You are viewing a plain text version of this content. The canonical link for it is here.
Posted to dev@subversion.apache.org by Julian Foad <ju...@btopenworld.com> on 2003/10/22 17:39:48 UTC

Re: svn rev 7469: FAIL (i686-pc-linux-gnu static ra_svn)

Enough of these 505KB messages.  I am going to unsubscribe from the "breakage" list until either failures like this are fixed quickly or (more reasonably) the failure message is made much smaller.  Downloading half-megabyte messages at 40kbps isn't fun.  (OK, it only takes a few minutes.  I can cope with it once or twice a day for a few days.  I have done for the last few days.  I'm just feeling cranky again.)

- Julian


brane@xbc.nu wrote:
> PASS: update apr
> PASS: update apr-util
> PASS: update subversion
> PASS: autogen.sh
> 
> BUILD: Revision 7469 on i686-pc-linux-gnu static
> 
[... some failure is in there somewhere ...]
[... for 505 KB total e-mail message size.]


---------------------------------------------------------------------
To unsubscribe, e-mail: dev-unsubscribe@subversion.tigris.org
For additional commands, e-mail: dev-help@subversion.tigris.org

Re: svn rev 7469: FAIL (i686-pc-linux-gnu static ra_svn)

Posted by Branko Čibej <br...@xbc.nu>.
Julian Foad wrote:

> Branko Čibej wrote:
>
>> In the meantime, I changed the test scripts to compress the tests.log
>> file before attaching it to the FAIl post. That should bring the size of
>> the posts down to about 40k, which should be acceptable.
>
>
> Thanks.  Because of that I am now re-subscribing.

Ah, that makes midnight coding worthwile. :-)

-- 
Brane Čibej   <br...@xbc.nu>   http://www.xbc.nu/brane/


---------------------------------------------------------------------
To unsubscribe, e-mail: dev-unsubscribe@subversion.tigris.org
For additional commands, e-mail: dev-help@subversion.tigris.org

Re: svn rev 7469: FAIL (i686-pc-linux-gnu static ra_svn)

Posted by Julian Foad <ju...@btopenworld.com>.
Branko Čibej wrote:
> In the meantime, I changed the test scripts to compress the tests.log
> file before attaching it to the FAIl post. That should bring the size of
> the posts down to about 40k, which should be acceptable.

Thanks.  Because of that I am now re-subscribing.

- Julian


---------------------------------------------------------------------
To unsubscribe, e-mail: dev-unsubscribe@subversion.tigris.org
For additional commands, e-mail: dev-help@subversion.tigris.org

Re: svn rev 7469: FAIL (i686-pc-linux-gnu static ra_svn)

Posted by kf...@collab.net.
Branko Čibej <br...@xbc.nu> writes:
> In the meantime, I changed the test scripts to compress the tests.log
> file before attaching it to the FAIl post. That should bring the size of
> the posts down to about 40k, which should be acceptable.

Thanks!

---------------------------------------------------------------------
To unsubscribe, e-mail: dev-unsubscribe@subversion.tigris.org
For additional commands, e-mail: dev-help@subversion.tigris.org

Re: svn rev 7469: FAIL (i686-pc-linux-gnu static ra_svn)

Posted by Branko Čibej <br...@xbc.nu>.
In the meantime, I changed the test scripts to compress the tests.log
file before attaching it to the FAIl post. That should bring the size of
the posts down to about 40k, which should be acceptable.

-- 
Brane Čibej   <br...@xbc.nu>   http://www.xbc.nu/brane/


---------------------------------------------------------------------
To unsubscribe, e-mail: dev-unsubscribe@subversion.tigris.org
For additional commands, e-mail: dev-help@subversion.tigris.org

Re: svn rev 7469: FAIL (i686-pc-linux-gnu static ra_svn)

Posted by Philip Martin <ph...@codematters.co.uk>.
Philip Martin <ph...@codematters.co.uk> writes:

> Branko Čibej <br...@xbc.nu> writes:
>
>> IMO checking how your changes behave on platforms you can't test
>> yourself is a matter of public hygiene.
>
> In this particular case one of the merge tests is failing, so it's
> probably my recent commit to merge_tests.py.

valgrind has identified the problem, it's a long standing bug in
dry-run merge, the svn_wc_diff_callbacks_t merge callbacks sometimes
don't set *state, so leaving a random value for the caller.

-- 
Philip Martin

---------------------------------------------------------------------
To unsubscribe, e-mail: dev-unsubscribe@subversion.tigris.org
For additional commands, e-mail: dev-help@subversion.tigris.org

Re: svn rev 7469: FAIL (i686-pc-linux-gnu static ra_svn)

Posted by Philip Martin <ph...@codematters.co.uk>.
Branko Čibej <br...@xbc.nu> writes:

> IMO checking how your changes behave on platforms you can't test
> yourself is a matter of public hygiene.

In this particular case one of the merge tests is failing, so it's
probably my recent commit to merge_tests.py.  It only fails for static
ra_svn on Linux, it doesn't fail on freebsd, nor does it fail on my
Linux box (and I've tried static ra_svn).  I assume it's platform, and
maybe machine, specific; I don't know how to investigate it further.

-- 
Philip Martin

---------------------------------------------------------------------
To unsubscribe, e-mail: dev-unsubscribe@subversion.tigris.org
For additional commands, e-mail: dev-help@subversion.tigris.org

Re: svn rev 7469: FAIL (i686-pc-linux-gnu static ra_svn)

Posted by Branko Čibej <br...@xbc.nu>.
kfogel@collab.net wrote:

>Branko Čibej <br...@xbc.nu> writes:
>  
>
>>>Enough of these 505KB messages.  I am going to unsubscribe from the
>>>"breakage" list until either failures like this are fixed quickly or
>>>(more reasonably) the failure message is made much smaller.
>>>      
>>>
>>You may unsibscribe from the breakage list if you promise never to
>>commit again. IMNSHO.
>>    
>>
>
>Erk, gently, dear :-).  IMHO, the method people use to detect breakage
>is up to them.  The breakage list is a convenience, but Julian isn't
>required to be subscribed to it.  He's perfectly free to hire a butler
>to read the list for him, or whatever.
>
>(I.e., you can't force someone to read mail anyway.  Let's judge a
>committer by their commits, by which measure Julian has nothing to
>apologize for.)
>
>I agree with your position that that we shouldn't change the breakage
>output; I'm just talking about the rest of the mail.  Let's not turn
>into a bunch of hair-trigger hardliners like that... other ... project
>:-).
>  
>
Actually, it's the other way around. I wouldn't mind changing the
breakage output *if* we can make sure that all relevant information is
there. I do believe that committers should keep an eye on the breakage
posts.

Yes, I know you can't force anyone to subscribe or read the mail, but
IMO checking how your changes behave on platforms you can't test
yourself is a matter of public hygiene.


-- 
Brane Čibej   <br...@xbc.nu>   http://www.xbc.nu/brane/


---------------------------------------------------------------------
To unsubscribe, e-mail: dev-unsubscribe@subversion.tigris.org
For additional commands, e-mail: dev-help@subversion.tigris.org

Re: svn rev 7469: FAIL (i686-pc-linux-gnu static ra_svn)

Posted by kf...@collab.net.
Branko Čibej <br...@xbc.nu> writes:
> > Enough of these 505KB messages.  I am going to unsubscribe from the
> > "breakage" list until either failures like this are fixed quickly or
> > (more reasonably) the failure message is made much smaller.
> 
> You may unsibscribe from the breakage list if you promise never to
> commit again. IMNSHO.

Erk, gently, dear :-).  IMHO, the method people use to detect breakage
is up to them.  The breakage list is a convenience, but Julian isn't
required to be subscribed to it.  He's perfectly free to hire a butler
to read the list for him, or whatever.

(I.e., you can't force someone to read mail anyway.  Let's judge a
committer by their commits, by which measure Julian has nothing to
apologize for.)

I agree with your position that that we shouldn't change the breakage
output; I'm just talking about the rest of the mail.  Let's not turn
into a bunch of hair-trigger hardliners like that... other ... project
:-).

-K

---------------------------------------------------------------------
To unsubscribe, e-mail: dev-unsubscribe@subversion.tigris.org
For additional commands, e-mail: dev-help@subversion.tigris.org

Re: svn rev 7469: FAIL (i686-pc-linux-gnu static ra_svn)

Posted by Julian Foad <ju...@btopenworld.com>.
Branko Čibej wrote:
> Julian Foad wrote:
> 
>>Enough of these 505KB messages.  I am going to unsubscribe from the
>>"breakage" list until either failures like this are fixed quickly or
>>(more reasonably) the failure message is made much smaller.
> 
> You may unsibscribe from the breakage list if you promise never to
> commit again. IMNSHO.

I sort of agree in principle.  But not in practice, right now.

> And judging by the fact that this failure seems to have been around for
> the last 4 days, it seems that some people _do_ commit and then don't
> check for breakage. Duh.


>>Downloading half-megabyte messages at 40kbps isn't fun.  (OK, it only
>>takes a few minutes.  I can cope with it once or twice a day for a few
>>days.  I have done for the last few days.  I'm just feeling cranky
>>again.)
> 
> Instead of ranting, why don't you propose how to shorten those failure
> messages instead? Or better yet, modify the test scripts, they're right
> there in the repo.

Yes, I know I should do that.  Perhaps I will some day soon.  Sorry I ranted.  As I said, I was in a cranky mood.

It's a valuable service.  Thanks for running it.

- Julian


---------------------------------------------------------------------
To unsubscribe, e-mail: dev-unsubscribe@subversion.tigris.org
For additional commands, e-mail: dev-help@subversion.tigris.org

Re: svn rev 7469: FAIL (i686-pc-linux-gnu static ra_svn)

Posted by Branko Čibej <br...@xbc.nu>.
Julian Foad wrote:

> Enough of these 505KB messages.  I am going to unsubscribe from the
> "breakage" list until either failures like this are fixed quickly or
> (more reasonably) the failure message is made much smaller.

You may unsibscribe from the breakage list if you promise never to
commit again. IMNSHO.

And judging by the fact that this failure seems to have been around for
the last 4 days, it seems that some people _do_ commit and then don't
check for breakage. Duh.

> Downloading half-megabyte messages at 40kbps isn't fun.  (OK, it only
> takes a few minutes.  I can cope with it once or twice a day for a few
> days.  I have done for the last few days.  I'm just feeling cranky
> again.)

Instead of ranting, why don't you propose how to shorten those failure
messages instead? Or better yet, modify the test scripts, they're right
there in the repo.


-- 
Brane Čibej   <br...@xbc.nu>   http://www.xbc.nu/brane/


---------------------------------------------------------------------
To unsubscribe, e-mail: dev-unsubscribe@subversion.tigris.org
For additional commands, e-mail: dev-help@subversion.tigris.org