httpd-dev mailing list archives

Site index · List index
Message view « Date » · « Thread »
Top « Date » · « Thread »
From "Dirk.vanGulik" <Dirk.vanGu...@jrc.it>
Subject Re: getwd stuff
Date Thu, 20 Feb 1997 08:17:45 GMT
Ok, I've now tried and tested it. On Solaris and FreeBSD absolutely
no problem whatso-ever (on the NFS dismount case I was refering to)
but I am still trying to track down the trouble on an OSF/1 machine
with nfs/amd.

In principle you are perfectly right; it is a short lived situation;
but a lot of mounting/dismounting by the amd (Funily enough AFS does not
seem to give the hassle) I still get occasional hangs of 30seconds
before an NFS timeout apprears. Even though the httpd has no right
to be in that directory at all.

But I guess the summary is that I would gove it a +0; not really in
favour but certainly not strongly against it. Anyone who relies on
nfs and  amd's under his webserver should check the logs for excessive
timeouts.

Dw.

> 
> I've already given my +1, but +1 again.
> 
> I do not think that Dirk.vanGulik's concern applies in this case because
> it is used in a short-lived situation and the cwd is reset for each
> SSI request, but I wouldn't mind him verifying that I understood his
> concern...
> 
> On Wed, 19 Feb 1997, Jim Jagielski wrote:
> 
> > I'm leaning towards removing the getwd() call and having the stat
> > simply look at the file, relatively...
> > 
> > Votes please... So far there was one concern
> > -- 
> > ====================================================================
> >       Jim Jagielski            |       jaguNET Access Services
> >      jim@jaguNET.com           |       http://www.jaguNET.com/
> >                   "Not the Craw... the CRAW!"
> > 
> 
> 


Mime
View raw message