| From | Subject | Date |
|---|---|---|
| Matthew Dillon | Re: How much are filesystem images trusted?
I'm assuming you just poked the bits in the on-media UNDO FIFO to
create the failure condition and it isn't a bug per-say, right?
I think an assertion is fine, or even just have the mount return
a failure. Would you like to code up your patch suggestion? We
can commit it after the release.
-Matt
Matthew Dillon
<dillon@backplane.com>
| Jul 19, 11:27 am 2008 |
| Matthew Dillon | HEADS UP - last minute (non-bug-fix) HAMMER change to ma ...
I've decided to remove some confusion from the HAMMER utility, and also
remove some of the master id code from the filesystem, which was
designed for multi-master mode but really added confusion to the
whole master-slave setup since we don't support multi-master mode
in this release.
PFS's will now only be master or slave, there will be no master id
(0-15) setting and there will not be a no-mirror mode. A single
whole-mount master id or no-mirror mode can be ...
| Jul 19, 10:58 am 2008 |
| Matthew Dillon | Re: 2.0 BRANCH FREEZE - TODAY IS THE LAST DAY FOR COMMIT ...
I think we have to release tomorrow even if it means only the master
site has the packages. pkgbox's redirect will simply have to remain
set to point to just the master site for the next few days.
I am going to make two last minute changes today. I'm going to fix
one issue with HAMMER related to upgrading and downgrading a mirror,
and I am going to see if I can easily fix the silly language selector
in the installer. I'm kinda tired of English being 'default'. ...
| Jul 19, 9:52 am 2008 |
| Sascha Wildner | Re: 2.0 BRANCH FREEZE - TODAY IS THE LAST DAY FOR COMMIT ...
It should. But I guess one has to set the correct font manually first.
Sascha
--
http://yoyodyne.ath.cx
| Jul 19, 10:57 am 2008 |
| Matthew Dillon | Re: 2.0 BRANCH FREEZE - TODAY IS THE LAST DAY FOR COMMIT ...
It's mispelled. I'll fix it.
-Matt
Matthew Dillon
<dillon@backplane.com>
| Jul 19, 9:41 am 2008 |
| Matthew Dillon | Re: HAMMER and NFS?
That will definitely confused NFS. kill -HUP wouldn't work in that
case because mountd will still think that you didn't change any of
the exports.
-Matt
| Jul 19, 9:29 am 2008 |
| Dionysus Blazakis | How much are filesystem images trusted?
I've been looking at the HAMMER code a bit. It seems the mount will
hang the kernel at recovery time if the tail of a undo record contains
a zero size. I've been told the filesystem is implicitly trusted, but
I think a failed assert would be better than the stuck while loop.
I have a small disk image to illustrate the hang at:
http://leaf.dragonflybsd.org/~dion/hammer.small.bz2
This obviously isn't a high priority, but I'm interested in hearing
opinions on it (does this kind of bug interest ...
| Jul 19, 9:17 am 2008 |
| Robert Luciani | Re: pkg_search patch for 2.0
done.
--
Robert Luciani
Chalmers University of Technology, SWE
Department of Computer Science and Engineering
http://www.rluciani.com/rluciani@gmail.com.asc
| Jul 19, 5:08 am 2008 |
| Matthias Schmidt | pkg_search patch for 2.0
Hi,
I want to get this patch in before 2.0 is released:
http://leaf.dragonflybsd.org/~matthias/pkg_search_2.0.diff
Could someone with access to pkgbox please create a link from
pkgsrc-2008Q2/ to stable/, such that pkg_search(1) can access
http://pkgbox.dragonflybsd.org/packages/DragonFly-2.0/stable
Cheers
Matthias
| Jul 19, 2:06 am 2008 |
| Matthew Dillon | Re: HAMMER and NFS?
Both of these are the same problem. Definitely a bug... I think it's
just that I am not storing the correct value for the st_size field of
the PFS softlink. PFS softlinks are special, HAMMER expands them in
place to a full transaction id + PFS format. So the readlink()
returns a string with a different size then stat shows in st_size.
NFS is probably depending on st_size being correct.
However there is a secondary issue here which you have to be aware
of. ...
| Jul 18, 7:16 pm 2008 |
| YONETANI Tomokazu | Re: HAMMER and NFS?
That and restarting nfsd+mountd did the trick it seems. My fstab contained
'noauto' keyword, so by the time the NFS daemons started they didn't know
about the subdirectories inside HAMMER filesystem and just sending SIGHUP
I confirmed that all of the problems above are fixed now!
Thanks.
| Jul 19, 12:32 am 2008 |
| YONETANI Tomokazu | HAMMER and NFS?
Hi.
Do you have some known limitation(or bugs) related to NFS-exporting of
HAMMER filesystem? I've been playing with /etc/exports so as I can
upgrade my router to 2.0, and found something strange:
- can't NFS-mount an exported directory inside a HAMMER filesystem,
only root directory can be NFS-mounted (no errors from mountd though)
- can't NFS-mount an exported pseudofs (Permission denied), or
NFS-mount a pseudofs inside an exported HAMMER filesystem
(again, no errors from ...
| Jul 18, 5:51 pm 2008 |
| YONETANI Tomokazu | Re: HAMMER and NFS?
> # grep /HAMMER
Of course this should have been:
| Jul 18, 6:06 pm 2008 |
| Justin C. Sherrill | Re: 2.0 BRANCH FREEZE - TODAY IS THE LAST DAY FOR COMMIT ...
And if you're booting a 2.0 system, please test pkg_radd, too.
| Jul 18, 7:02 pm 2008 |
| Dylan Reinhold | Re: 2.0 BRANCH FREEZE - TODAY IS THE LAST DAY FOR COMMIT ...
Thanks for reading my error messages for me. I totally misread it...
Its running now....
Dylan
| Jul 18, 11:53 pm 2008 |
| Erik Wikström | Re: 2.0 BRANCH FREEZE - TODAY IS THE LAST DAY FOR COMMIT ...
Perhaps the version should be part of the path:
http://pkgbox.dragonflybsd.org/2.0/All/
--
Erik Wikström
| Jul 19, 3:05 am 2008 |
| Sascha Wildner | Re: 2.0 BRANCH FREEZE - TODAY IS THE LAST DAY FOR COMMIT ...
Hmm, will you change http://pkgbox.dragonflybsd.org/All/ to point to the
2.0 packages?
I wonder, how can we make the All/ redirect handle users of previous
DragonFly releases needing older packages?
Sascha
--
http://yoyodyne.ath.cx
| Jul 18, 7:16 pm 2008 |
| Simon 'corecode' Sch ... | Re: 2.0 BRANCH FREEZE - TODAY IS THE LAST DAY FOR COMMIT ...
I think we should hold off the release until packages are built and have
propagated to all mirrors. This might take several days, though.
Packages are large and connectivity is low.
cheers
simon
| Jul 19, 7:32 am 2008 |
| Dylan Reinhold | Re: 2.0 BRANCH FREEZE - TODAY IS THE LAST DAY FOR COMMIT ...
Ok what am I doing wrong.. I just sync'ed with 2_0 and rebuilt, now
trying to test the release builds.
# uname -a
DragonFly dfly_vm1.ocnetworking.com 2.0.0-RELEASE DragonFly
2.0.0-RELEASE #0: Fri Jul 18 22:56:51 PDT 2008
dylan@dfly_vm1.ocnetworking.com:/usr/obj/usr/src/sys/DFLY1 i386
# cd /usr/src/nrelease
{dfly_vm1}
/usr/src/nrelease
# make installer release
Unable to find cdrtools-ossdvd-2.01.1.36nb2.tgz, use the following ...
| Jul 18, 11:12 pm 2008 |
| Sascha Wildner | Re: 2.0 BRANCH FREEZE - TODAY IS THE LAST DAY FOR COMMIT ...
This message tells you that the required package could not be found in
/usr/release/packages, and not that it is not installed.
Do make installer fetch.
Sascha
--
http://yoyodyne.ath.cx
| Jul 18, 11:32 pm 2008 |
| Justin C. Sherrill | Re: 2.0 BRANCH FREEZE - TODAY IS THE LAST DAY FOR COMMIT ...
Oh, yeah. Changed now. If someone could test, that would be helpful, as
Well, we could change pkg_radd to build a path using the output of uname,
and then analyze the incoming request path on pkgbox, and then redirect
those requests to the appropriate folder based on that. I'm not going to
get that working in the next 24 hours, though.
| Jul 19, 8:21 am 2008 |
| Justin C. Sherrill | Re: 2.0 BRANCH FREEZE - TODAY IS THE LAST DAY FOR COMMIT ...
Only two of the package mirrors have even started to update, so it's going
to be a few days until the new build updates, and then more days until all
the files are copied out.
pkgbox has a full set of packages - a good chunk of them are built on
1.13, so there will be a warning at install, but that's it. Building from
source (which I think is somewhat more common with pkgsrc) is already an
option in any case. I don't think there's a need to wait.
| Jul 19, 8:41 am 2008 |
| previous day | today | next day |
|---|---|---|
| July 18, 2008 | July 19, 2008 | July 20, 2008 |
