Thus said Richard Hipp on Mon, 06 Jan 2014 21:48:40 -0500:
just by clicking. Then again, maybe I'm the only person who ever wants
to do that so the missing capability is not important...
I think this is easily shown to be problematic by simply browsing to:
Thus said Andy Bradford on 06 Jan 2014 20:01:13 -0700:
Then click Tree-View which gives:
http://www.fossil-scm.org/fossil/dir?ci=13c71a97feb26586name=srctype=tree
One has to manually add expand at this point to reveal the files.
Perhaps Fossil should check it's query parameters
Hello,
I have a question regarding how to handle check-ins that include both a
merge into a branch from trunk (or another branch) and new code.
Specifically, when such an event happens, is it possible when using the
ui (or command line) to view a vdiff between the previous checkin
Thus said Richard Hipp on Sat, 04 Jan 2014 13:13:11 -0500:
From the link above click on family. That shows a small graph
which contains the [555c44eb5d] check-in together with its parents
and children. In that graph, click once on the primary parent
([4f32dced74]) then a
Thus said Richard Hipp on Sat, 04 Jan 2014 21:06:50 -0500:
Maybe this is what you want:
fossil up 4f32dced74
fossil merge b7ff13a0aa
fossil diff --from 555c44eb5d
Yes, this is one way of accomplishing what I want, thank you.
Andy
--
TAI64 timestamp: 400052c8c28d
Thus said Richard Hipp on Thu, 02 Jan 2014 11:32:22 -0500:
I hate having to support --disable-internal-sqlite, and I hate having
to add silly work-arounds in the code to accommodate distributions
trying to use an older SQLite with a newer Fossil. This impedes
progress and
Hello,
I thought I would try the new /tree URI that has been added to Fossil,
but I get the following error:
Files of check-in [daff9d2062]
SQLITE_ERROR: near WITHOUT: syntax error
Just in case others run into this error and scratch their head for a bit
as I did...I get this because I
Thus said Stephan Beal on Mon, 30 Dec 2013 20:41:38 +0100:
On a related note: i had intended to add an ls -1 style output to
/dir over the holidays (based on an ML suggestion back in November),
Isn't this already possible with CSS?
Thanks,
Andy
--
TAI64 timestamp: 400052c1dcd7
Thus said Michai Ramakers on Mon, 30 Dec 2013 21:16:11 +0100:
As for the visual appearance: what would be very useful already for me
is a clear visual difference between a dir- and file-name (in any
file-manager, for that matter).
I use the following CSS for this (provided by someone on
Thus said Stephan Beal on Mon, 30 Dec 2013 22:34:24 +0100:
Good find! Once again, it is proven that i have the memory of a
goldfish!
Hence my suggestion on the other thread that we have some skins that
actually represent these things. I too forget them and have to
Thus said David Rush on Tue, 24 Dec 2013 10:17:59 -0600:
I'm wondering if optional password manager support would be a welcomed
addition to fossil.
Rather than making Fossil aware of different user databases, why
not simply make a Fossil interface for passing user/authentication
Thus said David Rush on Tue, 24 Dec 2013 11:06:50 -0600:
The nice thing about running fossil as a CGI behind a web server is
you automatically get for free any authorization mechanism supported
by the web server.
What you don't get for free is the Fossil assignment of capabilities.
Thus said Jan Nijtmans on Mon, 23 Dec 2013 09:10:29 +0100:
Since the /info page in this case displays a kind of timeline, I would
expect this page to respect the hidden tag as well and have an
Unhide button. Started a new hidden-tag branch now for working on
that.
It's working much
Hello,
When clicking on the genealogy links from the /info page for a hidden
check-in (e.g. family, ancestors, descendants, branch), the timeline
does not display the hidden content which makes the view for hidden
artifacts less useful and there is no way to unhide it except
Thus said Stephan Beal on Sat, 21 Dec 2013 10:30:57 +0100:
[stephan@host:~/cvs/fossil/fossil]$ echo 'select uuid from blob order by
rid limit 1;' | f sqlite3
a28c83647dfa805f05f3204a7e146eb1f0d90505
Ok, I assumed it would involve finding the first rid, but wasn't certain
that this was right.
Thus said Richard Hipp on Fri, 20 Dec 2013 19:05:07 -0500:
http://www.fossil-scm.org/fossil/info/a28c83647d
Is there a command line option that will find this artifact? I thought
perhaps the root:trunk symbolic name would find it, but I was wrong. It
finds
Hello,
Now that the hidden tag feature has been merged into trunk, should there
also be a Setting to always show hidden check-ins in the timeline
display? I would find it useful in local clones of repositories where I
always want to see all of the timeline without resorting to
Thus said Ron Wilson on Wed, 18 Dec 2013 11:30:56 -0500:
This sounds like Fossil is using a tag, TAG_USER, to override the
value specified in the U card. If that is the case, then canceling
that tag should make the value of the U card visible, again.
Yes, this is about the tag, not
Thus said Ron Wilson on Wed, 18 Dec 2013 13:37:57 -0500:
Something that might be worth considering is whether TAG_USER should
actually override the U card value or supplement it. For example,
displaying ${TAG_USER} via ${manifest_U_card} or similar.
As Richard pointed out, the UI
Hello,
I've stumbled upon another potential bug in the ci_edit page wherein if
one edits the User for a check-in, it then displays an option to cancel
the user tag on successive edits:
X Cancel special tag user
If I choose to cancel it, it does actually cancel the user tag, however,
the
Thus said Gilles on Tue, 17 Dec 2013 00:25:47 +0100:
Oh, and for selectively adding,
fossil add $( fossil extra )
is nice, given that fossil extra uses the
ignore-glob to restrict the list.
I assume you meant fossil extras?
Fossil only needs the shortest non-ambiguous substring of
Thus said Ron Wilson on Mon, 16 Dec 2013 10:58:50 -0500:
Note that while you can prevent students from modifying built in
wiki pages, you won't be completely depriving them of the wiki
features. They will just have to create pageName.wiki files and
edit/commit them the same
Hello,
While testing, I created a new fossil and then in the first setup page,
change the project name to just 2 characters (e.g. TF). It didn't
disallow me from doing so, but now I see that it wants to load a Wiki by
the same name, but it doesn't meet the Wiki requirements and so I get
Thus said Ron Wilson on Fri, 13 Dec 2013 21:57:22 -0500:
Since you are providing a server with Fossil repos on it, your build
could work the way you want it for your students to use SSH for access
using stock builds of Fossil.
This is a really good point.
David, it is already possible to
Thus said Jan Nijtmans on Sat, 14 Dec 2013 11:06:15 +0100:
Then the ordering of the tags can be kept the same as it is now,
except that the bug you found is fixed.
Yes, thanks for the suggestion. I think is is better now:
http://www.fossil-scm.org/fossil/info/818c0acd07
Andy
--
TAI64
Thus said David Mason on Sat, 14 Dec 2013 10:38:46 -0500:
That's great! I had obviously misunderstood. I thought that the client
side had different interactions because it used to interact with the
shell, and now didn't.
I actually made it so the new Fossil server code emulates the old
Thus said David Mason on Sat, 14 Dec 2013 16:33:57 -0500:
cp master.fossil $s.fossil
If you create a ``master.fossil'' that is then copied, each project will
have the exact same project-id, and will not be unique. This may not be
what you want. I would move the creation of each fossil
Hello,
I think I've discovered a bug with the way tags are sorted when in the
ci_edit page. Basically, if the name of the tag sorts before the branch
tag, then it displays an option to cancel the tag. For example,
sym-blue, sym-bits, sym-black all sort before ``branch'' and so if
Thus said David Mason on Fri, 13 Dec 2013 13:37:40 -0500:
I am really looking forward to the release that includes the proper
ssh remote repositories.
Will you describe what you envision is a ``proper ssh remote
repository?''
The current SSH implementation in trunk does away
Thus said David Mason on Fri, 13 Dec 2013 13:37:40 -0500:
(If anyone's interested in the details, I'm going to set up accounts
that can only run fossil, and give them each their own fossil with
them, me, and the TA having access to the repository. Then they can
do/submit assignments
Thus said org.fossil-scm.fossil-us...@io7m.com on Fri, 06 Dec 2013 22:25:28
+:
Unless I'm mistaken, it seems that the Plain Gray (No Logo) skin is
missing the following CSS present in some form in all of the other
skins:
Yes, this was missing for some reason and definitely
Thus said Jan Nijtmans on Wed, 04 Dec 2013 10:49:44 +0100:
I'm not sure if some more protection is needed: Any change done
through this page can just as easily been undone through succeeding
edits. For example Branch closure could be hidden if the branch is
trunk, the new branch
Thus said Jan Nijtmans on Mon, 02 Dec 2013 09:52:01 +0100:
This feature is mainly intended for the mistake branch, and if I
make a mistaken commit then the succeeding commits on the same branch
are surely a mistake too!
But if someone simply wants to hide a branch of development from
Thus said Jan Nijtmans on Mon, 02 Dec 2013 09:52:01 +0100:
Any more feedback? If not I would like to merge this to trunk.
I did notice that there appears to be an unused variable:
http://www.fossil-scm.org/index.html/artifact/a366fc950d92e1b7d063855450561128adb149bc?ln=1758
It gets set on
Thus said Jan Nijtmans on Tue, 26 Nov 2013 12:50:52 +0100:
At this moment, only single nodes can be hidden, that's intended. I
see no problems making it propagating, it would make it easier to hide
big parts without anyone noticing. I'm not sure we want to encourage
that.
Tough choice.
Thus said Matt Welland on Tue, 26 Nov 2013 12:01:30 -0700:
Do you now have:
- a - c -
OR
- a c -
What about:
- a = c -
Where = is a new line style that indicates a hidden line of commits
between a and c?
Andy
--
TAI64 timestamp: 40005294f9ae
Thus said Ron Wilson on Tue, 26 Nov 2013 15:17:31 -0500:
At the same time, I find it strange to hide commits whose content is
still being used (unless c is just a reversion of those changes, but
then I'd just commit against a, leaving b as an abandoned
fork/branch)
I too had the
Hello,
There have been two recent changes to Fossil that result in the
following warning after cloning using the --once option:
$ fossil ci -m test
Cannot figure out who you are! Consider using the --user
command line option, setting your USER environment variable,
or setting a default
Thus said Andy Bradford on 25 Nov 2013 08:05:51 -0700:
I would like to propose the following change which does not save the
URL user as the admin user if --once is in force:
Sorry for the noise. I don't think Fossil should make any assumptions in
this case about who the admin user should
Thus said Jan Nijtmans on Mon, 25 Nov 2013 17:08:07 +0100:
Any objections to this change? At least it's not dangerous at all:
It's just as easy to unhide items than to hide them, just by
removing the hidden tag later.
Just some quick observations and questions.
First, if I hide
Thus said Rolf Ade on Sun, 24 Nov 2013 12:41:37 +0100:
But I'm asking for more. I not only want to make the result of my work
within the private branch public, I want to disclose the whole history
of that work.
Instead of a private branch, make a public branch in a private clone
with
Thus said Rolf Ade on Sun, 24 Nov 2013 12:41:37 +0100:
But I'm asking for more. I not only want to make the result of my work
within the private branch public, I want to disclose the whole history
of that work.
In addition to my last suggestion, to reduce the risk of accidental
Thus said Rolf Ade on Sat, 23 Nov 2013 21:36:02 +0100:
My test repository is a very simple one, only a few test commits. Is
this way - delete from private with an sqlite tool on the repository
database - really a working way without sideeffects, to make private
branches public?
Perhaps
Thus said Stephan Beal on Wed, 20 Nov 2013 10:24:02 +0100:
If the patch is relatively small, please just paste it to the list
(don't attach it - attachments get stripped). i don't see the prior
mail explaining the problem - maybe it already contains the patch?
The patch is already in
Thus said Martin Gagnon on Tue, 12 Nov 2013 23:26:59 -0500:
The vdiff page is more propitious to generate big diff since it can
cover many checkins, so I think it's logical to have the option to
hide detailed diff of the whole checkin by default.
Does anyone have any objections
Thus said John L. Poole on Tue, 19 Nov 2013 05:12:08 -0800:
The reason I'm writing to this list is to inquire if there will be a
new release coming shortly or if I should try to create an ebuild
that incorporates the patch that fixes this problem?
I think it would be
Thus said Ben Collver on Tue, 19 Nov 2013 21:24:33 -0800:
* Edit commit faf2ad2219
* Start new branch trunk
You need to also edit 816b791510bd638d and add the trunk branch there.
Andy
--
TAI64 timestamp: 4000528c5460
___
fossil-users mailing
Thus said Tim Chase on Sun, 17 Nov 2013 19:23:41 -0600:
I've been fumbling around and my google-fu is failing me. The help
from fossil help branch doesn't suggest anything obvious to me, and
the mv/rename command renames files, not branches. Is there a simple
way to rename a branch
Thus said Stephan Beal on Fri, 15 Nov 2013 22:33:44 +0100:
@Devs: it seems a new error might have come up in the recent
URL/user/password-related changes?
From what I can tell, it looks unrelated. Fossil does not exchange user
credentials when cloning without a user specified in
Thus said org.fossil-scm.fossil-us...@io7m.com on Fri, 15 Nov 2013 21:37:53
+:
So this time, it actually asked me for a password on cloning (which I
don't remember it ever doing before). I entered the password that was
printed when the repository was created on the server side and
Thus said org.fossil-scm.fossil-us...@io7m.com on Fri, 15 Nov 2013 20:47:47
+:
I'm suddenly seeing not authorized to write on attempting to push to
a newly created and cloned repository. Is there some way to get either
the server or the client to give more information?
fossil push
Thus said org.fossil-scm.fossil-us...@io7m.com on Fri, 15 Nov 2013 23:37:52
+:
Oh dear... I think we've found the problem. I'd cloned the original
repositories over SSH on the same server! I've no idea why I thought
it was over HTTP before.
That explains a lot. SSH URLs have full
Thus said org.fossil-scm.fossil-us...@io7m.com on Fri, 15 Nov 2013 21:37:53
+:
$ fossil user new m0 -R broken.fossil
contact-info:
password:
Retype new password:
What capabilities did you give m0? The default capabilities for a new
user are 'u' (Reader).
To give your user the
Thus said Ron Wilson on Thu, 14 Nov 2013 09:28:57 -0500:
I think it should clearly state it is asking for acceptance of the
proxy cert, when that is the cert being considered.
As far as I am aware, Fossil does not support SSL proxies, so there is
no SSL certificate to be accepted. In
Thus said Martin Gagnon on Tue, 12 Nov 2013 23:26:59 -0500:
My change consist in hiding the detailed diff when the
show-version-diffs setting set to off, just like the info
webpage (when clicking on a checkin link on timeline). I also add the
hide diffs and show diffs
Thus said Jan Nijtmans on Tue, 12 Nov 2013 16:52:57 +0100:
2013/10/29 Jan Danielsson jan.m.daniels...@gmail.com:
Yes, very useful. We only use client certificates, and apache is
set up to set REMOTE_USER from the client certificate, and fossil
uses REMOTE_USER without
Thus said Andy Bradford on 12 Nov 2013 11:28:22 -0700:
Would it be safe to assume that we should disable password
prompting if --ssl-identity is provided?
I can answer that question. NO. I currently use Fossil in this very
manner. I have a client/peer certificate
Thus said Jan Nijtmans on Tue, 12 Nov 2013 16:52:57 +0100:
Does anyone have objections integrating this into trunk? I think it
works fine, but I would appreciate more eyes having a final look at
it.
I'm currently seeing a segfault when I try to clone my SSL protected
(self-signed
Thus said Andy Bradford on 12 Nov 2013 22:14:46 -0700:
#2 0x1c04fd6c in fossil_warning (zFormat=0x0) at printf.c:1078
Something not being handled with the proxy I'm using, which hits these
lines:
http://www.fossil-scm.org/index.html/artifact/25e445fe1b3ff036fa53334eb4ce2784bb41eb04?ln=268
Thus said Jan Nijtmans on Tue, 12 Nov 2013 16:52:57 +0100:
I think it works fine, but I would appreciate more eyes having a final
look at it.
Here's another observation. Now that I got around the proxy problem (the
proxy I was using was returning 403), I get prompted to accept the
Thus said j. van den hoff on Sun, 10 Nov 2013 12:32:50 +0100:
the latter (`myname') is directly and consistently reported as the
user doing future commits and wiki as well as ticket changes. this is
perfectly how it should be. however: `myname' is reported as having
only reader
Thus said j. van den hoff on Sun, 10 Nov 2013 12:44:30 +0100:
in my local clone of the repo, however, the _same_ checkin (from
somebody (and somewhere) else) is reported as originating from me@myIP
(i.e. the `s' user of the local clone).
is this a bug or just a misinterpretation of
Thus said j. van den hoff on Wed, 06 Nov 2013 10:18:08 +0100:
the question to the developer(s) would be: is the present behaviour
natural from some higher perspective? or is it really an
inconsistency that should/could be fixed? I would think that the
default user setting
Thus said Jan Nijtmans on Thu, 07 Nov 2013 12:41:01 +0100:
I merged the url-password-fixes and the jan-httpsproxytunnel
branched together, and did various tests using https over a http
tunnel. This confirmed what I already guessed: The merge of those two
branches didn't have any
Thus said j. van den hoff on Wed, 06 Nov 2013 10:18:08 +0100:
fossil clone http://userid@server/repo myclone.fossil
fossil user new localname
fossil user default localname
all commits appear in the timeline with user localname (as they
should).
however, the GUI says logged in as
Thus said Jan Nijtmans on Tue, 05 Nov 2013 11:14:07 +0100:
I now added the --once option to fossil clone. I can only think of
one use-case: when cloning a repository where autosync is going to be
off, which is going to be synced automatically using a script
containing fossil sync
Thus said Andy Bradford on 05 Nov 2013 22:04:14 -0700:
Thanks for the help. Looks like it might be ready?
Nope, not ready yet... I just encountered a problem when cloning.
Andy
--
TAI64 timestamp: 40005279d649
___
fossil-users mailing list
Thus said Jan Nijtmans on Sat, 02 Nov 2013 12:11:51 +0100:
I just made a little modification preventing the prompt if the saved
password is already equal to the to-be-saved password. I will do some
more testing the coming days, but I think the pending-review branch
(another name would be
Thus said Jan Nijtmans on Fri, 01 Nov 2013 16:53:24 +0100:
[Be aware that I have only been looking at the Fossil code for a short
time, so what I say may be inaccurate]
1) If a password is necessary and there is a saved password, use
the saved password.
2) Otherwise, prompt
Thus said Ron Wilson on Fri, 01 Nov 2013 12:10:19 -0400:
Also, I would suggest that before prompting, Fossil determine if STDIN
and (STDERR or STDOUT) are associate with a console. If not, then
Fossil is not running interactively.
The normal password prompting already checked for the
Thus said Martin Gagnon on Fri, 01 Nov 2013 06:52:33 -0400:
It's ok for me, but I still think that specifying password in URL in
interactive shell a bad idea. E.g. It got saved in shell command
history.
I too stay away from passwords in command line shell commands and prefer
to use
Thus said Jan Nijtmans on Fri, 01 Nov 2013 16:53:24 +0100:
fossil clone [--once] http://u...@url.org
By the way, clone does not currently have a --once:
$ fossil clone --once http://tester:testing@localhost:8081/ clone.fossil
unknown repository: --once
Andy
--
TAI64 timestamp:
Thus said Jan Nijtmans on Sat, 02 Nov 2013 12:11:51 +0100:
Agreed. A single retry is enough for me too.
As it turns out, I can get it to prompt more than once easily enough,
but 2 is the limit here because we drop out of the http exchange loop if
we've seen 2 login failure messages. It
Thus said Andy Bradford on 02 Nov 2013 10:07:30 -0600:
I'm now investigating why it didn't prompt me to save the password (it
has nothing to do with your change), but you'll notice that it did
prompt me twice.
Haha, it was not prompting me because I entered the same failed
Thus said =?Big5?B?vEKq2qtG?= on Fri, 01 Nov 2013 13:55:50 +0800:
Great thanks, I was just wondering about that. I tried the fix and
works well for me, it now prompts for remembering password when
cloning with username+password.
As it turns out, after I committed the fix, I
Thus said Ron Wilson on Thu, 31 Oct 2013 09:54:29 -0400:
Error: login failed, missing or incorrect password for user xxx
Did Fossil ask you for the password after this error? If not, it
should have.
Fossil 1.27 will not do this, but Fossil from trunk will prompt for it.
Andy
--
Thus said =?Big5?B?vEKq2qtG?= on Fri, 01 Nov 2013 10:10:53 +0800:
Thanks to all for your response, I have tried the trunk and it does
work, here is test result/comparison:
I just now committed a fix to pending-review that should also correct
the missing prompt to save the password which
Thus said David Blanford on Mon, 28 Oct 2013 16:33:11 -0600:
Fossil internal error: repository does not exist or is in an
unreadable directory:
Can you cause this to happen again with a different fossil file? If so,
will you share the exact commands you typed to cause this?
Thanks,
Thus said Stephan Beal on Thu, 17 Oct 2013 16:49:53 +0200:
As far as the other changes go regarding the # symbol - i'm still
waiting for someone else to volunteer :). As a general rule, i don't
like to change features which have survived the test of time (even if
they're not 100%
Thus said sky5w...@gmail.com on Thu, 17 Oct 2013 13:06:02 -0400:
Understood, but the point was to not create a message file.
My example does not create a message file but instead tells fossil to
read the commit message from stdin (using a here document):
$ f ci -M - file EOF
This is a
Thus said j. van den hoff on Tue, 15 Oct 2013 18:39:50 +0200:
Accept certificate for host {ourserver} (a=always/y/N)
which he answered with y which leads to
server did not reply
I just tested this with my SSL environment (tcpserver+stunnel+fossil)
and it worked fine. I have a
Thus said Stephan Beal on Tue, 15 Oct 2013 19:08:30 +0200:
[stephan@host:~/cvs/fossil/fossil/src]$ grep 'server does not respond' *.*
The actual error is:
Accept certificate for host fossil (a=always/y/N)? y
server did not reply
$ grep 'server did not reply' *.*
http.c:fossil_fatal(server
Thus said B Harder on Thu, 10 Oct 2013 11:03:59 -0700:
I was getting no common ancestor.
I too get the warning when using your broken.fsl if I merge into the
feature branch from trunk:
$ f mer 823ef8
WARNING - no common ancestor: a
WARNING - no common ancestor: b
WARNING - no common
Hello,
Should these two pending-review items be included in 1.27:
http://www.fossil-scm.org/index.html/info/8ad0a8796a
http://www.fossil-scm.org/index.html/info/bdbf144ed3
Thanks,
Andy
--
TAI64 timestamp: 4000522e9141
___
fossil-users mailing
Hello,
It would seem there is a bug in update_cmd(). Cloning, syncing and other
operations will call user_select() to setup g.userUid among other
things. This does not happen with update_cmd() so when I'm viewing
/info/artifact through the web UI, it shows ``unknown @ IP'' as
Hello,
In an effort to simplify the code changes being made, I have separated
out the SSH transport changes from the development of the new (and
somewhat experimental) shared SSH account code. I did this so that it
might be easier to get the more critical SSH transport changes
Thus said Matt Welland on Sat, 07 Sep 2013 20:22:42 -0700:
What is the recommended ssh command line for Linux? 'ssh -q' seems to
work fine for me.
I typically don't change the ssh command line unless necessary and the
default of ``ssh -e none -T'' seems to work fine. I haven't seen a
Thus said David Given on Wed, 04 Sep 2013 22:07:32 +0100:
Based on various conversations in the past I have been working on a
semi-interesting features: using mnemonic encoding to turn ticket
hashes into something that humans can remember.
What about using the ICAO alphabet? It be
Thus said John Long on Wed, 28 Aug 2013 11:57:01 -:
There are two value added things digital signing provides over
hashing in this specific example when fossil uses SHA1. One, a person
is taking responsibility for a commit and saying I did this. Two,
PGP can use much stronger
Thus said Stephan Beal on Sat, 17 Aug 2013 14:03:45 +0200:
When building fossil:
make: *** No rule to make target `src/../manifest.uuid', needed by
`bld/VERSION.h'. Stop.
I usually encounter this when I move the manifest* files to open a
different repository in the same directory
Hello,
I've just committed code that will cause the SSH client and server to
use a single SSH connection, rather than spawning a new connection for
each Round-trip. This makes it significantly more performant as it is no
longer required to open up multiple connections. It will obviously
Thus said Chad Perrin on Wed, 14 Aug 2013 16:07:02 -0600:
Is there another option for this when developers basically do not have
access to the server where the main Fossil repo is located except by
way of clone/pull/push/sync?
Well, technically speaking, they *do* have access to the
Thus said Chad Perrin on Wed, 14 Aug 2013 16:57:53 -0600:
Have them point their browser to http://localhost:/
I guess I should consider running the web interface too, then, if
that's the only way to let them change their Fossil passwords.
I thought you already were running the
Thus said David Mason on Tue, 13 Aug 2013 13:12:47 -0400:
All I want is for my users to say:
fossil clone ssh://remote/proj.fossil clone.fossil
[snip]
It appears that I can do that (possibly with a small auxiliary script)
with the changes that Andy is working on.
You can probably
Thus said Chad Perrin on Tue, 13 Aug 2013 18:39:21 -0600:
Everything was working great until I tried to use the following in the
authorized_keys file for the user account hosting the Fossil repos:
command=/usr/local/bin/fossil,no-X11-forwarding,no-agent-forwarding
ssh-rsa key
Thus said Chad Perrin on Sun, 11 Aug 2013 06:38:45 -0600:
It should still log Fossil usernames though -- right? If so, that'll
do, at least for now.
Yes, it will do that.
Andy
--
TAI64 timestamp: 400052079daa
___
fossil-users mailing list
Thus said Stephan Beal on Sun, 11 Aug 2013 15:26:50 +0200:
One of the devs (Andy?) has been working on integrating ssh forced
commands with fossil so that ssh connections can use fossil's
authentication. i'm not sure what the status of that is, but from what
i've read it sounds
Thus said Andy Bradford on 11 Aug 2013 09:15:53 -0600:
# will prompt for Fossil user otheruser but login to SSH using local
USER fossil clone -h on -l otheruser ssh://remote//tmp/proj.fossil
clone.fossil
I'm also considering dropping -h altogether, and making neither options
global
Thus said Chad Perrin on Sun, 11 Aug 2013 10:10:40 -0600:
On Sun, Aug 11, 2013 at 03:24:40PM +0200, Rene wrote:
On 2013-08-11 14:49, Chad Perrin wrote:
Err . . . wait. Is it not logging the *user*, or just the IP
address? What would it log in place of the actual authenticated
Thus said Chad Perrin on Sat, 10 Aug 2013 18:07:28 -0600:
What's the quick/easy way to get Fossil set up so a small team can
push/pull/sync multiple Fossil repositories on the server without
having shell accounts?
At the moment, this type of SSH integration isn't as flexible as
801 - 900 of 969 matches
Mail list logo
