[torquedev] reverting 'hostname -f' change in torque.setup from September

Garrick Staples garrick at usc.edu
Fri Nov 30 12:18:10 MST 2007


> The command 'set server operators += bpendleton at server' will fail in the  
> torque.setup script.  It appears that when torque checks the ACL, it will  
> only check against the first column of the host in /etc/hosts, causing the  
> check to fail even though it should succeed.  CRI support ran into this  

Yes, it's a well known problem regarding matching forward and reverse DNS.

Note that nothing in torque reads /etc/hosts and compares columns.


> issue with customers fairly regularly, so a change was made to the setup  
> script which fixed the issue.  If hostname -f isn't portable, it should  
> removed, but a fix still needs to be implemented so that customers don't  
> run into this issue.  A better fix might be to fix the ACL check so that  
> it accepts all columns of /etc/hosts as valid.

It is definitely not portable.  It really breaks people's non-GNU machines.

I removed it from torque.setup a long time ago and was surprised that it was
added back again.


> Please let me know if I can provide you with more information or be of any  
> assistance on this issue.

If you can come up with a portable way to determine matching forward and
reverse DNS in a shell script, that would be great :)

-------------- next part --------------
A non-text attachment was scrubbed...
Name: not available
Type: application/pgp-signature
Size: 189 bytes
Desc: not available
Url : http://www.supercluster.org/pipermail/torquedev/attachments/20071130/8df4a64f/attachment.bin


More information about the torquedev mailing list