On Thu, Oct 23, 2014 at 5:06 PM, Eric Rubin-Smith wrote:
> Stephan Beal wrote:
> > We'd be interested in hearing back if you discover how an error on your
> > end (if indeed it is) is confusing fossil into trying to read
> non-manifest
> > files as manifests.
>
> Actually, your question sort of c
Stephan Beal wrote:
> We'd be interested in hearing back if you discover how an error on your
> end (if indeed it is) is confusing fossil into trying to read non-manifest
> files as manifests.
Actually, your question sort of confuses me. From my (very tenuous)
understanding of the code, it
On Thu, Oct 23, 2014 at 1:59 PM, Eric Rubin-Smith wrote:
> My current suspicion is that something about the PKI stuff on my
> server box has changed, which causes Tcl to have trouble connecting to a
> remote HTTPS server from my TH1 hooks. I.e. Fossil itself is no longer
> implicated as a contri
Stephan Beal wrote:
> FYI: i committed one on top of that. The advantage is that it's
> centralized, the disadvantage is that it hashes every manifest before
> parsing (to get the UUID, since parsing modifies it). Might be considered
> too expensive, considering how rare broken manifests are
On Wed, Oct 15, 2014 at 2:36 PM, Stephan Beal wrote:
> FYI: i committed one on top of that. The advantage is that it's
> centralized, the disadvantage is that it hashes every manifest before
> parsing (to get the UUID, since parsing modifies it). Might be considered
> too expensive, considering h
On Wed, Oct 15, 2014 at 2:27 PM, Eric Rubin-Smith wrote:
> Richard Hipp wrote:
>
> > Can you try compiling the tip of the "better-error-msgs" branch and put
> > that on your server, then let us know what the new error message is?
>
> Here we go:
>
> $ fossil push
> Push to https://eas@:10444/
> R
Stephan Beal wrote:
> You mention custom changes with regards to push scripts, which make me
> curious about that error message: i can find "push script failed" nowhere
> in fossil. Where is it coming from (if you know)?
The strings are escaped :)
eric@dev:~/Fossil-5ff4e33617/src$ grep -r 'push\
Eric Rubin-Smith wrote:
> $ fossil push
> Push to https://eas@:10444/
> Round-trips: 1 Artifacts sent: 2 received: 0
> Error: push script failed: syntax error in manifest
> [449bc674c3fc9c036db0d4dc71281b7cb900fe7d]
> Round-trips: 1 Artifacts sent: 2 received: 0
> Push finished with 818173
On Wed, Oct 15, 2014 at 8:19 PM, Richard Hipp wrote:
> On Wed, Oct 15, 2014 at 2:15 PM, Eric Rubin-Smith
> wrote:
>
>> Richard Hipp wrote:
>>
>> > Are you sure that is the manifest that is causing the problem? Perhaps
>> we
>> > should enhance the error message to include the UUID of the proble
On Wed, Oct 15, 2014 at 8:27 PM, Eric Rubin-Smith wrote:
> Error: push script failed: syntax error in manifest
> [449bc674c3fc9c036db0d4dc71281b7cb900fe7d]
>
You mention custom changes with regards to push scripts, which make me
curious about that error message: i can find "push script failed" n
Richard Hipp wrote:
> Can you try compiling the tip of the "better-error-msgs" branch and put
> that on your server, then let us know what the new error message is?
Here we go:
$ fossil push
Push to https://eas@:10444/
Round-trips: 1 Artifacts sent: 2 received: 0
Error: push script failed: sy
Richard Hipp wrote:
> Can you try compiling the tip of the "better-error-msgs" branch and put
> that on your server, then let us know what the new error message is?
Our emails crossed paths. I'll do as you suggest and let you know what
happens.
--
Eric A. Rubin-Smith
Aterlo Networks, Inc.
http
Stephan Beal wrote:
> Working on a patch now.
Sorry, I'm realizing that the Fossil version I mentioned (1.29) is
tainted with my own private changes to the C code (to get my TH1 hooks
working). I submitted them to this list, and Joe Mistachkin accepted
them -- but with significant changes. It
On Wed, Oct 15, 2014 at 2:15 PM, Eric Rubin-Smith wrote:
> Richard Hipp wrote:
>
> > Are you sure that is the manifest that is causing the problem? Perhaps
> we
> > should enhance the error message to include the UUID of the problem
> > manifest as part of the error message?
>
> In case you find
Richard Hipp wrote:
> Are you sure that is the manifest that is causing the problem? Perhaps we
> should enhance the error message to include the UUID of the problem
> manifest as part of the error message?
In case you find it useful for your debugging, I have continued plodding
along with my no
On Wed, Oct 15, 2014 at 7:56 PM, Richard Hipp wrote:
> Perhaps we should enhance the error message to include the UUID of the
> problem manifest as part of the error message?
>
Working on a patch now.
--
- stephan beal
http://wanderinghorse.net/home/stephan/
http://gplus.to/sgbeal
"Freedom
On Wed, Oct 15, 2014 at 1:47 PM, Stephan Beal wrote:
>
>>
>> Let me re-send you the manifest as a 'gz' file. Hopefully no code
>> between me and you will be too clever for its own good this time.
>>
>
> Good idea. Please send it to Richard, too.
>
>
"fossil test-parse" runs without error on the
On Wed, Oct 15, 2014 at 7:47 PM, Stephan Beal wrote:
> On Wed, Oct 15, 2014 at 7:44 PM, Eric Rubin-Smith
> wrote:
>
>> Let me re-send you the manifest as a 'gz' file. Hopefully no code
>>
> between me and you will be too clever for its own good this time.
>>
>
No, that one looks good:
[stephan
On Wed, Oct 15, 2014 at 7:44 PM, Eric Rubin-Smith wrote:
> How dare you suggest it! :) I don't think that's what's happening.
> Maybe that was your mail client when you saved my attachment, or mine
> when I initially sent it (though I doubt it, since I am using a
> Linux + mutt).
>
> Here's wh
Stephan Beal wrote:
> According my debugger, that uuid is 41 bytes long, including a
> carriage-return character. How on earth that happend is a mystery to me, as
> fossil does not use \r characters anywhere unless it's prescribed by a
> standard (e.g. HTTP headers).
>
> The C-card also contains
On Wed, Oct 15, 2014 at 7:31 PM, Stephan Beal wrote:
> F .fossil-settings/allow-symlinks a09a968bf05a50058f3ad50132730b719bc39e76
>
> That UUID is 40 bytes long, i.e. an SHA1 hash, which is syntactically
> correct So i'm not sure what he's complaining about. Will need to run this
> through the de
Can somebody please forward the bad manifest to me too?
On Wed, Oct 15, 2014 at 1:31 PM, Stephan Beal wrote:
> On Wed, Oct 15, 2014 at 7:22 PM, Stephan Beal
> wrote:
>
>> [stephan@host:~/cvs/fossil/cwal]$ f-mfparse ~/Downloads/manifest.txt
>> Parsing this manifest: /home/stephan/Downloads/manif
On Wed, Oct 15, 2014 at 7:22 PM, Stephan Beal wrote:
> [stephan@host:~/cvs/fossil/cwal]$ f-mfparse ~/Downloads/manifest.txt
> Parsing this manifest: /home/stephan/Downloads/manifest.txt
> f-mfparse: ERROR #128 (FSL_RC_CA_SYNTAX): Malformed UUID in F-card
>
> Now i need to expand the error reporti
On Wed, Oct 15, 2014 at 7:11 PM, Stephan Beal wrote:
> On Wed, Oct 15, 2014 at 7:03 PM, Eric Rubin-Smith
> wrote:
>
>> Keep in mind that this 'master' repo doesn't have any sandboxes open
>> against it. It just is there for syncing and the Transfer hooks. So
>> there's no sandbox in which to t
On Wed, Oct 15, 2014 at 7:03 PM, Eric Rubin-Smith wrote:
> Keep in mind that this 'master' repo doesn't have any sandboxes open
> against it. It just is there for syncing and the Transfer hooks. So
> there's no sandbox in which to turn on the 'manifest' setting.
>
i suspect it's choking on the
Stephan Beal wrote:
> > and then the manifest should be dumped to "manifest.uuid". Please post the
> > contents of that file for a checkin which fails this way, and i think that
> > will show us what's wrong.
And, just to add some fun, I just got a different error message while
trying to run 'syn
Stephan Beal wrote:
> > and then the manifest should be dumped to "manifest.uuid". Please post the
> > contents of that file for a checkin which fails this way, and i think that
> > will show us what's wrong.
Keep in mind that this 'master' repo doesn't have any sandboxes open
against it. It jus
On Wed, Oct 15, 2014 at 6:48 PM, Stephan Beal wrote:
> fossil set manifest on
>
> (if it's not already)
>
> and then the manifest should be dumped to "manifest.uuid". Please post the
> contents of that file for a checkin which fails this way, and i think that
> will show us what's wrong.
>
My mi
On Wed, Oct 15, 2014 at 6:31 PM, Eric Rubin-Smith wrote:
> I tried this, which doesn't give much useful info:
>
> $ fossil test-parse-manifest ./manifest
>
> I assume it's the remove fossil complaining, though -- not the local one.
>
That's my assumption, too, but can you paste that one for us?
I haven't changed much recently about my repository topology. One central
"master" repo that I push to from my "dev" repo. No one else is doing any
pushing. The master repo has hooks set up for publishing my commits etc to
a remote HTTP REST API.
The only change I made recently is to open up my
30 matches
Mail list logo