[9fans] Venti for Fossil

[9fans] Venti for Fossil

[lucio]

I am attempting to set up Fossil reliably, but I'm encountering some
stumbling blocks.

I have one incidental question, first of all: does not setting
/env/venti cause Fossil to omit the Venti interactions?

That aside, I have configured a single partition for Venti arenas:

	--rw-r----- S 0 proxima proxima 2146770944 Jul  3 10:57 sd01/v.arena

with

	--rw-r----- S 0 proxima proxima  209715200 Jul  3 10:57 sd00/v.idx0
	--rw-r----- S 0 proxima proxima  209715200 Jul  3 10:57 sd00/v.idx1

for the index.

The following commands:

	venti/fmtarenas arena.  /dev/sd01/v.arena
	venti/fmtisect  isect.0 /dev/sd00/v.idx0
	venti/fmtisect  isect.1 /dev/sd00/v.idx1

produce results along the following lines:

	clearing the partition
	configuring /dev/sd01/v.arena with arenas=3 for a total storage of bytes=2146435072 and directory bytes=65536
	adding arena arena.0 at [335872,537206784)
	adding arena arena.1 at [537206784,1074077696)
	adding arena arena.2 at [1074077696,1610948608)
	clearing the partition
	configuring index section isect.0 with space for index config bytes=65536
	clearing the partition
	configuring index section isect.1 with space for index config bytes=65536
	using 51118 buckets of 51118; div=84021

which I find a little worrying, index-size-wise.

Starting venti is surprising:

	venti/venti -c venti.conf -B 1m -C 1m -I 1m -w -a tcp!127.1!venti

which returns (note the different numbers):

	add arena arena.0 at [1048576,537903104)
	add arena arena.1 at [537903104,1074757632)
	add arena arena.2 at [1074757632,1611612160)
	configured index=main with arenas=3 and storage=1610563584

Whether this explains the problems I have encountered with archive
snapshots (lost blocks, but I also saw some I/O errors :-(, I would
prefer to ask more competent people.

I'm appending the script I use to set up Venti and Fossil from
scratch.  May it be useful to others as well as attract any
improvements (and land on the WIKI, if really useful).

(I have been playing a bit, trying to produce reproducible results,
but I have encountered too many difficulties, so the script that
follows may not be exactly what I used last.  It ought to be pretty
close, though.)

---- cut here ----
#!/bin/rc
cd /sys/lib/sysconfig/huddle/init
venti/fmtarenas arena.  /dev/sd01/v.arena
venti/fmtisect  isect.0 /dev/sd00/v.idx0
venti/fmtisect  isect.1 /dev/sd00/v.idx1
venti/fmtindex  venti.conf
venti/venti -c venti.conf -B 1m -C 1m -I 1m -w -a tcp!127.1!venti
venti=tcp!127.1!venti
echo -n '(venti)' Sleeping...; sleep 30; echo
echo y | fossil/flfmt /dev/sd01/fossil
echo -n '(flfmt)' Sleeping...; sleep 5; echo
fossil/fossil -c '. flauth'
echo -n '(fossil)' Sleeping...; sleep 5; echo
con -l /srv/fscons << EOF
create /active/adm adm adm d775
EOF
mount -c /srv/fossil /n/fossil
sed 's/^([^:]+):([^:]+)/\2:\2/' /adm/users > /n/fossil/adm/users
con -l /srv/fscons << EOF
users -r /active/adm/users
EOF
9fs kfs
disk/mkfs -a -s /n/kfs /sys/lib/sysconfig/proto/allproto | disk/mkext -u -d /n/fossil
con -l /srv/fscons << EOF
users -w
EOF
echo 'Done!'
---- cut here ----

flauth is, very simply:

---- cut here ----
srv -p fscons
srv fossil
fsys main config /dev/sd01/fossil
fsys main open -AWP
fsys main
---- cut here ----

venti.conf is:

---- cut here ----
index main
isect /dev/sd00/v.idx0
isect /dev/sd00/v.idx1
arenas /dev/sd01/v.arena
---- cut here ----

and the output, for reference:

- Omitted - Things are too far out of kilt! -

++L

Re: [9fans] Venti for Fossil

[okamoto]

I'm engaging this venti+fossil fighting now. ☺

> The following commands:
> 
> 	venti/fmtarenas arena.  /dev/sd01/v.arena
> 	venti/fmtisect  isect.0 /dev/sd00/v.idx0
> 	venti/fmtisect  isect.1 /dev/sd00/v.idx1
> 
> produce results along the following lines:
> 
> 	clearing the partition
> 	configuring /dev/sd01/v.arena with arenas=3 for a total storage of bytes=2146435072 and directory bytes=65536
> 	adding arena arena.0 at [335872,537206784)
> 	adding arena arena.1 at [537206784,1074077696)
> 	adding arena arena.2 at [1074077696,1610948608)
> 	clearing the partition
> 	configuring index section isect.0 with space for index config bytes=65536
> 	clearing the partition
> 	configuring index section isect.1 with space for index config bytes=65536
> 	using 51118 buckets of 51118; div=84021

Why the number of index sector is 2, but not 3.

I'm using whole of 120GB space to v.arenas, and it created 228 arenas like:

adding arean v.arenas0 at [335872, 537206784)
....
adding arena v.arenas227 at [121870032896, 122406903808)

What number of index sectors I have to make?

Kenji

Re: [9fans] Venti for Fossil

[andrey mirtchovski]

> I'm using whole of 120GB space to v.arenas, and it created 228 arenas like:
>
> adding arean v.arenas0 at [335872, 537206784)
> ....
> adding arena v.arenas227 at [121870032896, 122406903808)
>
> What number of index sectors I have to make?
>

you should have a sector for each arena, perhaps?

this is something I wouldn't mind being explained to me -- I'm using a 80GB
hard driver with just a single 60GB arena. which one is better: many small
arenas on the same disk, or a single, large one?

I did it with a single arena because I thought the next one should reside on
a separate disk...

or maybe I'm confused -- what venti considers an 'arena' is not the same as
an 'arena' partition on the hard drive, dedicated to venti:

index=main
total arenas=121 active=2
total space=64959397888 used=764195110
clumps=480677 compressed clumps=412542 data=1447303506 compressed
data=733912459


anyway,
I like these usage statistics -- I've thrown about 3gb of different data to
the venti (like 300+mb fonts duplicated in /lib and in /usr/andrey) and it
still grows very, very slowly...

Re: [9fans] Venti for Fossil

[okamoto]

>> I'm using whole of 120GB space to v.arenas, and it created 228 arenas like:
>>
>> adding arean v.arenas0 at [335872, 537206784)
>> ....
>> adding arena v.arenas227 at [121870032896, 122406903808)
>>
>> What number of index sectors I have to make?
>>
>
> you should have a sector for each arena, perhaps?

I'm now reformatting arena partition, because I got error messages
when I dispatched venti/fmtindex /sys/lib/sysconfig/dntf/venti.conf:

err2: arena set has wrong magic number: 0 expected a9e4a5e7
err2: corrupted arena partition header: arena set has wrong magic number:
	0 expected a9e4a5e7
fatal venti/fmtindex error:
	can't initialization config /sys/lib/sysconfig/dntf/venti.conf:
	corrupted arena partition header: arena set has wrong magic
	number: 0 expected a9e4a5e7.

I think magic number a9e4a5e7 is of ArenaPartMagic, which means I named
each arena wrong.   Then my next command line for fmtarenas is:
venti/fmtarenas arenas. /dev/sdC1/v.arenas
Of course, I switched dma on for the disk, this time, and made log
of the process, because I'll have so many arenas.   ?

Kenji

Re: [9fans] Venti for Fossil

[andrey mirtchovski]

On Wed, 30 Jul 2003 okamoto@granite.cias.osakafu-u.ac.jp wrote:

> I think magic number a9e4a5e7 is of ArenaPartMagic, which means I named
> each arena wrong.   Then my next command line for fmtarenas is:
> venti/fmtarenas arenas. /dev/sdC1/v.arenas
> Of course, I switched dma on for the disk, this time, and made log
> of the process, because I'll have so many arenas.   ?

I try to stay as close to the instructions as possible -- simple is better.
My venti.conf has the following:

% cat venti.conf
index main
isect /dev/sdC0/isect
arenas /dev/sdC0/arenas
%

my partitions on the disk are:

  9fat                   0 20482       (20482 sectors, 10.00 MB)
  swap               20482 1273392     (1252910 sectors, 611.77 MB)
  fossil           1273392 25599169    (24325777 sectors, 11.59 GB)
  arenas          25599169 153492606   (127893437 sectors, 60.98 GB)
  isect          153492606 160071596   (6578990 sectors, 3.13 GB)
  nvram          160071596 160071597   (1 sectors, 512 B )

with this setup I haven't had any problems with venti.

andrey

[9fans] VMWare related question..

[Ishwar Rattan]

I have VMWare-4.0 workstation installed under Win2K.

I downloaded the file vmware.zip from Plan 9 download page.

How do I use it under the simulator?

-ishwar

[9fans] generating ssh keys for a user

[Ishwar Rattan]

How can the public and private keys can be created
for a user (eg Glenda) using aux/ssh_genkey? It
complains as permission denied..

-ishwar

Re: [9fans] generating ssh keys for a user

[David Presotto]

ssh_genkey is obsolete.  man 8 rsa.  Look at the examples.

[9fans] rsa key again..

[Ishwar Rattan]

I did this (per man page)

  auth/rsagen -t 'service=ssh' >key
  auth/rsa2ssh key | ssh unix 'cat >>.ssh/authorized_keys'
  cat key > /mnt/factotum/ctl

last line produces error message

  cat: write error copying key: unknown proto rsa

-ishwar

Re: [9fans] rsa key again..

[David Presotto]

[-- Attachment #1: Type: text/plain, Size: 37 bytes --]

Sounds like you have an old factotum.

[-- Attachment #2: Type: message/rfc822, Size: 2430 bytes --]

From: Ishwar Rattan <ishwar@pali.cps.cmich.edu>
To: 9fans@cse.psu.edu
Subject: [9fans] rsa key again..
Date: Wed, 6 Aug 2003 09:17:22 -0400 (EDT)
Message-ID: <Pine.LNX.4.56.0308060911170.14638@pali.cps.cmich.edu>


I did this (per man page)

  auth/rsagen -t 'service=ssh' >key
  auth/rsa2ssh key | ssh unix 'cat >>.ssh/authorized_keys'
  cat key > /mnt/factotum/ctl

last line produces error message

  cat: write error copying key: unknown proto rsa

-ishwar

Re: [9fans] Venti for Fossil

[okamoto]

> I'm now reformatting arena partition,
>Then my next command line for fmtarenas is:
> venti/fmtarenas arenas. /dev/sdC1/v.arenas
> Of course, I switched dma on for the disk, this time, and made log
> of the process, because I'll have so many arenas.   ?

This time, the time for formatting /dev/sdC1/v.arena of 120GB was
reduced to 1 hour, and that for index was to 4 min!   This time, I have
no error message from venti/fmtindex command.
Ok, next is making fossil.

Kenji

Re: [9fans] Venti for Fossil

[Lucio De Re]

On Wed, Jul 30, 2003 at 01:14:57PM +0900, okamoto@granite.cias.osakafu-u.ac.jp wrote:
>
> I'm engaging this venti+fossil fighting now. ?
>
I've added a "-n" option to venti/fmtarenas so that I can specify the
number of arenas instead of the size.  It's a small fix and I believe
it to be useful at least for experimentation.  Should I submit it
here?  Can anyone think of a reason _not_ to add it?

I couldn't find a good way of extending this to pick a different:

		tabSize = 64 * 1024;
			/* BUG: should be determine from number of arenas */

(sic) as the relationship isn't obvious (to me).  If anyone can
suggest how to do it, I'll be happy to include it in my patch.

++L

PS: venti/syncindex fails (for me) as soon as there is something in
Venti.  Is this expected?

Re: [9fans] Venti for Fossil

[okamoto]

When I want to close venti or fossil server, I just kill those
processes?   I read venti/sync command, and I think I got it
right.

Kenji

Re: [9fans] Venti for Fossil

[jmk]

On Wed Jul 30 23:48:31 EDT 2003, okamoto@granite.cias.osakafu-u.ac.jp wrote:
> When I want to close venti or fossil server, I just kill those
> processes?   I read venti/sync command, and I think I got it
> right.
>
> Kenji

There's a fossil 'sync' command too.

Re: [9fans] Venti for Fossil

[okamoto]

> On Wed Jul 30 23:48:31 EDT 2003, okamoto@granite.cias.osakafu-u.ac.jp wrote:
>> When I want to close venti or fossil server, I just kill those
>> processes?   I read venti/sync command, and I think I got it
>> right.
>> 
>> Kenji
> 
> There's a fossil 'sync' command too.

Thanks, I also got a reply from Andrey to see fossilcons(8). ☺
I'm now copying whole of our file server to dunite fossil+venti,
which have two disks /dev/sdC0: 40GB, /dev/sdC1: 120GB).
Fossil file system is on 40GB disk, and whole of arenas are 
on 120GB disk, index section of the venti is on 40GB disk, too.
It seems to be working...

Some interesting commands seem to be added, for example
unhalt!   Halt may have different meaning from that of the old
file server.  What is the reason of those difference?

Kenji

Re: [9fans] Venti for Fossil

[okamoto]

> Some interesting commands seem to be added, for example
> unhalt!   Halt may have different meaning from that of the old
> file server.  What is the reason of those difference?

Aha!
probably, this asumes kfs is running for the fossil file server.
I thought why it is neccessary when fossil file server is stopped
where that fossil server is the only one file server in the Plan 9
network...

Kenji

Re: [9fans] Venti for Fossil

[okamoto]

One more question.

If I have a filesystem which I don't want to write to venti, such as
other in the old file server, I think I should not dispatch the command of

fsys other config /dev/sdC0/other

,instead, I should just do

fsys other open

without config command to this file system.

However, if I have no config command, how the system knows where is
other file system?

kenji

Re: [9fans] Venti for Fossil

[Fco.J.Ballesteros]

[-- Attachment #1: Type: text/plain, Size: 325 bytes --]

This is what we use. main is archived to venti; once is not.

srv -p fscons
srv -p fscmd
msg -m 100 -p 15
srv boot
fsys main config /dev/fs/fossil
fsys main open -c 20000
fsys main
snaptime -s 60 -a 0500 -t 6048
listen il!*!17008
listen tcp!*!564
fsys once config /dev/fs/once
fsys once open -c 2000
fsys main

[-- Attachment #2: Type: message/rfc822, Size: 2098 bytes --]

From: okamoto@granite.cias.osakafu-u.ac.jp
To: 9fans@cse.psu.edu
Subject: Re: [9fans] Venti for Fossil
Date: Thu, 31 Jul 2003 18:11:09 +0900
Message-ID: <2b5d0d09c2c3151feedf42f54271b1d0@granite.cias.osakafu-u.ac.jp>

One more question.

If I have a filesystem which I don't want to write to venti, such as
other in the old file server, I think I should not dispatch the command of

fsys other config /dev/sdC0/other

,instead, I should just do

fsys other open

without config command to this file system.

However, if I have no config command, how the system knows where is
other file system?

kenji

Re: [9fans] Venti for Fossil

[okamoto]

Thanks nemo!

> This is what we use. main is archived to venti; once is not.
>
> srv -p fscons
> srv -p fscmd
> msg -m 100 -p 15
> srv boot
> fsys main config /dev/fs/fossil
> fsys main open -c 20000
> fsys main
> snaptime -s 60 -a 0500 -t 6048
> listen il!*!17008
> listen tcp!*!564
> fsys once config /dev/fs/once
> fsys once open -c 2000
> fsys main

I'm afraid, in this case, the once file system will be cleaned up by
the line of
> fsys once config /dev/fs/once
?

Kenji

Re: [9fans] Venti for Fossil

[Fco.J.Ballesteros]

> I'm afraid, in this case, the once file system will be cleaned up by
> the line of

What do you mean by `clean up' in this  context?
It for sure retains its previous contents...

Re: [9fans] Venti for Fossil

[okamoto]

>> I'm afraid, in this case, the once file system will be cleaned up by
>> the line of
>
> What do you mean by `clean up' in this  context?
> It for sure retains its previous contents...

I was afraid "it for sure _not_ retained its previous contents".

Kenji

Re: [9fans] Venti for Fossil

[jmk]

On Thu Jul 31 05:22:34 EDT 2003, nemo@plan9.escet.urjc.es wrote:
> This is what we use. main is archived to venti; once is not.
>
> srv -p fscons
> srv -p fscmd
> msg -m 100 -p 15
> srv boot
> fsys main config /dev/fs/fossil
> fsys main open -c 20000
> fsys main
> snaptime -s 60 -a 0500 -t 6048
> listen il!*!17008
> listen tcp!*!564
> fsys once config /dev/fs/once
> fsys once open -c 2000
> fsys main

There is also an undocumented '-V' flag to 'fsys name open' which
someone has added in an attempt to do an open without making a connection
to a venti server. However, a quick look at the code leads me to believe
that it doesn't quite do the trick and may cause null references if used
in conjunction with some other console commands. But I could be wrong.

This shows the rather fragmented development history of fossil and it's
fair to say that a lot of the fossil internals and the venti library code
it uses are not loved very much.

Re: [9fans] Venti for Fossil

[rog]

> [...] a lot of the fossil internals and the venti library code
> it uses are not loved very much.

i blame all that mixed case...

Re: [9fans] Venti for Fossil

[okamoto]

>> [...] a lot of the fossil internals and the venti library code
>> it uses are not loved very much.
>
> i blame all that mixed case...

I lean to Jim's stand, because file server is the heart of
Plan 9 system, and we expect most robustness to it,
most deeply thinked, and most finest codes it comprises.
All those beyond my ability, and I expect someone can
do it for us.    What I can give them?   Only my respects to them.
I'm sory for all...

Anyway, now I know the present status of venti+fossil,
and I'll touch it with most careful attention from now.
I'll continue to try to live with this venti+fossil and old robust
and loved file server in our Plan 9 network.

Kenji

Re: [9fans] Venti for Fossil

[Dan Cross]

> Anyway, now I know the present status of venti+fossil,
> and I'll touch it with most careful attention from now.
> I'll continue to try to live with this venti+fossil and old robust
> and loved file server in our Plan 9 network.

Yeah, I haven't switched over yet either because it seems like venti
and fossil are still getting some of the kinks worked out, and I can
live with the limitations of Ken's fileserver.  Once things stabilize
a bit more, I'll switch.

	- Dan C.

Re: [9fans] Venti for Fossil

[okamoto]

I'm confusing...

I have disks as:
--rw-r----- S 0 okamoto okamoto    10453504 Jul 18 17:15 /dev/sdC0/9fs
--rw-r----- S 0 okamoto okamoto           0 Jul 18 17:15 /dev/sdC0/ctl
--rw-r----- S 0 okamoto okamoto 40020664320 Jul 18 17:15 /dev/sdC0/data
--rw-r----- S 0 okamoto okamoto  3498247168 Jul 18 17:15 /dev/sdC0/fossil
--rw-r----- S 0 okamoto okamoto 10737417216 Jul 18 17:15 /dev/sdC0/other
--rw-r----- S 0 okamoto okamoto 40015954944 Jul 18 17:15 /dev/sdC0/plan9
-lrw------- S 0 okamoto okamoto           0 Jul 18 17:15 /dev/sdC0/raw
--rw-r----- S 0 okamoto okamoto  6442450944 Jul 18 17:15 /dev/sdC0/v.index0
--rw-r----- S 0 okamoto okamoto  6442450944 Jul 18 17:15 /dev/sdC0/v.index1
--rw-r----- S 0 okamoto okamoto  6442450944 Jul 18 17:15 /dev/sdC0/v.index2
--rw-r----- S 0 okamoto okamoto  6442451968 Jul 18 17:15 /dev/sdC0/v.index3
--rw-r----- S 0 okamoto okamoto            0 Jul 18 17:15 /dev/sdC1/ctl
--rw-r----- S 0 okamoto okamoto 122942324736 Jul 18 17:15 /dev/sdC1/data
--rw-r----- S 0 okamoto okamoto 122935002624 Jul 18 17:15 /dev/sdC1/plan9
-lrw------- S 0 okamoto okamoto            0 Jul 18 17:15 /dev/sdC1/raw
--rw-r----- S 0 okamoto okamoto 122934970368 Jul 18 17:15 /dev/sdC1/v.arenas

I run floppy bootstrapped fossil+venti, which seems to be fine.

I want to mount the other file system to /n/duniteother directory, and then,
my /sys/lib/dntf/dntfproto file is:  (Thanks nemo!)

srv -p fscons
srv -p fscmd
srv boot
fsys main config /dev/sdC0/fossil
# 36864 = (312MB - 8MB*3)*1024*1024/8192
# 29696 = (256MB - 8MB*3)*1024*1024/8192
fsys main open -c 26000
fsys main
users -r /active/adm/users
snaptime -s 60 -a 0500
listen il!*!17008
listen tcp!*!564
fsys other config /dev/sdC0/other
fsys other open -c 2000
fsys main

and I dispatched the command:

mount -c /srv/boot /n/duniteother other
and then,
ls -l /n/duniteother.

I got
d-rwxrwxrwx M 31 adm adm 0 Jul 31 17:04 /n/duniteother/active
d-r-xr-xr-x M 31 adm adm 0 Jul 31 17:04 /n/duniteother/archive
d-r-xr-xr-x M 31 adm adm 0 Jul 31 17:04 /n/duniteother/snapshot
I want just it is done like that for /n/boot using /srv/boot.

In this case, I don't use venti for this 'other' file system, and I don't want
to have subdirectories of active, archive and snapshot.
By the way, am I safe to use /n/dioriteother/active for such a usual
/n/duniteother?  This means can I just replace /n/dioriteother by
/n/dioriteother/active.

Kenji

Re: [9fans] Venti for Fossil

[okamoto]

It's interesting to touch fossil+venti in our Plan 9
network!   Just for applause to the developpers.
Sense of terminal's easiness and functional file server!

Kenji

Re: [9fans] Venti for Fossil

[okamoto]

As I have accumulated some of /n/dump, I'd like to add the
older /n/dump portions from the old 4ed file server.

Is is safe just to copy all the /n/dump/xxxx/xxxx to those
corresponding new fossil+venti file server?

Kenji

Re: [9fans] Venti for Fossil

[okamoto]

> Is is safe just to copy all the /n/dump/xxxx/xxxx to those
> corresponding new fossil+venti file server?

No, there are complete file trees at individual /n/dump/xxxx/xxxx.
How I can copy all the older 'dump' files to the venti+fossil file server?

Kenji

Re: [9fans] Venti for Fossil

[okamoto]

>> Is is safe just to copy all the /n/dump/xxxx/xxxx to those
>> corresponding new fossil+venti file server?
>
> No, there are complete file trees at individual /n/dump/xxxx/xxxx.
> How I can copy all the older 'dump' files to the venti+fossil file server?

I found nemo's message in this list:

>I copied all the dumps in our worm fs to venti. Our 9fs script uses
>vacfs to mount pre-fossil dumps and binds them along with fossil archive.

How did you copy those?

Kenji

Re: [9fans] Venti for Fossil

[okamoto]

> I found nemo's message in this list:
>
>>I copied all the dumps in our worm fs to venti. Our 9fs script uses
>>vacfs to mount pre-fossil dumps and binds them along with fossil archive.
>
> How did you copy those?

I also found nemo's post around January of this year, which solves my
problem.    Thanks Jim!
Nemo is, probably, now diving somewhere on this planet.

Huuuh, HOT! too HOT!

Kenji

Re: [9fans] Venti for Fossil

[jmk]

On Wed Aug  6 01:50:25 EDT 2003, okamoto@granite.cias.osakafu-u.ac.jp wrote:
> ...
> Huuuh, HOT! too HOT!
>
> Kenji

Here it is too wet. I think you will be maybe getting a typhoon soon
to make it cooler.

--jim

Re: [9fans] Venti for Fossil

[boyd, rounin]

rain, i need RAIN!!

Re: [9fans] Venti for Fossil

[andrey mirtchovski]

On Wed, 6 Aug 2003 jmk@plan9.bell-labs.com wrote:

> Here it is too wet. I think you will be maybe getting a typhoon soon
> to make it cooler.
>

nothing to complain about here -- it's 25 degrees with chances of snow
by the end of the month :P

Re: [9fans] Venti for Fossil

[okamoto]

> On Wed Aug  6 01:50:25 EDT 2003, okamoto@granite.cias.osakafu-u.ac.jp wrote:
> Here it is too wet. I think you will be maybe getting a typhoon soon
> to make it cooler.

Yeah, I'm afraid of it, because we'll have a meeting of this weekend
with audio freaks who love to make tube amplifier by themselves. ☺

Kenji

Re: [9fans] Venti for Fossil

[Lucio De Re]

On Thu, Jul 31, 2003 at 12:17:44PM -0400, jmk@plan9.bell-labs.com wrote:
>
> This shows the rather fragmented development history of fossil and it's
> fair to say that a lot of the fossil internals and the venti library code
> it uses are not loved very much.

I would make time to help redevelop Fossil and even Venti if someone
would like to take a lead.  Maybe we could start by integrating
Nemo's fixes into Fossil (I had a superficial look at them, and
they look good, but divergent) and we may even redevelop in a less
platform-dependent fashion, where possible.

I was thinking that Fossil makes an excellent foundation for more
than a Plan 9 file service; for example, my pet project is workflow
management and I was thinking of using Fossil's "independent"
authentication and access control to manage the ownership of workflow
transactions.  Not that I have given it much thought, unfortunately.

++L

Re: [9fans] Venti for Fossil

[okamoto]

When I want to close venti or fossil server, I just kill those
processes?   I read venti/sync command, and I think I got it
right.

Kenji

Re: [9fans] Venti for Fossil

[steve.simon]

>
>Huuuh, HOT! too HOT!
>
>Kenji
>

I doubt its any consolation but its
way too hot here too (UK)

-Steve

Re: [9fans] Venti for Fossil

[boyd, rounin]

way too hot/humid here too.

Re: [9fans] Venti for Fossil

[C H Forsyth]

[-- Attachment #1: Type: text/plain, Size: 108 bytes --]

i quite like it.
good weather for G&T, Pimms, or real ale.
water too, if you won't take WC Fields' advice.

[-- Attachment #2: Type: message/rfc822, Size: 1919 bytes --]

From: steve.simon@snellwilcox.com
To: 9fans@cse.psu.edu
Subject: Re: [9fans] Venti for Fossil
Date: Wed, 6 Aug 2003 09:40:43 +0100
Message-ID: <3d58a73ac2f9921c3bd9ffec81627beb@yourdomain.dom>

>
>Huuuh, HOT! too HOT!
>
>Kenji
>

I doubt its any consolation but its
way too hot here too (UK)

-Steve

Re: [9fans] Venti for Fossil

[Derek Fawcus]

On Wed, Aug 06, 2003 at 09:40:43AM +0100, steve.simon@snellwilcox.com wrote:
> >
> >Huuuh, HOT! too HOT!
> >
> >Kenji
> >
>
> I doubt its any consolation but its
> way too hot here too (UK)

Yeah - got burned over the weekend.  However that doesn't change that the
sea's around our little island are genrally 'Too COLD,  too COLD!',  hmm
where's that dry suit...

DF

Re: [9fans] Venti for Fossil

[David Presotto]

Ka pupuhi te hau ka maringi te wai

Re: [9fans] Venti for Fossil

[okamoto]

> Ka pupuhi te hau ka maringi te wai
 I guess it Hawaiian, doesn't it?

When I'm in Hawaii, I always feel much puzzled by their names
which has many voels similar to Japanese.   The fact itself makes
me more confused, because I can't image the names from somewhat
similar pronouciations.  Paradox, isn't it?
Hot climate probably makes me mad, sorry eveyone of off-topic.

kenji

Re: [9fans] Venti for Fossil

[andrey mirtchovski]

On Thu, 7 Aug 2003 okamoto@granite.cias.osakafu-u.ac.jp wrote:

> Hot climate probably makes me mad, sorry eveyone of off-topic.
>

I guess you can call it "Hot Topic", har har :)

Re: [9fans] Venti for Fossil

[boyd, rounin]

>  I guess it Hawaiian, doesn't it?

yes, the maoris are polynesian.

Re: [9fans] Venti for Fossil

[Skip Tavakkolian]

> Yeah - got burned over the weekend.  However that doesn't change that the
> sea's around our little island are genrally 'Too COLD,  too COLD!',  hmm
> where's that dry suit...

During the last two weeks it's been hot enough here that I've been
able to swim in the Puget Sound wearing a cheap wet suite.

Maybe there is something to this global warming claim?!!