ace_4.9c saceserver on solaris?

Ed Griffiths edgrif at sanger.ac.uk
Thu Sep 27 06:46:35 EST 2001


Dave,

> What??  Yes this is correct, I just tested.  saceserver won't start unless
> it has _write_ access to the database directory.
> 
> This is quite a change from the rpc aceserver's behavior, and a surprising
> one.  I see now that wdoc/SOCKET_aceserver.html says,

Currently the server will not continue if it can't write to the log file, that
is the only file it actually needs to write to. I can add configuration
parameter if required to disable this. My thinking was that if you are running a
server against a database you really should be logging what its doing.

Let me know if you would like this option.


> I wonder, could a more specific error message be given when this problem is
> encountered, instead of just "Connection refused"?

The reason that you get this cryptic message is that the server has already
exitted so there is nothing for the client to connect to. I could introduce some
mechanism for attempting to report early errors in server start up by perhaps
writing to /tmp but it may all just get rather arcane.

If you would like to be able to start the server in a truly "read only" way I
can do that easily, I think it should be a configuration file parameter though
so that administrators have to choose it.

What do you think ?

cheers Ed

 ------------------------------------------------------------------------
| Ed Griffiths, Acedb development, Informatics Group,                    |
|               The Sanger Centre, Wellcome Trust Genome Campus,         |
|               Hinxton, Cambridge CB10 1SA, UK                          |
|                                                                        |
| email: edgrif at sanger.ac.uk  Tel: +44-1223-494780  Fax: +44-1223-494919 |
 ------------------------------------------------------------------------





More information about the Acedb mailing list