confirm bug still exists?

James Harrison oscartheduck at gmail.com
Wed Dec 30 08:08:15 UTC 2009


http://www.freebsd.org/cgi/query-pr.cgi?pr=112818&cat=

I have an NFS server, /etc/exports reads:

[root at centos ~]# cat /etc/exports 
/home/james/portsold	172.16.128.133(rw)


I mount it on my FreeBSD box:

[root at server /usr]# mv ports portsold
[root at server /usr]# mkdir ports
[root at server /usr]# mount 172.16.128.137:/home/james/portsold ports
[root at server /usr]# ls -l | grep ports
drwxr-xr-x  69 500   500        4096 Dec 30 07:22 ports

On the FreeBSD server:

drwxr-xr-x 69 james james 4096 Dec 29  2009 portsold


Using portupgrade-devel:

[root at server /usr]# portupgrade -a
[Updating the portsdb <format:bdb_btree> in /var/db/pkg ... - 21055 port entries found .........1000.........2000.........3000.........4000.........5000.........6000.........7000.........8000.........9000.........10000.........11000.........12000.........13000.........14000.........15000.........16000.........17000.........18000.........19000.........20000.........21000 ..... done]
--->  Upgrading 'mysql-client-5.0.88' to 'mysql-client-5.0.89' (databases/mysql50-client)
--->  Building '/usr/ports/databases/mysql50-client'


...etc, etc, etc.

The /usr/ports directory isn't writeable:

[root at server /]# portsnap fetch update
Looking up us.portsnap.FreeBSD.org mirrors... 2 mirrors found.
Fetching snapshot tag from portsnap1.FreeBSD.org... done.
Fetching snapshot metadata... done.
Updating from Tue Dec 29 22:20:49 UTC 2009 to Wed Dec 30 05:02:19 UTC 2009.
Fetching 4 metadata patches... done.
Applying metadata patches... done.
Fetching 0 metadata files... done.
Fetching 2 patches.. done.
Applying patches... done.
Fetching 0 new ports or files... done.
portsnap: Directory is not writable: /usr/ports
[root at server /]# cd usr/ports
[root at server /usr/ports]# touch something
touch: something: Permission denied


However, it's true that I did do a little hinkying around to get stuff to partially build with an NFS mounted directory.

Anybody able to confirm that the bug still exists as described in the PR?


James


More information about the freebsd-ruby mailing list