
I've build a debian package using default debian options (1.4.6)
Hartmut Reuter, dnia 2008-04-07 16:59 napisal:
Are you sure your network interface is used in GBit/s mode with Debian
and not just 100MBit-mode?
1) Iface is in 1000Mib mode
2) copying files from local disk to AFS structure (iface is omitted in
this test) was slow, not the network copying
--
Derrick Brashear, dnia 2008-04-07 17:04 napisal:
and I've
compiled from source with no options for ./configure except from --prefix
In both cases the result was the same - slow speed around 8-12MiB (copying
from local disk to AFS structure)
Parameters you gave to afsd, in both (CentOS and
On Mon, Apr 7, 2008 at 10:15 AM, Michał Droździewicz
[EMAIL PROTECTED] wrote:
Derrick Brashear pisze:
Not what I expected. When you self-compiled 1.4.6 on Debian, I assume
you downloaded a tarfile from OpenAFS and did ./configure; make, yes?
What options, if any, to configure?
I've
On Mon, Apr 7, 2008 at 11:16 AM, Michał Droździewicz
[EMAIL PROTECTED] wrote:
Derrick Brashear, dnia 2008-04-07 17:04 napisal:
and I've
compiled from source with no options for ./configure except from
--prefix
In both cases the result was the same - slow speed around 8-12MiB
Derrick Brashear, dnia 2008-04-07 17:40 napisal:
The same in both configs.
Well, the kernel config options certainly aren't if you're using
CentOS's kernel in one case and Debian's in another.
:�§
I've compiled debian kernel package using kernel config from CentOS but
this was no help at
On Mon, Apr 7, 2008 at 11:55 AM, Michał Droździewicz
[EMAIL PROTECTED] wrote:
Derrick Brashear, dnia 2008-04-07 17:40 napisal:
The same in both configs.
Well, the kernel config options certainly aren't if you're using
CentOS's kernel in one case and Debian's in another.
:�§
What happens if you compare with memory cache in both cases? Could it
be the HD driver? A strace with the times for the different syscalls
might be interresting. And, eh, good luck.
Harald.
___
OpenAFS-info mailing list
OpenAFS-info@openafs.org
Derrick Brashear, dnia 2008-04-07 18:13 napisal:
Well, all that's left is compiling CentOS' kernel on Debian; If you're
willing it's certainly a valuable data point.
I'll try to test it tomorrow and will submit some new data.
--
Mike D
___
Harald Barth, dnia 2008-04-07 18:21 napisal:
What happens if you compare with memory cache in both cases? Could it
be the HD driver? A strace with the times for the different syscalls
might be interresting. And, eh, good luck.
Can't be HD driver - dd in both cases (Debian and CentOS) shows
openafs-info@openafs.org
Sent: 08-04-07 23:15:10
Subject: Re: [OpenAFS] Speed difference between OpenAFS 1.4.x on Debian and
CentOS
Derrick Brashear pisze:
Not what I expected. When you self-compiled 1.4.6 on Debian, I assume
you downloaded a tarfile from OpenAFS and did ./configure
Hello all,
Recently I've spent some time testing speed of OpenAFS server
installation on different machines and operating systems.
First I've tried to install OpenAFS with Debian 4.0 (etch) on 3
different machines (beginning from old Celeron, through Pentium 4 and at
the end on Xeon 3GHz).
For starters: I'm replying to the list - maybe somebody would be interested.
Sergio Gelato, dnia 2008-04-06 21:20 napisal:
* Michał Droździewicz [2008-04-06 10:07:18 +0200]:
First I've tried to install OpenAFS with Debian 4.0 (etch) on 3
different machines (beginning from old Celeron, through
I care about kernel, not OS. What kernel version on those machines?
Lets say that the performance was like 250% better with 1.4.6 on CentOS 5
than with 1.4.2 on Debian 4.0. I've even compiled 1.4.6 and 1.5.34 on Debian
4.0 but performance was the same.
36 matches
Mail list logo
