1. FTP: A requestor is asking if the files transferred from a source regardless of the permissions will be "converted" to the permissions as defined in the destination server/location. Which led me to some digging.
So basically, FTP permissions can be set. But not without affecting all users, i.e., globally as this is done via inetd. Or you can check these:
- http://forums11.itrc.hp.com/service/forums/questionanswer.do?threadId=913770&admit=109447626+1217982179357+28353475
- http://forums11.itrc.hp.com/service/forums/questionanswer.do?admit=109447626+1217991636938+28353475&threadId=603250
Sridhar Bhaskarla May 26, 2004 18:47:14 GMT
--------------------------------------------------------------------------------
Hi,
Set the umask in the ftp session itself. For ex.,
ftp>umask 022
ftp>put file1 file2
-Sri
Griselda Sanchez May 26, 2004 18:59:24 GMT
--------------------------------------------------------------------------------
Thanks Sri for your advice, but Is there any way to apply this permanent?
Mel Burslan May 26, 2004 19:04:19 GMT
--------------------------------------------------------------------------------
other than what Sri said, you can not set some ftp sessions' umask to something while keeping the others at something else.
in /etc/inetd.conf, you will find a line similar to this,
-Sri
Griselda Sanchez May 26, 2004 18:59:24 GMT
--------------------------------------------------------------------------------
Thanks Sri for your advice, but Is there any way to apply this permanent?
Mel Burslan May 26, 2004 19:04:19 GMT
--------------------------------------------------------------------------------
other than what Sri said, you can not set some ftp sessions' umask to something while keeping the others at something else.
in /etc/inetd.conf, you will find a line similar to this,
ftp stream tcp nowait root /usr/lbin/ftpd ftpd -L 002
where the 002 at the tail end, is the default umask for files put to the server via ftp. you can change this value to set the umask but it will effect every user.
Michael Tully May 26, 2004 19:06:46 GMT
--------------------------------------------------------------------------------
There is no way to set it permanently, other than enforcing these things from within a script wrapper.
Sridhar Bhaskarla May 26, 2004 19:26:20 GMT
--------------------------------------------------------------------------------
Hi,
As far as I know, you can do it either for all the users by configuring ftpd in inetd.conf with -u option or for a directory using ftpaccess file.
Look at 'man ftpaccess' on directory level permissions. Search for "upload".
-Sri
Bill Hassell May 26, 2004 22:15:26 GMT
--------------------------------------------------------------------------------
There is no way to change this for a single user. ftpaccess can allow or disallow the use of umask, but not preset the mask. You'll need to script the ftp transfer and includxe the required permission changes in your script. This is easy in Unix, a real pain with Macs and PCs, especially if the end user insists on using a fancy GUI interface.
But, I guess, it would much better to set things within the source thru umask [set it via .profile].
2. IP address [SC#21810036]: This has to be done on a Solaris box. An IP - virtual - (although two were requested), needs to be "un-bind" from the NIC. Here are the some of the info I got:
[root@server:/root]
# ifconfig -a ["pipe"] grep -e 192.44.122.154 -e 192.44.122.226
....
hme0:26: flags=1000842 mtu 1500 index 2
inet 192.44.122.226 netmask fffffe00 broadcast 192.44.123.255
....
[root@server4013:/root]
# ping 192.44.122.226
no answer from 192.44.122.226
[root@server4013:/root]
# netstat -r
Routing Table: IPv4
Destination Gateway Flags Ref Use Interface
-------------------- -------------------- ----- ----- ------ ---------
192.44.123.0 server1.com U 1 84389 hme0:4
6192.44.123.0 server2.com U 1 0 hme0:47
192.44.123.0 server3.com U 1 0 hme0:48
192.44.123.0 server4.com U 1 0 hme0:49
192.44.123.0 192.44.123.79 U 1 0 hme0:50
192.44.123.0 server5.com U 1 0 hme0:51
192.44.122.0 server4013.com U 1 940909 hme0
.............
default 192.44.122.4 UG 11260332localhost localhost UH 417021843 lo0
Routing Table: IPv6
Destination/Mask Gateway Flags Ref Use If
--------------------------- --------------------------- ----- --- ------ -----
fe80::/10 fe80::a00:20ff:fef9:81d4 U 1 0 hme0
ff00::/8 fe80::a00:20ff:fef9:81d4 U 1 0 hme0
default fe80::a00:20ff:fef9:81d4 U 1 0 hme0
localhost localhost UH 1 1736 lo0
[root@server4013:/root]
#
I also got some info from this site: http://www.tek-tips.com/faqs.cfm?fid=4083
But how?!?
jephilc:
Hi
1) There are two ways to remove logical interfaces...
Either do this for each one you want to remove (as root)
# ifconfig hme0:1 down unplumb
Michael Tully May 26, 2004 19:06:46 GMT
--------------------------------------------------------------------------------
There is no way to set it permanently, other than enforcing these things from within a script wrapper.
Sridhar Bhaskarla May 26, 2004 19:26:20 GMT
--------------------------------------------------------------------------------
Hi,
As far as I know, you can do it either for all the users by configuring ftpd in inetd.conf with -u option or for a directory using ftpaccess file.
Look at 'man ftpaccess' on directory level permissions. Search for "upload".
-Sri
Bill Hassell May 26, 2004 22:15:26 GMT
--------------------------------------------------------------------------------
There is no way to change this for a single user. ftpaccess can allow or disallow the use of umask, but not preset the mask. You'll need to script the ftp transfer and includxe the required permission changes in your script. This is easy in Unix, a real pain with Macs and PCs, especially if the end user insists on using a fancy GUI interface.
But, I guess, it would much better to set things within the source thru umask [set it via .profile].
2. IP address [SC#21810036]: This has to be done on a Solaris box. An IP - virtual - (although two were requested), needs to be "un-bind" from the NIC. Here are the some of the info I got:
[root@server:/root]
# ifconfig -a ["pipe"] grep -e 192.44.122.154 -e 192.44.122.226
....
hme0:26: flags=1000842
inet 192.44.122.226 netmask fffffe00 broadcast 192.44.123.255
....
[root@server4013:/root]
# ping 192.44.122.226
no answer from 192.44.122.226
[root@server4013:/root]
# netstat -r
Routing Table: IPv4
Destination Gateway Flags Ref Use Interface
-------------------- -------------------- ----- ----- ------ ---------
192.44.123.0 server1.com U 1 84389 hme0:4
6192.44.123.0 server2.com U 1 0 hme0:47
192.44.123.0 server3.com U 1 0 hme0:48
192.44.123.0 server4.com U 1 0 hme0:49
192.44.123.0 192.44.123.79 U 1 0 hme0:50
192.44.123.0 server5.com U 1 0 hme0:51
192.44.122.0 server4013.com U 1 940909 hme0
.............
default 192.44.122.4 UG 11260332localhost localhost UH 417021843 lo0
Routing Table: IPv6
Destination/Mask Gateway Flags Ref Use If
--------------------------- --------------------------- ----- --- ------ -----
fe80::/10 fe80::a00:20ff:fef9:81d4 U 1 0 hme0
ff00::/8 fe80::a00:20ff:fef9:81d4 U 1 0 hme0
default fe80::a00:20ff:fef9:81d4 U 1 0 hme0
localhost localhost UH 1 1736 lo0
[root@server4013:/root]
#
I also got some info from this site: http://www.tek-tips.com/faqs.cfm?fid=4083
But how?!?
The answer: It was a futile effort. Nothing much to read about.
"only one Interface was there : 192.44.122.226
removed the interface"
What can get from this?!? SO, I have to know this by hook or by crook or by research. And I found this: http://www.experts-exchange.com/OS/Unix/Solaris/Q_21713340.html
jephilc:
Hi
1) There are two ways to remove logical interfaces...
Either do this for each one you want to remove (as root)
# ifconfig hme0:1 down unplumb
or if you want to specify the address you can do this:
# ifconfig hme0 removeif 172.16.17.241
[ The above shows two ways of removing the 241 address.
2) To see what type of network interface you've got for example, for an ultra system run this as root
# /usr/platform/sun4u/sbin/prtdiag -v
You'll see a line like this for an hme interface
0 PCI-1 33 1 network-SUNW,hme
Check dmesg for messages after a boot which will tell you the speed the interface initialized and whether it was full or half duplex etc.
3) Use dmesg or look in /var/adm/messages to tell you at what speed and duplex setting the interface is running at
4) To hardcode the interface at a specific speed and duplex, you use the ndd command as root, but most of the time the interface will probably be set to autonegotiate. The autonegotiate facility is extremely useful especially if you have legacy systems on your network that might be running at 10 Mbps as your system will automatically talk to them OK. If you turn off this facility, and you are running legacy network interfaces, you could lose connectivity to them or cause errors, so it would be wise to check this first.
First check whether the general autoneg capability is set to 1 by running
# ndd /dev/hme adv_autoneg_cap
You'll see a line like this for an hme interface
0 PCI-1 33 1 network-SUNW,hme
Check dmesg for messages after a boot which will tell you the speed the interface initialized and whether it was full or half duplex etc.
3) Use dmesg or look in /var/adm/messages to tell you at what speed and duplex setting the interface is running at
4) To hardcode the interface at a specific speed and duplex, you use the ndd command as root, but most of the time the interface will probably be set to autonegotiate. The autonegotiate facility is extremely useful especially if you have legacy systems on your network that might be running at 10 Mbps as your system will automatically talk to them OK. If you turn off this facility, and you are running legacy network interfaces, you could lose connectivity to them or cause errors, so it would be wise to check this first.
First check whether the general autoneg capability is set to 1 by running
# ndd /dev/hme adv_autoneg_cap
If it's set to 1 then you need to turn it off by doing:
# ndd -set /dev/hme adv_autoneg_cap 0
# ndd -set /dev/hme adv_autoneg_cap 0
then if you want to manually set it to 100 full duplex run this:
# ndd -set /dev/hme adv_100fdx_cap 1
# ndd -set /dev/hme adv_100fdx_cap 1
and make sure the other settings are set to 0. See the other available settings by listing the possible variables with this command:
# ndd /dev/hme \?
The ones marked (read and write) are the ones you can set.
Good luck
John
01.28.2006 at 02:45PM PST, ID: 15814641
jephilc:
Hi again
Apologies but I forgot to mention the /etc/hostname.hme0 file.
If the entries had been made permanent for the logical interfaces, then you will need to edit this file to remove the relevant interface entries. This will be needed so that the interfaces are permanently removed and do not re-appear when you reboot. In my original posting, I was assuming that you wanted to do the operation on the fly.
If rebooting isn't a problem, then all you really need to do in order to remove the interfaces is to edit /etc/hostname.hme0, remove the entries and reboot.
Hope this helps
John
Well, I hope did...
# ndd /dev/hme \?
The ones marked (read and write) are the ones you can set.
Good luck
John
01.28.2006 at 02:45PM PST, ID: 15814641
jephilc:
Hi again
Apologies but I forgot to mention the /etc/hostname.hme0 file.
If the entries had been made permanent for the logical interfaces, then you will need to edit this file to remove the relevant interface entries. This will be needed so that the interfaces are permanently removed and do not re-appear when you reboot. In my original posting, I was assuming that you wanted to do the operation on the fly.
If rebooting isn't a problem, then all you really need to do in order to remove the interfaces is to edit /etc/hostname.hme0, remove the entries and reboot.
Hope this helps
John
Well, I hope did...
No comments:
Post a Comment