I have followed the instructions for setting up the nfs service on my MBWEII but it doesn't work. Since there is apparently no rpcinfo installed, I cannot really *check* if the service is running OK. All I get while I try to access it from the client is 'mount: RPC: Timed out'
I can log in to the device via ssh w/o problems and other nfs mounts from my main Ubuntu server work fine, but the MBWE refuses to cooperate :-(
Any ideas?
I get the same problem when trying to mount from my OpenBSD box. Anyone has a clue what is wrong or a way to troubleshoot this?
Need more info about how you installed. Basically you should follow http://mybookworld.wikidot.com/nfs-server
To check if NFS server is running do:
ps auxwwwww | grep rpc
You should see the processes you started with the nfs.sh script.
Check also /etc/exports, it should look like:
/shares/internal/PUBLIC *(rw,no_root_squash,sync)
You can use rpcinfo from your client:
rpcinfo -p your_mybook_IP
If you don't see nfs, reboot the mybook
Installed following those instructions.
Strange, I get different results every time I run the script manually…
this is what I have:
[root@MyBookWorld ~]# ps auxwwwww | grep rpc
root 1015 0.0 0.6 608 188 ? Ss 11:16 0:00 /usr/sbin/rpc.mountd
root 1020 0.0 0.0 0 0 ? S< 11:16 0:00 [rpciod/0]
root 1021 0.0 1.1 560 332 ? Ss 11:16 0:00 /usr/sbin/rpc.statd
root 1025 0.0 0.4 544 124 ? S 11:16 0:00 /usr/sbin/rpc.rquotad
root 1066 0.0 0.9 656 296 pts/0 R+ 11:23 0:00 grep rpc
[root@MyBookWorld ~]#
should the nfsd look like this and not rpc.nfsd?
[root@MyBookWorld ~]# ps auxwwwww | grep nfs
root 1017 0.0 0.0 0 0 ? S 11:16 0:00 [nfsd]
root 1068 0.0 1.0 656 320 pts/0 S+ 11:23 0:00 grep nfs
[root@MyBookWorld ~]#
Copied that exports to look identical.
rpc info from the client:
program vers proto port
100000 2 tcp 111 portmapper
100000 2 udp 111 portmapper
100005 1 udp 767 mountd
100005 1 tcp 770 mountd
100005 2 udp 767 mountd
100005 2 tcp 770 mountd
100005 3 udp 767 mountd
100005 3 tcp 770 mountd
100003 2 udp 2049 nfs
100003 3 udp 2049 nfs
100003 2 tcp 2049 nfs
100003 3 tcp 2049 nfs
100021 1 udp 3075 nlockmgr
100021 3 udp 3075 nlockmgr
100021 4 udp 3075 nlockmgr
100021 1 tcp 4430 nlockmgr
100021 3 tcp 4430 nlockmgr
100021 4 tcp 4430 nlockmgr
100024 1 udp 776 status
100024 1 tcp 779 status
100011 1 udp 778 rquotad
100011 2 udp 778 rquotad
#
:( buhu
mount_nfs: bad MNT RPC: RPC: Timed out
Must have missed something basic or it's the OpenBSD doing something crazy.
Check if the user and group "rpc" exist. The portmap version installed in the mybook requires it, but you have to create the user yourself (I don't remember if this is in the wiki).
The lines should look like:
/etc/passwd:
rpc:x:29:29:RPC Service User:/var/lib/nfs:/sbin/nologin
/etc/group:
rpc:x:29:
Then reboot the mybook (with "reboot"). You should see the same rpc.info that you got.
If you still have a timeout, check your firewall configuration on the client, you may be blocking UDP to/from your MyBook.
Hi,
no, that's not in the wiki. Had some progress after adding that, now I get this in /var/log/messages:
Nov 26 20:09:57 MyBookWorld user.notice syslog: authenticated mount request from 192.168.1.99:987 for /shares/internal/PUBLIC (/shares/internal/PUBLIC)
but still timeout on the client. I think the pf rules are correct but I guess I'll have to look deeper.
Wohoo success!!
I had forgot to add the client in the hosts file on the MBWE. Found out when I did a tcpdump and saw that it asked my dns to resolve my client IP.
Thanks for you help. Hopefully this will help some others as well.
Well, I'm away for a while, so I can't check this out. But it sure sounds very promissing.
I have two question, though.
1. The ids for the user/group rpc don't necessarily have to be 29, have they?
2. Do one really have to include an entry of the client into /etc/hosts for reverse lookup? Or did you mean /etc/hosts.allow
And one additional basic question about the Linux in the box. Where the heck is rpcinfo? Isn't it supposed to be a part of glibc?
If you don't have the client in the hosts-file the MBWE tries to lookup your hostname, if it can't find it it times out after a while, 30 seks? I guess rpc don't like that.
Another symptom of this problem is that it takes a long time to ssh in to the machine, it tries to lookup your hostaname after you type the password but times out after a while and gives you a prompt anyway.
OK, I see. But then, out of curiosity: The name put into /etc/hosts really doesn't matter, right? I could give the machine any name I like as long as I am on my private network and I don't have a central DNS repository? I have allways thought that this nfs stuff works based on numerical ip addresses. Or does nfs use the name-to-number lookup somewhere in-between?
OK, I can confirm that these two modifications are necessary pre-conditions.
My nfs is now fully functional.
These are the commands to add a new rpc group and a new rpc user:
addgroup -g 29 rpc
adduser -h /var/lib/nfs -s /sbin/nologin -G 29 -D -H
The used values are from the example above.
The name and id of the rpc user cannot be set with this command. You have to edit them by hand, if you wish.
The /etc/hosts file has to be edited by hand to look roughly like this one:
127.0.0.1 localhost
192.168.1.2 sol sol.solar-system.net
192.168.1.100 mercury mercury.solar-system.net
192.168.1.101 venus venus.solar-system.net
192.168.1.102 earth earth.solar-system.net
192.168.1.103 mars mars.solar-system.net
Where sol is the name of the MyBook and all the other names are names of computers which want to mount the file system of sol via nfs.
found out that the hosts file is overwritten at every boot as well :( There is another post in the forum explaingin how to get rid of that "feature"
strangely… after rebooting the MyBook i dont need to add it to the hostfile though.. everything seems to work fine now.
Hi, the "adduser" command given above by Z80 to enable NFS is not quite right - the group must be a name, and the username is missing. It should be:
adduser -h /var/lib/nfs -s /sbin/nologin -G rpc -D -H rpc
Further, I found that in /etc/exports, I had to give a real network number rather than 0.0.0.0/24, in my case:
/shares/internal/PUBLIC 192.168.0.0/24(rw,sync)
Many thanks to all for excellent information.
Hello All,
I have a couple of problems with setting up NFS (based on these insturctions http://mybookworld.wikidot.com/nfs-server).
First off, none of these files exist:
start-stop-daemon --start --exec /usr/sbin/rpc.mountd
start-stop-daemon --start --exec /usr/sbin/rpc.nfsd
start-stop-daemon --start --exec /usr/sbin/rpc.statd
start-stop-daemon --start --exec /usr/sbin/rpc.lockd
start-stop-daemon --start --exec /usr/sbin/rpc.rquotadSecondly, I can't edit this file /etc/exports which is just a soft link to this file /var/etc/exports with the following line:
/shares/internal/PUBLIC 192.168.0.0/16(rw,sync,insecure)It won't let me edit the file even as root.
I'm running firmware 2.00.15.
Does anyone have a good guid to setting up NFS?
Please note, NFS is no longer included on newer firmware versions 2.00.15+. You can recover NFS from the backup images in the rescue thread.
see the nfs page: http://mybookworld.wikidot.com/nfs-server
Hi all.
I installed new fw 2.0.18 (my mybook have 2 day of life) and i installed all files for nfs server.
The nfs server on mybook is up and running but when a try to access via my linux box, the message is "RPC timed out".
I read all pages and i not found the solution.
Can you help me?
Tnx!
Just added NFS to my MyBookWorld and had some trouble with file permissions.
I didn't notice this anywhere here and thought this might help out.
My nfs mount was working but I couldn't copy files to directories I just created.
I got things working when I exported with this line.
/shares/internal/STORAGE 192.168.0.0/16(rw,sync,insecure,all_squash,anonuid=33,anongid=33)Hi all,
I just got my MBWE 1TB with firmware 01.00.16 but I have problems setting up nfs. I have read all the instructions, but I have many binaries missing. Firstly my system doesn't have modprobe. Why's that? Also start-stop-daemon is missing…
Hello.. this is my first post.. and would like to say im totally new to unix based systems.
i bought this mybook 500gb NAS over 6 months ago, after my maxtor SS NAS crashed.
when i found out i had to register and use mionet i imminently searched the net for a new firmware, so i could use it as a normal mediaserver / file server.
i had my xbox, ps3 and wifi media player all connected to my NAS and it worked like a dream.
Now im sat with a white box with a pretty blue light as a book end, holding up my books.
can anyone help ??
First i will list my equipment as good as i can.
mybook : 500GB (with blue ring lights) with fw 2.00.19 installed. and upgraded with instructions from this page
http://mybookworld.wikidot.com/firmware-2-00-15
computer : atg (from dell running XP)
programs : on computer running both PuTTY ans WinSCP. can accsess mybook and login with root.
i followed the guide for installing optware and it works ok.
followed the guide for installing NFS server, but no accsess.
when i try to run the nsf,sh script i get the following:
[root@MBMS ~]# /etc/init.d/nfs.sh start
Starting NFS
/sbin/portmap already running.
727
start-stop-daemon: unable to start /usr/sbin/rpc.mountd: Permission denied
start-stop-daemon: unable to start /usr/sbin/rpc.nfsd: Permission denied
start-stop-daemon: unable to start /usr/sbin/rpc.statd: Permission denied
start-stop-daemon: unable to start /usr/sbin/rpc.lockd: Permission denied
start-stop-daemon: unable to start /usr/sbin/rpc.rquotad: Permission denied
/etc/init.d/nfs.sh: line 16: /usr/sbin/exportfs: Permission denied
Please help… and please be very specific as my knowledge of unix commands is limited.
Thanks to all who actually read the whole post, hope someone can help.
Kinds regards tba_
tba666,
Had the same problem. You've copied over the files, but you need to run chmod 755 on each of them so they are executable. Should be good to go after that.
500gg version FW 2.00.19
i've got it all setup following the TUT, set the perms etc and edited the exports file, but when i come to run the nfs.sh i get :
bash: /etc/init.d/nfs.sh: /bin/sh^M: bad interpreter: No such file or directory
anyone got an idea why?
Thanks in advance
fixed it, due to me creating the file in windows then SCP'ing it across it put the ^M at the end of the file, to fix you will need to open it in vi remove the ^M at the end of each line and then use :set fileformat=unix
only error im getting now is cannot register service RPC unable to receive;errno = connection refused
Cheers
