[9fans] sources down

[9fans] sources down

[geoff]

The venti arena on sources filled last night and we
noticed it this morning.  I've added another disk for
the arenas and am working to get the sources fossil
reinitialised from the dump of Nov 2nd.

I think things are almost restored, but I thought that
several times earlier today too.

Re: [9fans] sources down

[geoff]

sources is still down and probably will be until Saturday afternoon or
evening.  The venti store seems to be intact and the fossil that
sources exports seems to be healthy, but the standalone fossil that
sources boots from and does not export is corrupted.  I've been trying
to patch things up but the corruption is winning.  In the worst case,
we may have to reinitialise and reload that fossil.

Re: [9fans] sources down

[geoff]

sources is back up and appears to be working normally for users on the
outside.  It's been a learning experience and getting sources repaired
shouldn't take so long in the future.

Re: [9fans] sources down

[erik quanstrom]

thanks, geoff!

- erik

[9fans] Sources down

[Rodolfo kix Garcia]

Hi!

sources seems to be down:


  Object not found

The object / does not exist on this server.

errstr: '/bin/ip/httpd/sources' does not exist
uri host:
header host:
actual host: plan9.bell-labs.com



Saludos,

kix.

Re: [9fans] Sources down

[Richard Miller]

> sources seems to be down:

I don't think so.

term% 9fs sources
post...
term% lc /n/sources
9grid	contrib	du		fastos	patch	wiki
adm		dist	extra	lsr		plan9	xen
term% date
Sun Nov 30 12:38:11 GMT 2008
term%

Re: [9fans] Sources down

[Roman Zhukov]

web interface is actually down

--
Roma

Re: [9fans] Sources down

[Charles Forsyth]

[-- Attachment #1: Type: text/plain, Size: 46 bytes --]

it works for me. which page are you accessing?

[-- Attachment #2: Type: message/rfc822, Size: 2068 bytes --]

From: "Roman Zhukov" <rzhukov@gmail.com>
To: "Fans of the OS Plan 9 from Bell Labs" <9fans@9fans.net>
Subject: Re: [9fans] Sources down
Date: Mon, 1 Dec 2008 00:09:33 +0500
Message-ID: <881467ce0811301109p5b790e67tfe01bae12eef08e2@mail.gmail.com>

web interface is actually down

--
Roma

Re: [9fans] Sources down

[geoff]

Browsing the source via http is disabled again, probably temporarily.
I'm looking into what's consuming (or limiting) the bandwidth on our
internet connection and into improving the performance of our outside
web server.

Re: [9fans] Sources down

[Francisco J Ballesteros]

We are maintaining at sources.lsub.org
a mirror for sources' plan9 and google code inferno.

9fs sources.lsub.org should work from plan9.

hth

On Sun, Nov 30, 2008 at 9:34 PM,  <geoff@plan9.bell-labs.com> wrote:
> Browsing the source via http is disabled again, probably temporarily.
> I'm looking into what's consuming (or limiting) the bandwidth on our
> internet connection and into improving the performance of our outside
> web server.
>
>
>

[9fans] sources down

[hiro]

could somebody press the button at bell-labs?
thank you.

Re: [9fans] sources down

[Steve Simon]

i havent seen sources working for 6 months at least. there are mirrors however:

e.g. at http://plan9.io

-Steve


> On 28 Dec 2018, at 7:07 pm, hiro <23hiro@gmail.com> wrote:
> 
> could somebody press the button at bell-labs?
> thank you.

Re: [9fans] sources down

[Skip Tavakkolian]

[-- Attachment #1: Type: text/plain, Size: 654 bytes --]

I've not had any success for a while with bell-labs sources. I've been
using 9p.io.
https://9legacy.org is also a great resource.

while I'm thinking about it, is it possible for 9legacy.org to use
letsencrypt?   Go's acme/autocert package (golang.org/x/crypto/acme/autocert)
works great.

On Fri, Dec 28, 2018 at 11:30 AM Steve Simon <steve@quintile.net> wrote:

> i havent seen sources working for 6 months at least. there are mirrors
> however:
>
> e.g. at http://plan9.io
>
> -Steve
>
>
> > On 28 Dec 2018, at 7:07 pm, hiro <23hiro@gmail.com> wrote:
> >
> > could somebody press the button at bell-labs?
> > thank you.
>
>
>

[-- Attachment #2: Type: text/html, Size: 1266 bytes --]

Re: [9fans] sources down

[Ethan Gardener]

On Fri, Dec 28, 2018, at 7:43 PM, Skip Tavakkolian wrote:
> while I'm thinking about it, is it possible for 9legacy.org to use letsencrypt?   Go's acme/autocert package (golang.org/x/crypto/acme/autocert) works great.

"This package is a work in progress and makes no API stability promises."

Re: [9fans] sources down

[hiro]

the reason i ask is because i was trying to go through the sources
dump. sadly that isn't available at all now.

also, russ used to have this very nice browseable kernel history. that
is also gone :(

Re: [9fans] sources down

[David du Colombier]

> the reason i ask is because i was trying to go through the sources
> dump. sadly that isn't available at all now.

The history of /n/sources/plan9 is available as a Git repository here:

https://github.com/0intro/plan9

However, this doesn't include the history of the rest of /n/sources.
I have a full copy of /n/sourcesdump and I'd like to make it
available on 9p.io, but I didn't have time to do it so far.

> also, russ used to have this very nice browseable kernel history. that
> is also gone :(

The Plan 9 kernel history is available here: https://9p.io/sources/extra/9hist/
This includes the code, but unfortunately not the web service nor the comments,
which were stored on Russ' side (on swtch.com).

--
David du Colombier

Re: [9fans] sources down

[Skip Tavakkolian]

[-- Attachment #1: Type: text/plain, Size: 518 bytes --]

It's all relative. For example, compared to the current President of the
United States, this API is extremely stable :)

On Fri, Dec 28, 2018, 11:58 AM Ethan Gardener <eekee57@fastmail.fm wrote:

> On Fri, Dec 28, 2018, at 7:43 PM, Skip Tavakkolian wrote:
> > while I'm thinking about it, is it possible for 9legacy.org to use
> letsencrypt?   Go's acme/autocert package (
> golang.org/x/crypto/acme/autocert) works great.
>
> "This package is a work in progress and makes no API stability promises."
>
>

[-- Attachment #2: Type: text/html, Size: 972 bytes --]

Re: [9fans] sources down

[Skip Tavakkolian]

[-- Attachment #1: Type: text/plain, Size: 893 bytes --]

In all seriousness, it's not too difficult to integrate in. I used Russ'
letsencrypt package before this one. Both have  been straight forward.
There are samples too.

With Go modules and vendoring, package version can be locked in
(notwithstanding security fixes)

On Fri, Dec 28, 2018, 1:15 PM Skip Tavakkolian <skip.tavakkolian@gmail.com
wrote:

> It's all relative. For example, compared to the current President of the
> United States, this API is extremely stable :)
>
> On Fri, Dec 28, 2018, 11:58 AM Ethan Gardener <eekee57@fastmail.fm wrote:
>
>> On Fri, Dec 28, 2018, at 7:43 PM, Skip Tavakkolian wrote:
>> > while I'm thinking about it, is it possible for 9legacy.org to use
>> letsencrypt?   Go's acme/autocert package (
>> golang.org/x/crypto/acme/autocert) works great.
>>
>> "This package is a work in progress and makes no API stability promises."
>>
>>

[-- Attachment #2: Type: text/html, Size: 1679 bytes --]

Re: [9fans] sources down

[Steve Simon]

[-- Attachment #1: Type: text/plain, Size: 1190 bytes --]

i have a copy if the diffs russ’s web pages used, though i dont think the annotations are included. it lived on russ’s server i think, not the labs one.

maybe its on the internet archive?

-Steve


> On 28 Dec 2018, at 9:23 pm, Skip Tavakkolian <skip.tavakkolian@gmail.com> wrote:
> 
> In all seriousness, it's not too difficult to integrate in. I used Russ' letsencrypt package before this one. Both have  been straight forward. There are samples too. 
> 
> With Go modules and vendoring, package version can be locked in (notwithstanding security fixes)
> 
>> On Fri, Dec 28, 2018, 1:15 PM Skip Tavakkolian <skip.tavakkolian@gmail.com wrote:
>> It's all relative. For example, compared to the current President of the United States, this API is extremely stable :)
>> 
>>> On Fri, Dec 28, 2018, 11:58 AM Ethan Gardener <eekee57@fastmail.fm wrote:
>>> On Fri, Dec 28, 2018, at 7:43 PM, Skip Tavakkolian wrote:
>>> > while I'm thinking about it, is it possible for 9legacy.org to use letsencrypt?   Go's acme/autocert package (golang.org/x/crypto/acme/autocert) works great.
>>> 
>>> "This package is a work in progress and makes no API stability promises."
>>> 

[-- Attachment #2: Type: text/html, Size: 2366 bytes --]

Re: [9fans] sources down

[hiro]

after checking the git mirror, it turns out the change i'm interested
in must have happened before it's beginning at 2002.

On 12/28/18, Steve Simon <steve@quintile.net> wrote:
> i have a copy if the diffs russ’s web pages used, though i dont think the
> annotations are included. it lived on russ’s server i think, not the labs
> one.

i heard from Fish also that this is available, but I'm still waiting
in case somebody might still pop up with the annotations...

Re: [9fans] sources down

[Steve Simon]

i have wondered if it would be possible to apply the historic plan9 kernel diffs and regenerate these ancient kernels.

no real use but it would be interesting to see the code in its context.

> On 29 Dec 2018, at 10:05 am, hiro <23hiro@gmail.com> wrote:
> 
> after checking the git mirror, it turns out the change i'm interested
> in must have happened before it's beginning at 2002.
> 
>> On 12/28/18, Steve Simon <steve@quintile.net> wrote:
>> i have a copy if the diffs russ’s web pages used, though i dont think the
>> annotations are included. it lived on russ’s server i think, not the labs
>> one.
> 
> i heard from Fish also that this is available, but I'm still waiting
> in case somebody might still pop up with the annotations...

Re: [9fans] sources down

[David du Colombier]

> i have wondered if it would be possible to apply the historic plan9
> kernel diffs and regenerate these ancient kernels.

The diff files are ed scripts generated with "diff -e", so it should be
possible to regenerate the original files with a bit of scripting.

--
David du Colombier

Re: [9fans] sources down

[hiro]

Does somebody know if Russ still reads this list?
I would be very interested if he still has the annotations...

Re: [9fans] sources down

[David du Colombier]

I've just converted the 9hist diff files to a Git repository:

https://github.com/0intro/9hist

--
David du Colombier

Re: [9fans] sources down

[Steve Simon]

do you have a script that you used to generate the 9hist repository?
I always planned to ingest it into my venti.

I have already put a few old floppy and cdrom from my past in
venti as limited /yyyy/mmdd/usr/steve/... so the 9hist could go in
as /yyyy/mmdd/sys/src/9/...

Thanks,

-Steve

Re: [9fans] sources down

[David du Colombier]

> do you have a script that you used to generate the 9hist repository?
> I always planned to ingest it into my venti.
>
> I have already put a few old floppy and cdrom from my past in
> venti as limited /yyyy/mmdd/usr/steve/... so the 9hist could go in
> as /yyyy/mmdd/sys/src/9/...

Yes, but beware it's a throw-away script. I'll send it to you privately.

--
David du Colombier

Re: [9fans] sources down

[Devon H. O'Dell]

[-- Attachment #1: Type: text/plain, Size: 604 bytes --]

This “send it to you privately” ethos is a problem on this list. Why not
make it public?

—dho

On Sun, Dec 30, 2018 at 2:55 PM David du Colombier <0intro@gmail.com> wrote:

> > do you have a script that you used to generate the 9hist repository?
> > I always planned to ingest it into my venti.
> >
> > I have already put a few old floppy and cdrom from my past in
> > venti as limited /yyyy/mmdd/usr/steve/... so the 9hist could go in
> > as /yyyy/mmdd/sys/src/9/...
>
> Yes, but beware it's a throw-away script. I'll send it to you privately.
>
> --
> David du Colombier
>
>

[-- Attachment #2: Type: text/html, Size: 956 bytes --]

Re: [9fans] sources down

[Lucio De Re]

I guess it's a choice we need to be free to make. Sometimes the effort
involved in making something sufficiently polished for publication is
too much, when one is too self-conscious to dump a dog's breakfast for
the Internet to vomit over forever (no reflection over David's
coding).

There isn't a good place to draw that line, Devon, let's just
encourage posting, maybe offer to help tidying up. If that doesn't
work, then one must just accept the choice made by the originator.

Lucio.

PS: To push the point a bit: forty odd years ago the anecdote I heard
involved an APL one-liner that erased itself on completion. I think
that is "justified".



On 12/31/18, Devon H. O'Dell <devon.odell@gmail.com> wrote:
> This “send it to you privately” ethos is a problem on this list. Why not
> make it public?
>
> —dho
>
> On Sun, Dec 30, 2018 at 2:55 PM David du Colombier <0intro@gmail.com>
> wrote:
>
>> > do you have a script that you used to generate the 9hist repository?
>> > I always planned to ingest it into my venti.
>> >
>> > I have already put a few old floppy and cdrom from my past in
>> > venti as limited /yyyy/mmdd/usr/steve/... so the 9hist could go in
>> > as /yyyy/mmdd/sys/src/9/...
>>
>> Yes, but beware it's a throw-away script. I'll send it to you privately.
>>
>> --
>> David du Colombier
>>
>>
>


-- 
Lucio De Re
2 Piet Retief St
Kestell (Eastern Free State)
9860 South Africa

Ph.: +27 58 653 1433
Cell: +27 83 251 5824
FAX: +27 58 653 1435

Re: [9fans] sources down

[Steve Simon]

[-- Attachment #1: Type: text/plain, Size: 1184 bytes --]

hi,

one “send it privately” event does not an ethos make. 

i am speaking for Davide here, but i do not think he feels the it was worth publishing widely, being a throw away script.

 i am going to spend some time and try and beautify it, if i succeed i will post my version.

if in the meantime, you are desperate for a copy you could contact him directly and he may well send you a copy.

none of us are required to publish everything we write. 

-Steve

> On 31 Dec 2018, at 8:21 am, Devon H. O'Dell <devon.odell@gmail.com> wrote:
> 
> This “send it to you privately” ethos is a problem on this list. Why not make it public?
> 
> —dho
> 
>> On Sun, Dec 30, 2018 at 2:55 PM David du Colombier <0intro@gmail.com> wrote:
>> > do you have a script that you used to generate the 9hist repository?
>> > I always planned to ingest it into my venti.
>> > 
>> > I have already put a few old floppy and cdrom from my past in
>> > venti as limited /yyyy/mmdd/usr/steve/... so the 9hist could go in
>> > as /yyyy/mmdd/sys/src/9/...
>> 
>> Yes, but beware it's a throw-away script. I'll send it to you privately.
>> 
>> -- 
>> David du Colombier
>> 

[-- Attachment #2: Type: text/html, Size: 2060 bytes --]

Re: [9fans] sources down

[David du Colombier]

Well, here is it: http://www.9legacy.org/9legacy/tools/9hist-to-git.sh

The notes are inside the script.

--
David du Colombier

Re: [9fans] sources down

[hiro]

On 12/31/18, Devon H. O'Dell <devon.odell@gmail.com> wrote:
> This “send it to you privately” ethos is a problem on this list. Why not
> make it public?

I generally agree, but in this case he was forced to spend more work
(cleaning up, and documenting it).
>From my experience with scripts of the same length the work to write
it is probably less than the work to clean it up and document it. And
the task has to be done only once anyway, so not worth it (Unless we
get the annotations).

Re: [9fans] sources down

[Lucio De Re]

There's the other side of the coin, too, just so everyone can tell I'm
cursed with seeing more than one side of situations like this: no one
can learn much from "private exchanges".  So David's freedom not to
publish a one-off ought to be weighed against the benefit that someone
may be able to build on the knowledge he or she may have picked up
from it.

But I'm making a mountain out of a molehill and I apologise for that.
At the same time, I would like to thank David: his contributions are
going mostly unrewarded, the occasional show of gratitude is probably
a good thing.

Lucio. :-)

On 12/31/18, hiro <23hiro@gmail.com> wrote:
> On 12/31/18, Devon H. O'Dell <devon.odell@gmail.com> wrote:
>> This “send it to you privately” ethos is a problem on this list. Why not
>> make it public?
>
> I generally agree, but in this case he was forced to spend more work
> (cleaning up, and documenting it).
> From my experience with scripts of the same length the work to write
> it is probably less than the work to clean it up and document it. And
> the task has to be done only once anyway, so not worth it (Unless we
> get the annotations).
>
>


-- 
Lucio De Re
2 Piet Retief St
Kestell (Eastern Free State)
9860 South Africa

Ph.: +27 58 653 1433
Cell: +27 83 251 5824
FAX: +27 58 653 1435

Re: [9fans] sources down

[hiro]

no i agree lucio, i failed in believing what you described would be
obvious to everybody involved.
the script itself is rather simple. but the outcome is more important:
thanks to david for making sure everybody can find the plan9 history
on google/github without having to get too deeply involved.

Re: [9fans] sources down

[arnold]

"David du Colombier" <0intro@gmail.com> wrote:

> > i have wondered if it would be possible to apply the historic plan9
> > kernel diffs and regenerate these ancient kernels.
>
> The diff files are ed scripts generated with "diff -e", so it should be
> possible to regenerate the original files with a bit of scripting.
>
> --
> David du Colombier

How does that work if lines were deleted?

Thanks,

Arnold

Re: [9fans] sources down

[Kurt H Maier]

On Sun, Dec 30, 2018 at 11:58:20PM -0700, arnold@skeeve.com wrote:
>
> How does that work if lines were deleted?
>
> Thanks,
>
> Arnold
>

$ cat one two
a b c
d e f
g h i

a b c
g h i
j k l

$ diff -e one two
3a
j k l
.
2d


khm

Re: [9fans] sources down

[arnold]

> On Sun, Dec 30, 2018 at 11:58:20PM -0700, arnold@skeeve.com wrote:
> >
> > How does that work if lines were deleted?

Kurt H Maier <khm@sciops.net> wrote:
> [ example session, deleted ]

Let me restate the question.  When one has only the "new" file and
the ed script that created it from the "old" one, and said script
says "delete lines N through M", how does one recover the lines
that were deleted? (With context or unified diffs, the deleted text
is there.)

Thanks,

Arnold

Re: [9fans] sources down

[David du Colombier]

> Let me restate the question.  When one has only the "new" file and
> the ed script that created it from the "old" one, and said script
> says "delete lines N through M", how does one recover the lines
> that were deleted? (With context or unified diffs, the deleted text
> is there.)

I don't think you can "reverse" the ed scripts produced with "diff -e",
unlike unified diff files.

However, in 9hist, the files are always reconstructed forward,
starting from the complete original file.

--
David du Colombier

Re: [9fans] sources down

[arnold]

"David du Colombier" <0intro@gmail.com> wrote:

> > Let me restate the question.  When one has only the "new" file and
> > the ed script that created it from the "old" one, and said script
> > says "delete lines N through M", how does one recover the lines
> > that were deleted? (With context or unified diffs, the deleted text
> > is there.)
>
> I don't think you can "reverse" the ed scripts produced with "diff -e",
> unlike unified diff files.
>
> However, in 9hist, the files are always reconstructed forward,
> starting from the complete original file.

OK, then it makes sense then.

Much thanks!

Arnold

Re: [9fans] sources down

[hiro]

> In all seriousness, it's not too difficult to integrate in. I used Russ'
> letsencrypt package before this one. Both have  been straight forward.
> There are samples too.

yeah, but with dp9ik around nowadays why not just use 9p access for
the really inclined.
somebody who doesn't even run plan9 might not need the extra security
certificate vouched for by the president (of some unstable country of
your chosing).

please don't concentrate on the web interface's web security so much,
i just want the data some way or another - primarily without web
involvement.

Re: [9fans] sources down

[Ethan Gardener]

[-- Attachment #1: Type: text/plain, Size: 716 bytes --]

Oh, when you put it that way, I can't disagree! :D


On Fri, Dec 28, 2018, at 9:15 PM, Skip Tavakkolian wrote:
> It's all relative. For example, compared to the current President of
> the United States, this API is extremely stable :)>
> On Fri, Dec 28, 2018, 11:58 AM Ethan Gardener
> <eekee57@fastmail.fm wrote:>> On Fri, Dec 28, 2018, at 7:43 PM, Skip Tavakkolian wrote:
>>  > while I'm thinking about it, is it possible for 9legacy.org to use
>>  > letsencrypt?   Go's acme/autocert package
>>  > (golang.org/x/crypto/acme/autocert) works great.>>
>>  "This package is a work in progress and makes no API stability
>>  promises.">>


--
The lines in the sand aren't where I thought they were.


[-- Attachment #2: Type: text/html, Size: 1934 bytes --]

Re: [9fans] sources down

[hiro]

hitler jokes work to relativize anything. but eekee's point was valid.

On 12/28/18, Skip Tavakkolian <skip.tavakkolian@gmail.com> wrote:
> It's all relative. For example, compared to the current President of the
> United States, this API is extremely stable :)
>
> On Fri, Dec 28, 2018, 11:58 AM Ethan Gardener <eekee57@fastmail.fm wrote:
>
>> On Fri, Dec 28, 2018, at 7:43 PM, Skip Tavakkolian wrote:
>> > while I'm thinking about it, is it possible for 9legacy.org to use
>> letsencrypt?   Go's acme/autocert package (
>> golang.org/x/crypto/acme/autocert) works great.
>>
>> "This package is a work in progress and makes no API stability promises."
>>
>>
>

Re: [9fans] sources down

[David du Colombier]

> while I'm thinking about it, is it possible for 9legacy.org to use
> letsencrypt? Go's acme/autocert package
> (golang.org/x/crypto/acme/autocert) works great.

Actually, what I need would be an HTTP server supporting
SNI so I could serve the appropriate certificate according to
the requested hostname.

The machine hosting 9legacy.org is also hosting other websites
and currently only presents a valid X.509 certificate for 9grid.fr.

--
David du Colombier

Re: [9fans] sources down

[Lucio De Re]

On 12/29/18, David du Colombier <0intro@gmail.com> wrote:
> Actually, what I need would be an HTTP server supporting
> SNI so I could serve the appropriate certificate according to
> the requested hostname.
>
As far as I can tell (coffee still working it's way to my brain) Go's
HTTPS server does that just fine. Surely all you need is the file
server?

> The machine hosting 9legacy.org is also hosting other websites
> and currently only presents a valid X.509 certificate for 9grid.fr.
>
And, like me, you don't like using multi-domain certificates?

I think the answer is a simple one, but off the cuff, I don't have one.

Personally, I think HTTPS is overrated, or has just made itself overrated.

Lucio.

Re: [9fans] sources down

[Ethan Gardener]

On Sat, Dec 29, 2018, at 3:58 AM, Lucio De Re wrote:
> Personally, I think HTTPS is overrated, or has just made itself overrated.

I've heard from multiple sources that this kind of single-ended authorization isn't worth much, but I'm not so sure.  My current ISP re-encrypts all data from imgur with their own certificate, which means I know the connection is being interfered with.

Re: [9fans] sources down

[Lucio De Re]

On 12/29/18, Ethan Gardener <eekee57@fastmail.fm> wrote:
> On Sat, Dec 29, 2018, at 3:58 AM, Lucio De Re wrote:
>> Personally, I think HTTPS is overrated, or has just made itself
>> overrated.
>
> I've heard from multiple sources that this kind of single-ended
> authorization isn't worth much, but I'm not so sure.  My current ISP
> re-encrypts all data from imgur with their own certificate, which means I
> know the connection is being interfered with.
>
I like to think that I'm as sharp a security specialist as I'm likely
to encounter, despite living in a bit of a backwater.

My ISP (one of, thankfully) managed to arrange a certified X.509
certificate for a domain I have absolute control over. Since then, the
web site has had its ups and down, last of which (I really can't raise
enough anger to take them on) they won't let me install the
certificate myself: it needs root privileges and they won't grant them
to a user.

Thing is, the web site has no need whatever for user security, we hold
no private information,

Lucio.

PS: That was a bit O.T. But I bet I could run the site on a Plan 9 server...

Re: [9fans] sources down

[sl]

i think what annoys people is things (experiments, code, entire
operating systems) that are intended to be kept private neverthess
stil get mentioned.  the happiest scenario is the one we just
witnessed: after an exchange of several more messages on the list, the
bits got published.

thanks, david.

sl