On Tuesday 08 June 2004 16:14, Mike wrote:

But did you try writing to it. I had the same experience as you. Plug it in and it is automagically recognized. Remove it, and it's gone. Then I tried to write a file to it. Nope, can't do it. Tried to change permissions. Not allowed. Heck, not even root would do it.

I've seen that behavior. USB-stick that was not writeable in Linux, but it was in Windows, the reason for that... It had been formated with NTFS... Just run and do a quick mkreiserfs on it and complain about it not being read- or writeable in Windows (then it is Linux that damaged it, normal logic there)

Nice to look at, but totally useless.

You could read from it? regards j -- Jonas Helgi Palsson

Mike wrote:

Nope.. It WAS writeable in linux and winders. Guess again. It's formatted with fat NOT ntfs. Has been since I bought it. Worked fine in 9.0, and for that matter still does. Can't even change permissions as root.

You are trying to change permissions on a FAT volume? FAT doesn't have permissions. What you see is something that linux puts there depending on how you mounted the volume. Regarding USB sticks, I've seen that they come formatted in a strange setup, from the vendor. Like they have 4 partitions, yet windows displays one. I don't think it's useful to know what it really does, I don't care. I have erased all partitions, and created the FAT filesystem from linux, with one FAT copy only, to save some KB. Become root. To erase the whole partition table on the stick: dd if=/dev/zero of=/dev/look_at_the_device_name bs=512 count=1 To look_at_the_device_name you can type "mount" after SUSE has mounted it and see the line with /media/usb-storage-odd-.... Device name can be /dev/sda, for example, ONLY if you don't have any SCSI disk, or /dev/sdX. After you wipe the partition table, I think you'd better take it out and insert it again. To create one FAT partition: fdisk /dev/sda "n" of type "e W95 FAT16 (LBA)" if you have a stick >32 M or "4 FAT16 <32M". I think that for sticks larger that 512M you need "c W95 FAT32 (LBA)" Write changes with "w" Create the FAT filesystem like this: mkdosfs -f 1 /dev/sda1 -f 1 means only one copy of FAT. Flash-sticks are reliable, they are not diskettes, so having only one copy of FAT saves some KB. Other useful parameters: -F 32 for sticks larger that 512M -s and -S to save space (less slack) it's better to have a small sector size (512 bytes) and a small number of sectors per cluster. Fragmentation is not something you worry about on a stick. Space for files is allocated in clusters, that's the DOS term (blocks in Linux). Because files are not exact multiples of cluster size, there is unused space in the last cluster. That's called slack space and you want to reduce that, by having small clusters. However, the number of clusters is limited, so you need to test starting with -s 2 -S 512 -s 4 -S 512 -s 8 -S 512 Until you don't get an error. You will have a stick that's OPTIMALLY formatted for space and can be used with Linux and Windows without any problem.

Thanks Silviu. It's work for me. On Sat, 2004-06-12 at 03:44, Mike wrote:

On Friday 11 June 2004 11:41, Silviu Marin-Caea wrote:

...

Mike wrote:

...

Nope.. It WAS writeable in linux and winders. Guess again. It's formatted with fat NOT ntfs. Has been since I bought it. Worked fine in 9.0, and for that matter still does. Can't even change permissions as root.

You are trying to change permissions on a FAT volume? FAT doesn't have permissions. What you see is something that linux puts there depending on how you mounted the volume.

The drive is automounted via subfs. I'm trying to get SUSE 9.1 to allow me to write to the drive. It won't do it. It tells me that writing to the drive isn't supported. What a crock. Try to change permissions on the drive, via a right click on properties, and it won't allow it. Can't change permissions, can't even do it as root. Switching to another console logged in as root, and it won't let me chmod it, or chown it. Writing to the usb stick isn't allowed.

...

Regarding USB sticks, I've seen that they come formatted in a strange setup, from the vendor.

I've had the drive for over a year. Same format, 1 partition. Guess again. Doesn't need to be formatted.

I guess it's not going to happen until enough folks have problems with it. I've already sent in feedback to SUSE, but other than the auto-response message, I've not heard or seen anything.

Mike

-- Powered by SuSE 9.1 Kernel 2.6.4 KDE 3.2.1 Kmail 1.6.2 For SuSE Mondo/Mindi backup support go to http://www.mikenjane.net/~mike 9:36pm up 6 days 13:58, 4 users, load average: 1.01, 1.10, 1.14 -- Jason Lim Asia-Pacific Broadcasting Union

Miguel Afonso Oliveira wrote:

Dear All,

I've been getting the messages:

kernel: svc: unknown procedure (6)

on my server and clients. Does any one know what this is and how to solve it?

All the best,

MAO

There used to be a message like that associated with NFS quite a while ago. It would be interested to know what version/s of SuSE you are running. If you are trying to use nfs and what "chkconfig nfs", chkconfig nfslock" and "chkconfig nfsserver" say. Any relevant stuff reported in /var/log/messages would also help and the output of "rpcinfo -p" on each box, also "rpcinfo -p " from server and clients. e.g barrabas:/ftp/jun04 # rpcinfo -p program vers proto port 100000 2 tcp 111 portmapper 100000 2 udp 111 portmapper [Just portmapper is running on this side] [Below, NFS, NIS etc. are running on host bumble] barrabas:/ftp/jun04 # rpcinfo -p bumble program vers proto port 100000 2 tcp 111 portmapper 100000 2 udp 111 portmapper 100024 1 udp 610 status 100024 1 tcp 613 status 100004 2 udp 635 ypserv 100004 1 udp 635 ypserv 100004 2 tcp 638 ypserv 100004 1 tcp 638 ypserv 600100069 1 udp 993 fypxfrd 600100069 1 tcp 995 fypxfrd 391002 2 tcp 32768 sgi_fam 100003 2 udp 2049 nfs 100003 3 udp 2049 nfs 100003 4 udp 2049 nfs 100003 2 tcp 2049 nfs 100003 3 tcp 2049 nfs 100003 4 tcp 2049 nfs 100021 1 udp 32770 nlockmgr 100021 3 udp 32770 nlockmgr 100021 4 udp 32770 nlockmgr 100021 1 tcp 32769 nlockmgr 100021 3 tcp 32769 nlockmgr 100021 4 tcp 32769 nlockmgr 100005 1 udp 649 mountd 100005 1 tcp 652 mountd 100005 2 udp 649 mountd 100005 2 tcp 652 mountd 100005 3 udp 649 mountd 100005 3 tcp 652 mountd 100009 1 udp 712 yppasswdd Regards Sid. -- Sid Boyce .... Hamradio G3VBV and keen Flyer ===== LINUX ONLY USED HERE =====

Miguel Afonso Oliveira wrote:

Sid Boyce wrote:

...

Miguel Afonso Oliveira wrote:

...

Dear All,

I've been getting the messages:

kernel: svc: unknown procedure (6)

on my server and clients. Does any one know what this is and how to solve it?

All the best,

MAO

Dear Sid,

I'm running SuSE Linux 9.1 on every machine. I know this has something to do with nfs and rpc but what exactly I really don't know. I'm running nfs-utils 1.0.6 and output of rpcinfo on the clients is:

program vers proto port 100000 2 tcp 111 portmapper 100000 2 udp 111 portmapper 100021 1 udp 4000 nlockmgr 100021 3 udp 4000 nlockmgr 100021 4 udp 4000 nlockmgr 100024 1 udp 4000 status 100021 1 tcp 4000 nlockmgr 100021 3 tcp 4000 nlockmgr 100021 4 tcp 4000 nlockmgr 100024 1 tcp 4000 status 100007 2 udp 4004 ypbind 100007 1 udp 4004 ypbind 100007 2 tcp 4004 ypbind 100007 1 tcp 4004 ypbind

while on the server is:

program vers proto port 100000 2 tcp 111 portmapper 100000 2 udp 111 portmapper 100004 2 udp 834 ypserv 100004 1 udp 834 ypserv 100004 2 tcp 834 ypserv 100004 1 tcp 834 ypserv 100009 1 udp 835 yppasswdd 100003 2 udp 4001 nfs 100003 3 udp 4001 nfs 100003 4 udp 4001 nfs 100227 3 udp 4001 nfs_acl 100003 2 tcp 4001 nfs 100003 3 tcp 4001 nfs 100003 4 tcp 4001 nfs 100227 3 tcp 4001 nfs_acl 100005 1 udp 4002 mountd 100005 1 tcp 4002 mountd 100005 2 udp 4002 mountd 100005 2 tcp 4002 mountd 100005 3 udp 4002 mountd 100005 3 tcp 4002 mountd 100021 1 udp 4000 nlockmgr 100021 3 udp 4000 nlockmgr 100021 4 udp 4000 nlockmgr 100024 1 udp 4000 status 100021 1 tcp 4000 nlockmgr 100021 3 tcp 4000 nlockmgr 100021 4 tcp 4000 nlockmgr 100024 1 tcp 4000 status 100011 1 udp 41743 rquotad 100011 2 udp 41743 rquotad 100011 1 tcp 41743 rquotad 100011 2 tcp 41743 rquotad

I also get the message:

ypbind[----]: Unknown signal: 0

on the clients from time to time. On the server I also get the very disconcerting message:

ypserv[---]: refused connect from 127.0.0.1:836 to procedure ypproc_clear (,;0)

from time to time. I've checked time and time again but ypserv is allowed from hosts.allow.....

All the best,

MAO

For NFS, presumably "showmount -e <hostname>" from one box to another yields the correct results and you can nfs mount those. For NIS, check /etc/nsswitch.conf to make sure nis is against the relevant entries ... one of my boxes (Mandrake 10.0) shows passwd: files nisplus nis shadow: files nisplus nis group: files nisplus nis #hosts: db files nisplus nis dns hosts: files nisplus nis dns # Example - obey only what nisplus tells us... #services: nisplus [NOTFOUND=return] files #networks: nisplus [NOTFOUND=return] files #protocols: nisplus [NOTFOUND=return] files #rpc: nisplus [NOTFOUND=return] files #ethers: nisplus [NOTFOUND=return] files #netmasks: nisplus [NOTFOUND=return] files bootparams: nisplus [NOTFOUND=return] files ethers: files netmasks: files networks: files protocols: files rpc: files services: files netgroup: nisplus publickey: nisplus automount: files nisplus Something is mis-configured and not allowing nis cache to be cleared. I haven't got around to setting up nisplus on my other boxes so far after new installs and upgrades, but I used to have it working across 3 of them, 2 set as clients and slave servers. Regards Sid. -- Sid Boyce .... Hamradio G3VBV and keen Flyer ===== LINUX ONLY USED HERE =====