2009/11/7 Willem Jan Palenstijn <w...@usecode.org>:
>
> On Sat, Nov 07, 2009 at 11:02:07AM +0000, Dr. David Kirkby wrote:
>>
>> I built Sage 4.2 on Solaris 10 on a Sun Netra T1 running the first release of
>> Solaris 10.
>>
>> Whilst computing things like 1+1 work fine in command line mode, they do not
>> work in the notebook. The input is accepted in the notebook, but no output is
>> ever displayed.
>>
>>[...]
>>
>> 2009-11-07 10:49:32+0000 [HTTPChannel,2,192.168.1.11] got EOF subprocess must
>> have crashed...
>> 2009-11-07 10:49:32+0000 [HTTPChannel,2,192.168.1.11] The authenticity of
>> host
>> 'localhost (127.0.0.1)' can't be established.
>> 2009-11-07 10:49:32+0000 [HTTPChannel,2,192.168.1.11] RSA key fingerprint is
>> 55:45:b6:ed:05:bb:e1:be:21:0a:0a:93:87:15:4d:41.
>> 2009-11-07 10:49:32+0000 [HTTPChannel,2,192.168.1.11] Are you sure you want
>> to
>> continue connecting (yes/no)?
>> 2009-11-07 10:49:32+0000 [HTTPChannel,2,192.168.1.11] import
>> os;os.chdir("/tmp/tmpnMPbma");
>> 2009-11-07 10:49:32+0000 [HTTPChannel,2,192.168.1.11]
>> execfile("_sage_input_2.py")Host key verification failed.
>> 2009-11-07 10:49:32+0000 [HTTPChannel,2,192.168.1.11]
>
> It looks like the ssh connection is failing. Try if it works after first
> manually ssh-ing to sa...@localhost once to answer the 'Are you sure you want
> to continue connecting?' question.
>
>
> -Willem Jan
That did fix it. I did not actually logm in as sa...@localhost, since
I would not have known any password for 'sage1' as there was no user
'sage1' in /etc/paswd, but I did say "yes" to the 'Are you sure you
want to continue connecting?'" question.
Is this only observed on Solaris? It seems strange that it should be.
This should be documented, since it is far from obvious for someone to
know how to solve - or perhaps I am just thick!
Every attempt I ever made to build Sage on my Sun Blade 2000 has
always resulted in there being no server running on port 8000. This
time, where I used another machine to build Sage, the server did at
least run. But there could be many reasons this has behaviour has
changed.
* A Sun Netra T1 was used not a Blade 2000 (Both are SPARC though).
* My Blade 2000 runs the latest Solaris 10 - this machine the first
version of Solaris 10.
* This time browser and host were on different computers.
* I believe you have made some changes to the notebook in Sage
recently, though whether that includes the server I do not know.
* This was built using gcc 4.4.2 - every other attempt has been made
with an older gcc, though recent attempts have used 4.4.1, so the
changes should be minimal.
I'm not at home now (I had to go to my fathers yesterday to use his
Internet connection, as mine had failed and I was scheduled to play a
game of chess for a team).
When I get home later today, I'll implement Gonzalo Tornaria's
suggestions to changes for the script that builds the binary, and move
the resulting binary to another machine to test it out.
If you want, I could create a zone on 't2' and run the server publicly on 't2'.
Dave.
--~--~---------~--~----~------------~-------~--~----~
To post to this group, send an email to sage-devel@googlegroups.com
To unsubscribe from this group, send an email to
sage-devel-unsubscr...@googlegroups.com
For more options, visit this group at http://groups.google.com/group/sage-devel
URL: http://www.sagemath.org
-~----------~----~----~----~------~----~------~--~---
