> > is that all files are equal, and doing away with abominations like
> > symlinks, device nodes and ioctls.
>
> the 9P semantics for those things would be `wrong' (for the suggested use)
> anyway.
> a remote 9P's device node would access the remote device, not the local one,
> which
> is usefu
> I learned something sad about the motorola and trolltech "open"
> phones. You can get the phone, get the source, build an OS.
>
> You can't load the OS you build.
>
> This just dropped them into the bucket for me. Not sure about open
> moko, I expect it is similar.
>
> Too bad. Cell phone comp
> The difference wrt lock/unlock is that it does not spin. If the lock
> cannot be set, the
> thread is put to sleep in queue waiting for the lock. So, it's better
> to use qlock in
> general than it is to use lock. (IIRC, lock is used to protect the
> data structure of the QLock,
> that might giv
> p.s. met the coraid guys at linux world yesterday. Very, very fine
> stuff. I love the fact that they are 'the linux storage company'.
they wouldn't let me put bunnies on the signage!
- erik
you might check out 9p2000.u. plan 9 doesn't do symlinks.
- erik
i think that you want to read
http://www.cs.bell-labs.com/magic/man2html/5/stat
also look at the /sys/include/libc.h for the following comments
/* bits in Qid.type */
/* bits in Dir.mode */
- erik
> BTW: I'm still interested in functional differences between
> FUSE and 9P. For example, does 9P support all *nix style
> inode types (ie. symlinks, devices, pipes, etc) ?
there are no inodes in 9p. so no.
- erik
> I'm host a lot of web applications which share 99% of their code.
> Disk space is not the issue, but bandwidth on remote backup.
> So my idea is to let an filesystem automatically link together
> equal files in the storage, but present them as separate ones.
> Once an file gets changed, it will
why not just tie the fid to a "file" version. then it would be easy
to handle offsets.
- erik
On Thu Aug 9 21:47:02 EDT 2007, [EMAIL PROTECTED] wrote:
> Hi,
>
> I am trying to understand the general practice, under
> Plan 9, for synthetic filesystems for serving up a "file"
> which:
> 1) retur
qlock(2) doesn't block all threads in a proc. i'm terrible
at explaining code, but here it goes
from the source, you'll see that the thread library sets
up (/sys/src/libthread/main.c:35)
_qlockinit(_threadrendezvous);
what does this do? in /sys/src/libc/9sys/qlock.c we see
that thi
> > if you want a really cheep aoe device. use vblade on a pair of
> > fs(3)-mirrored sata drives. i do that at home.
>
> Yes, that's on my wish list. And if I read the documentation
> correctly, using two vblades would give me RAID-5 with a live spare.
fs(3) has no raid5.
> Sadly, by the tim
> I hope this doesn't sound like teaching grandma to suck eggs, but
> would laptop-sized disks not help at least with dimensions, I'm not
> sure about price?
that's a good point.
our first product used laptop drives. at the time one could get more
bytes/u of rack space with laptop drives. that'
On Thu Aug 9 12:57:04 EDT 2007, [EMAIL PROTECTED] wrote:
> > and it's stable.
>
> or if it's not, at least you stand half a chance of finding and fixing it.
sorry. that wasn't clear.
i ment stable as in the scheduler doesn't change with the seasons.
- erik
> For the morbidly curious, is there any documentation regarding coraid's use
> of plan9 that you all wouldn't mind sharing? I'd love to see how the
> decision was made to use plan9, and what features were critical to the
> decision making process. It would be an interesting read, for me anyway.
>
> hello
>
> i found this one interesting:
>
> http://www.amazon.com/LaCie-Ethernet-External-Drive-300963/dp/B000BKJ5Z0
>
> 1TB for about 650USD (about 400USD second hand)
>
> is still a bit expensive, but may be it works. I don't know if coraid
> has something like it (i think the 1U appliance
coraid is pleased to announce a plan 9 aoe initiator.
the patch has been submitted as /n/sources/patch/aoe-initiator.
the initiator has been tested with linux vblade, plan 9 vblade, and
coraid sr hardware. we hope you use it with coraid sr hardware. ☺
there are two parts, the initiator proper, w
> If I recall correctly Rob Pike's comments in this forum a long time
> ago apropos the issue of customisation, I would suggest that such a
> chording pad would be more frustrating than useful. It goes almost
> without saying that the generality of such a device lends itself to
> insane re-configu
> Stupid question but ... how do we do AoE on Plan 9 at this point? I
> have not been paying attention.
thanks for asking. it's not a stupid question. the plan 9 initiator is
not quite ready but it will be pretty soon. the way i'm using it,
it seems pretty stable.
i am able to pxe boot a syste
i've put vblade, a software AoE target, on sources
in /n/sources/contrib/quanstro/src/vblade. this is a
personal project.
vblade serves a file over AoE. that is, the file
appears on the network as a block device.
i've currently got a small filesystem on my linux
machine mounted on an aoe targ
> We send mail from terminals all the time. cron is a backup, to run
> the queues to retry mail that didn't get delivered on the first
> atttempt.
not everone has a cpu server.
- erik
you've tried telnet tcp!mx.e271.net!smtp to verify that you can
get packets through?
i think that mail support on a terminal will be left wanting. upas
depends on cron.
- erik
switch($service){
case cpu
if (test -e /mnt/term/mnt/wsys) {
# rio already running
wsys = /mnt/term^`{cat /mnt/term/env/wsys}
bind -a /mnt/term/mnt/wsys /dev
if(test -f /m
ymmv. some computers don't mind being replugged. we do it to servers
all the time.
at home i use a cheep kvm ($29 at frys) to avoid replugging machines.
- erik
> >
> > > if i unplug the ps2 mouse from running terminal box, and then plug it in
> > again,
> >
> > I don't think you should be doin
we've been having reoccuring problems when sending troff output
to one hp 4xxx printer, but not an hp 5. the error message
from the printer is
minooka; cat log/prlisa.st
printer startup
%[ status: idle ]%
8192 sent
%[ status: busy ]%
16384 sent
the source on sources doesn't necessarly match the binaries
on sources.
- erik
i think the problem is that the pci id in your motherboard is not recognized.
i think you will need the following lines added to /sys/src/9/pc/sdata.c and
recompile
the kernel. (to be sure, could you send me the output of pci from your
machine.)
unfortunately, this will only be possible if you
> either grouping (well-behaved dogs) and interns or well-behaved (dogs
> and interns)
> in a rather ambiguous way. Can we bring bad-behaved interns?.
>
> The "and" is redundant in the presence of the "et" in the "et cetera",
> unambiguously meaning
> that there is a grouping or class of equivalen
> > bring friends, family, well-behaved dogs and interns, etc.
> Well behaved dogs and interns... Hmm, as an intern I am not sure
> I like the class of equivalence established here... Thanks I am
> bad-behaved.
there is no distributive property of adjectives over nouns except maybe
if the adjecti
/07, erik quanstrom <[EMAIL PROTECTED]> wrote:
>
> > "old" is a pervasive epithet with computer platforms
> > and sw, but i'm not sure it addresses any of the reasons
> > some bits are better than others.
>
> well, that's a good point. The bit
> On 7/31/07, maht <[EMAIL PROTECTED]> wrote:
>
> > It's not plan9's fault though I think, I've never found much use for a
> > PDA esp. when my phone does most of the PIM functions.
>
> do a greenphone or openmoko port? bitsy is so old, it's moldy.
>
> ron
as one who works with pee cees, i'll r
On Mon Jul 30 14:24:41 EDT 2007, [EMAIL PROTECTED] wrote:
> hello
>
> i have had problems too with sata, is your controller ahci capable?
>
> try stardard mode, compatible mode settings on your bios and see if it works.
>
> i have an intel ich7 without ahci and it works in standard mode, compati
the plan 9 install didn't recognize your controller. this could mean it's not
supported. it could also mean that you just need to change some bios settings.
what chipset does your motherboard have?
- erik
On Mon Jul 30 06:11:06 EDT 2007, [EMAIL PROTECTED] wrote:
> Hi,
>
> I am trying to insta
for some reason adding a user added to /adm/keys a few days ago also
added 11 bogus entries to my /adm/keys file. i submitted a patch with
some defensive code (perhaps not defensive enough) to protect against
this.
bad entries in /adm/keys will cause /mnt/keys to contain illegal file names.
and c
unfortuantely, it's not obvious to me which memmove is triggering
this panic.
- erik
ladd# fcpu1: registers for validateaddress 25320
aFLAGS=10202 TRAP=E ECODE=2 PC=F0195E51u SS=5F66 USP=F027F830
l AX 0082801C BX F082D098 CX DX F082D098
t SI 5000 DI 0001E000 BP 0002
CS
the length limit is actually 13 characters. the field is 14 characters wide and
null terminated.
- erik
i was just looking into that. thanks. we have a winner.
many thanks. why is it called the "secstore key" and not the "secstore
password"?
- erik
> I don't know about len limits, but mine has really weird characters and
> works fine in nvram. This may seem silly, but, have you checked out that
> your password is nvram is indeed ok? [that was a mistake I made]
> You could try after rewriting it.
well, the password does authenticate with the
i get this error
ladd# auth/secstore -nG factotum
ladd# echo $status
secstore 1517: invalid password in nvram
however if i do it this way, it works:
ladd# auth/secstore -G factotum
secstore password:
secstore
i have a longish password, b
> - Why was 8 1/2 abandoned? Is it because the graphics is much advanced
> over bitmap operations? Does plan9 use Xorg now? I think at least the
> idea of having a single private console sounds good.
plan 9 uses rio, which is a decendent of 8½. plan b, a close relative of plan
9,
uses omero.
>
heh. this is exactly the dance i was worried about.
for some reason, i didn't notice the -n option to
secstore.
alles klar.
- erik
the sshserve(1) (sic.) man page mentions how to generate a host
key on-the-fly, but doesn't mention where to store the key.
where is the recommend place to stash that key?
- erik
obviously that would fix the problem by overwriting your entire
venti store. if you're using venti, you need to run fshalt(no man page)
before you kill the emulator.
- erik
this looks like an error one gets when the machine is shutdown
without shuting down fossil and venti gracefully. are you running
the latest version of venti? you may want to run venti/checkindex.
- erik
by using static, alts is on the heap and not on the stack.
plan 9 doesn't default to huge stacks, so this makes a difference.
also, alts is automatically zeroed because it's on the heap.
the only difference between
void
fu(void)
{
static Alt alts[NKALT+1];
no need to be sorry.
we're in the same boat. my dsl in washington is still dead.
new dsl in georgia has the voice pair connected (which
i didn't want) but bell has forgotten to connect the
dsl pair to the dslam.
fun.
- erik
> First of all, we need a means to identify resources, both local ones and
> remote ones.
> The (ssh-like) URI idea seems to be good enough for it.
>
> cat [EMAIL PROTECTED]:/dev/microphone > [EMAIL PROTECTED]:/dev/speakers
>
> Ok... it's very naive, I admit; but it's just an example.
plan 9 can
i haven't been able to connect to 9fans.net for the past several days.
- erik
On Tue Jul 17 13:08:26 EDT 2007, [EMAIL PROTECTED] wrote:
> On 7/17/07, erik quanstrom <[EMAIL PROTECTED]> wrote:
>
> > why can't you just put the (known) offsets in your plan9.ini like i am doing
> > for aoe? it's distasteful, but it does get the job done.
> The boot cd does not help me -- cd is partitioned. I think the
> question still needs a good answer -- you have a cd and a disk image,
> you boot plan 9 without benefit of 9load, how do you get #S/sdD0/data
> set up? you need to write to ctl. What's the sensible way to do this?
>
> ron
why can'
hey, it wouldn't be too hard to port prep to linux, right?
- erik
silly ideas:
1. couldn't you just use a kfs in a fdisk/fat partition?
2. another approach would be to use russ' boot-to-rc script
trick.
3. prep the disk on a plan 9 machine.
i was fighting a similar problem this morning. in my case
the problem is that it is quite difficult to convince 9load
t
; ps -a | grep acid
quanstro 60520:00 0:00 616K Broken acid 6047
quanstro 60550:00 0:00 616K Broken acid 6047
quanstro 60720:00 0:00 156K Preadgrep acid
; ; /n/dump/2005/0220/386/bin/acid 6052
/proc/6052/text:386 plan 9 executable
/sys/lib/a
On Fri Jul 13 12:07:13 EDT 2007, [EMAIL PROTECTED] wrote:
> Does anyone know who to contact to talk about video cards that aren't
> functioning correctly with the vesa driver?
>
> Thanks
mail the list.
- erik
could you lay out a specific scenario where one would use
your proposed filesystem?
- erik
as geoff has pointed out, the flaw in this logic is that cs doesn't
make the connection, it just proposes a dialable address.
it turns out our problem was traced to someone reusing an inuse ip address.
- erik
On Sun Jul 8 10:50:21 EDT 2007, [EMAIL PROTECTED] wrote:
> in this case "net!somehost!
On Thu Jul 12 11:08:11 EDT 2007, [EMAIL PROTECTED] wrote:
> when i first saw that image a couple years ago i wondered
> what type of keyboard that was in front of Dorward. i think
> i recall seeing the same type in some other lab pictures too.
> does anybody know?
if it's not a unicomp model m p/n
are there any uses for this (anymore) on the pc architecture?
- erik
in this case "net!somehost!9fs" could be translated il!somehost!9fs
or tcp!somehost!9fs. the current algorithm is to look through the possibities
and see if there exists somehost on protocol $protocol for service 9fs.
the first match wins. this depends on the order in the table.
the current algor
it appears that net! is hueristic.
we use il here and this is the network table (from /sys/src/cmd/ndb/cs.c)
that works for us. i'm not sure why Nilfast was eliminated from the current
table, but the current ndb/cs doesn't work for us.
i would suspect that tcp is timing out in an antisocial way
r
On Sat Jul 7 21:19:33 EDT 2007, [EMAIL PROTECTED] wrote:
> Thinking about how some of the other platforms handle it, and even
> other OSes on x86, has anyone looked at trying to leverage a different
> bootloader? Something like grub, maybe?
>
> -Jack
this won't work.
remember the whole probl
i didn't intend to start that thread. especially since i don't have
time to do too much with 9load right now.
- erik
the problem is not the usb stick, but the fact that plan 9 does not recognize
the usb stick as a floppy disk as it is not using the bios calls. it is the
bios calls
that map requests to the floppy disk and map them to the usb.
thus 9load cannot find plan9.ini.
the easy (but oh-so-wrong) might b
most machines these days have 10x that much memory. it should
be speedy enough to use strstr(2) once you've loaded them into
memory. and even loading them into memory should take no
more than a few seconds at 80MB/s.
a more elegant solution would be to reduce each document to
a set of stemmed w
> One thing I like about the idea using 9p even in smaller systems is,
> that you can define one file to be something like a "serial terminal".
> If one has a lot of such embedded devices connected to a network, one
> can easily have a "command line" for each of them without the need to
> connect e
i'm not sure you can blame the ide/sata controller for this problem.
you're failing before cpuidprint() in main.c so interrupts haven't even
been enabled. you might try *nomp=1 in your plan9.ini file.
- erik
> So, maybe three boxes at, say, $750 total, maybe another $150 for a
> monitor, keyboard and mouse. We need some crappy 8-port switch for
> $50. Let's continue the dirt cheap theme and go a couple of 80 GB
> drives striped for the bulk of the filesystem and a 200 GB drive for
> the venti arenas
this is a gotcha with a standalone setup. it's not supposed to
be that easy to confuse the fs.
if you can at all swing a stand-alone fileserver, you'll really
be glad you did. ken's fileserver will run on old pIII and pII
hardware that current cpu kernels have trouble with. at home
i have a $15
> Hi all,
> what could be the cause of the following phenomenon:
> After every reboot my CPU server cannot authenticate
> anybody. auth/debug shows that the authsrv "does not
> agree" or has wrong tickets. After wrkey, changeuser,
> reboot, it works again.
>
> Best regards,
>
> Matthias
i'm not
ah, we're out of phase 180 degrees.
the particular problem i happened to catch the other day found
nserver*.apple.com unreachable starting from the root. i tried with
dig on linux, too. that didn't work either. but perhaps trying harder
would make sense.
i noticed that you aren't seeing negati
> Actually, I'm guessing that this is a webfs question.
>
> How do you open a local html file with abaco. If I type
> "file:/path/to/file", I get "unsupported url type".
>
> Thanks.
>
> Greg
you are correct.
you can manoever around this problem with a plumbing rule like
typeis text
d
post. write. think. what i ment was:
this crashes drawterm every time for me.
; cd /mnt/term/mnt
; ls -l factotum
- erik
this does it every time for me.
; cd /mnt/term
; ls -l factotum
- erik
> I would say that not understanding plan9 fileserver concept is my
> weakist point
these papers may be of some use:
/sys/doc/9.ps
/sys/doc/net/net.ps
/sys/doc/auth.ps
/sys/doc/prog4.ps
/sys/doc/acme/acme.ps
page is the file viewer. so "page /sys/doc/9.ps
i realize this is apolitical and offtopic. i apologize in advance.
geoff's improvements in dns are really quite nice. dns appears do a good
job in the face of well-behaved servers, but there are some ill behaved ones
for popular sites that give me occasional fits.
for example, www.apple.com dec
html.c:247,253 - /n/dump/2007/0702/usr/quanstro/src/abaco/html.c:247,252
{
Irule *i;
Point pt0, pt1;
- int j;
i = (Irule *)b->i;
pt0 = subpt(b->r.min, p->pos);
html.c:254,263 - /n/dump/2007/0702/usr/quanstro/src/abaco/html.c:253,259
pt0.y += Space;
i'll look at it this evening.
label is a Runestr, not a char*. what you want is
void
updlabel(void)
{
int fd, n;
fd = open("/dev/label", OWRITE);
if(fd == -1)
return;
fprint(fd, "%#.*S", title.nr, title.r);
close(fd);
}
- erik
why not make your linux swap partition temporarly into a dos partition.
then install plan 9 where you had the dos partition. then teach grub
to boot plan 9 and you can have your linux swap back.
- erik
he's copying the boot stuff to the hard drive because although bios can
access the usb cdrom, 9load cannot. it doesn't know how. this means
9load can't load a kernel unless he puts it someplace else.
- erik
> - here the installer asks about some file, so I satisfy this demand
> by giving !sdC0!dos!9pcflop.gz
i think you mean "sdC0!dos!9pcflop.gz".
> - So I hit ctrl-alt-del nad reboot to linux. Here I mount /dev/hda2
> as plan9 partition and copy plan9.iso.bz2 to it. Reboot again and
>
> Also another question, what is the exact filesize of the CD image. I
> mean the ISO file. I download it twice but got two different sizes of
> image. One of them is 267 MB (280,584,192 bytes). Is it correct? Or
> there may be some errors while downloading.
the downloadable image is rebuilt every
> But my point was that any protocol other than TCP or unix would
> seem to go unused, and break consistency with other URI types
> (none that I know of allow you to name a protocol). If there
> comes a time when it's a must, something (as suggested) like
> this:
i think you've invented a new
>
> Wouldn't a heuristic be enough here? if the host specified is the localhost
> then forst try to attach to a Unix domain socket, if that fails then try
> an IN connection to the localhost instead, the port number would become
> the socket's path.
this would lead to nondeterministic behavior.
>
> > but what is the point of encoding a dial string+path as a url?
>
> Integration in web-like environments, ie. mozilla ?
>
> I'm going to add an minimal 9p client into mozilla. So you can
> simply put an 9p url into the browser and have fun ;-)
mozilla, unlike emacs, is not an operating sys
> The scheme should be "9p", obviously. Where to put server name
> and port is also obvoius. So for the simple TCP case, we can
> use something like:
>
> 9p://localhost:9000/my/path
>
the format for a dial string is proto!server!port. e.g.
tcp!atlas.coraid.com!http.
going with your style t
abort()+0x0 /sys/src/libc/9sys/abort.c:6
_assert(s=0x324f9)+0x3a /sys/src/libc/port/_assert.c:12
destck(p=0xdfffe3b0)+0x29 /sys/src/cmd/ndb/dnresolve.c:278
xmitquery(qp=0xdfffe4ac,depth=0x1,medium=0x0,inns=0x0,obuf=0x62c46d0,len=0x34)+0xa6
/sys/src/cmd/ndb/dnresolve.c:910
p=0xdfffe0b0
On Fri Jun 29 15:27:25 EDT 2007, [EMAIL PROTECTED] wrote:
> I am thinking about buying a laptop to run Plan9 on and was looking at
> ThinkPad T30's. The wiki hardware compatibility list says that this
> model is supported by Plan9. Has anyone had any bad experiences with
> this laptop and Plan9 tha
you might mention these problems to the v9fs-developer list.
i believe it's on sourceforge.
- erik
> The Holy Code of the Sacred Developers contain the hard-coded Values that
> lead to True Enlightenment.
you say this with sarcasm. but i think there's some truth to it.
hard-coded Values are ones you don't need to think about.
my first reaction to russ' note was that vim hasn't so much provide
it sounds like you want γ correction to accommodate your hardware's
quirks, not a different color in sam.
- erik
that's pretty massive packet loss. tcp is likely to suffer mightily.
i used dialup until last year. the rtt time to the gatway was 150ms and
the provider was proud enough of their network to push the default mtu
to 1500 bytes. i never measured any packet loss.
perhaps you would have better lu
> Even with a local root, importing stuff over (say) wifi can be painful.
perhaps there's something about your setup that's different than
mine. even mounting root over a wireless connection is not a problem.
my desktop machine is connected a shared (with a dozen other computers)
wireless bridge.
rather than go to these extremes, why not use a local root?
run replica in a cron job if necessary. a paired-down root
can fit in 32MB more-or-less.
i use import so i can run acme locally, and cpu in acme windows to
access machines at work. this way only command stdin, stdout, stderr
and the fil
> 9P is just great for use when latency is reasonable (or not too bad,
> with cfs), but to go further
> away and still be comfortable using remote files, I'd say we need
> another protocol.
> I'd love to be proved wrong :)
are there any protocols that deal well with latency?
the only way i know t
> Although Plan9 file model, as nice as it is breaks horribly when
> we start talking about HD multimedia content. The performance bottleneck
> is quite noticeable at 50MBpbs :-(
50 MB/s == 1 gigabit
i think this will be a problem as soon as there exist gigabit link to actual
content providers.
we haven't deleted any log files since 2004 and yet the largest
is 1GB. all told, the log files eat < 1% of our total space and
every so often somebody wants to know something that only the
log files can answer.
it seems false economy. disk space is generally not a scarce resource.
- erik
no. that shouldn't do anything except shunt log information to /dev/null.
if you are using either venti or ken's fs, this won't save you any space as the
old log file is already in an arena (or in the dump).
- erik
what do you mean by "the fileserver part is no longer accessible?"
- erik
> Now, after a replica/pull update and a reboot, it seems that the
> fileserver part is no longer accessible, either. I installed Plan 9 on
> this machine several months ago using a CD that was rather older. Have
> there be
> > Steve Simon's post further down is significantly more interesting than the
> > normal /. 'lol u iz informanative' drivel.
>
> yet, the moderator gives it a score of 0 compared to a 1 for the
> drivel. why?
This isn't Plan 9. This is Slashdot. We don't need rules.
- erik
> ok, i'm still not sure what you're trying to do but thought i'd
> mention that bridge(3) is not something that implements an "Ethernet
> bridge" as in ANSI/IEEE 802.1d
i don't have the 802.1d standard. i can imagine the mss business isn't
standard and it seems a real bridge needs to participate
601 - 700 of 1584 matches
Mail list logo
