Alan Cox announced 2.4.9-ac10, and Mike Fedyk asked if there were any plans to merge Robert Love's patch to allow network devices to contribute to the entropy pool for /dev/random. Robert replied:
While I would love to see the network entropy patch merged into 2.4 (its an option -- plus, since it standardizes the handfull of drivers that currently _do_ contribute to the entropy pool, its a nice cleanup) I am aiming for 2.5.
I keep the patch up to date with both Linus's and Alan's tree. Patches against the current kernel can be found at http://tech9.net/rml/linux/ - 2.4.9-ac10 patches are up.
Peter J. Braam announced:
I have just released InterMezzo 1.0.5.2. This is a further test version ramping up for a stable Linux 2.4 release in 1.0.6. We now believe we have fixed most problems in InterMezzo related to Linux 2.4. This code is released under the GPL.
WHAT IS INTERMEZZO?
InterMezzo is a high availability file system which replicates directory trees among systems. It provides disconnected operation, journal recovery and kernel level write back caching. It can use the rsync algorithm for synchronization. It uses protocols somewhat similar to Coda's.
This release includes a kernel rpm (2.4.9-ac5). The 2.4 -ac series includes intermezzo and this kernel includes a minor extra intermezzo patch to pure -ac.
WHAT'S NEW IN THIS RELEASE?
- it works with ordered data, provided you use the latest ext3 (Stephen Tweedie)
- replicates ACL's and other extended attributes (requires kernel with EA, not packaged) (Shirish Phatak)
- better handling of InterMezzo specific metadata (Phil Schwan and me)
WHAT'S NEXT?
- data on demand
- better handling of NFS servers
- DAFS style network packets
- better handling of (false) conflicts for laptop users
DISCLAIMER:
Read the file COPYING in the distribution to see the conditions under which this software is made available. Please use this version at your own risk and exercise care (back up your systems etc). [With the 2.2.19 kernels fewer problems are known, but a 2.2.19 kernel RPM is not included.]
WHERE TO GET IT:
You can get sources and rpms from
ftp://ftp.inter-mezzo.org/pub/intermezzo/current
Documentation is included and available at: http://www.inter-mezzo.org/
Or get code from the intermezzo project on sourceforge. Check out the CVS tag r1_0_5_2.
Pavel Machek asked if this code was released under the standard GPL,
or a variant. Peter replied, Standard GPL -
and there are a lot of copyright holders already, so it's solid.
Matt Domsch of Dell announced:
Dell is pleased to provide digest forms of the linux-kernel and linux-scsi mailing lists. These lists, along with some Dell-specific lists, are hosted at http://lists.us.dell.com/, using the GNU Mailman list software, on a 2x1GHz Dell PowerEdge 2450 server.
Two digest forms of linux-kernel are provided.
- linux-kernel-digest - sent every 100KB as needed
- linux-kernel-daily-digest - sent only once per day, somewhat under 1MB per.
Please direct any feedback about these lists to mailman@lists.us.dell.com.
Juan Piernas Conovas looked over the ext3 0.0.7a source and discovered that
it was impossible to use a block device as a journal. Christian Borntræger
pointed out that ext3 was already at version .99, and that Juan should look
over some more recent code. But Juan replied that he was using kernel version
2.2, and ext3 0.0.7a was the most recent code for that kernel. Christian
replied, good point. I am not sure if
ext3 is still maintained for linux 2.2, but I doubt it. Andrew or Stephen
should be able to answer this question. But with 0.0.7a it is not possible
to have an external journal.
Andrew Morton agreed that it wouldn't
be possible to have an external journal with that version, and added, Stephen is actively maintaining ext3 for the 2.2 kernels,
but it is definitely in "maintenance mode".
Stephen C. Tweedie added,
Critical bugfixes only for 2.2. All current
development is 2.4 only. There are _tons_ of improvements in 2.4 which will
not be back-ported; 2.4 is the official current ext3.
Elsewhere he
added:
In the 2.2 code (ext3-0.0.7a), the journal has to be part of the filesystem. The current e2fsprogs and 2.4 ext3 code (ext3-0.9.9) does support journals on other devices, but there's more support to be added before it's truly complete (in particular we want to support several filesystems journaling to a single shared journal spool.)
The separate-journal-disk code hasn't been tested as much as the default journal-on-main-filesystem code, and it's not officially supported, so the format might change in the future. But it should work.
Andreas Dilger also mentioned, There was a
patch for 2.2 journal devices long ago, but I'm not sure if it was updated
to work with the new "format" of the external journal. It was posted to
ext2-devel by Marcelo Tossatti, probably 6 months ago or more.
H. Peter Anvin said:
kernel.org will be moving to a new location at the Internet Software Consortium (ISC) this coming Wednesday, September 19. For the duration of the move, ftp/www.kernel.org, filehub.kernel.org and its associated services will be unavailable.
Mirror sites will continue to be available, of course, and http requests will be redirected to a web page which has the mirror sites listed.
The new IP addresses will be as follows:
- 204.152.189.113: ftp.kernel.org, www.kernel.org, ns2.kernel.org
- 204.152.189.114: filehub.kernel.org
- 204.152.189.115: www.linuxbase.org
- 204.152.189.116: www.li18nux.org
- 204.152.189.117: nic.os.org
- 204.152.189.118: www.linuxeda.com
- 204.152.189.119: www.lanana.org
- 204.152.189.120: www.freestandards.org
Aaron Lehmann asked, Is IP-based virtual
hosting still in fashion?
H. Peter replied, It is if you care about anything but HTTP and mail.
Sujal Shah asked:
The Linux Coda drivers and the ext3 patches don't seem to get along very well, at least in Linux 2.4.7. I've got a stock 2.4.7 kernel with a patch applied to the USB drivers (for a sony digital camera; see http://www.sujal.net/tech/linux/ just a change in unusual_devs.h).
After I applied the ext3 patches from http://www.uow.edu.au/~andrewm/linux/ext3/. Basically, when an application tries to write to a file system mounted via coda, the application terminates with "Memory Fault" returned to the terminal. THe file system still thinks it's busy (can't umount).
Jan Harkes replied, Yeah, I know, and it will
probably work when you don't enable data-journalling. Coda's kernelmodule
currently uses generic_file_read and generic_file_write on it's containerfiles,
which works for many filesystems. However, ext3fs (and tmpfs and several
others) have a filesystem specific write implementation and don't really
like being called with the generic functions. The patch is simple, but
I haven't made it yet.
A few hours later, he made and posted his
patch, but Florian Schaefer replied, I just
tried your patch together with coda-debug-client-5.3.15-1, kernel-2.4.9 and
ext3-2.4-0.9.6-249 and still get the BUG message
. Jan replied No, different bug.
He explained Your bug is unusual, because it happens during the mount, everybody
else successfully mounted Coda and could read files, it would just die when
trying to write anything.
He summarized, I'm
clueless how this could have happened, and I can't see how it could happen.
Perhaps a patch got botched, or your kernel is mixing up objects from before
the patch with later ones.