Hot answers tagged nfs-server
12
sync and async have different meanings for the two different situations.
sync in the client context makes all writes to the file be committed to the server.
async causes all writes to the file to not be transmitted to the server immediately, usually only when the file is closed. So another host opening the same file is not going to see the changes made by ...
5
The issue you're highlighting is not a flaw, it's a feature! Yes it's a feature of the SSH protocol and giving that improperly configured SSH service can be used for a wide range of exploits like:
Bypassing host firewalls
Bypassing IP restrictions
Remotely accessing services which listens only on 'localhost' interfaces [mysql?]
As an example of common ...
4
That's a very old document your looking at it talks about kernel version 2.4 which came out in 2001, a lot have changes have happened in the last 12 years. Although some things remain the same.
I only have CentOS 6.x boxes to play with which uses nfsv4 by default. To allow the connection via an intermediate machine I had to export the filesystem with ...
4
Did you issue the command using sudo so you have elevated privileges?
Rsync might also work better for something like that.
I'd caution against considering it a backup, as it's just a copy. Backups protect against corruption issues or deleted files. Making copies is more like protecting availability of resources.
4
yum -y install nfs-utils portmap
Prepare and modify /etc/exports for sharing files, that would similar to as shown below
/home/NFS-files
192.168.100.0/24(ro,sync)
/home/NFS-share */26(rw,sync)
/ISO 192.168.100.0/24(ro,sync)
service portmap start
service nfs start
3
Mac OS X 10.4 Tiger uses NetInfo database, which was discontinued in 10.5 Leopard. For 10.4:
nicl . -delete 116 clients 172.18.150.130
nicl . -delete 116 clients 172.18.150.138
nicl . -append 116 clients 172.18.150.132
nicl . -append 116 clients 172.18.150.134
Then restart the NFS server.
2
It's more of a hindsight suggestion, but in general you should place snapshots of directory dir residing in a parentdir in parentdir or in subdirectory of parentdir, not in dir. So that you'd have directory hierarchy similar to this:
parentdir
├── dir
│ ├── Apps
│ └── Clients
└── snapshots
├── @GMT-2012.01.30-13.00.01
│ ├── Apps
│ └── ...
2
There's been some work recently to speed up directory access, which will be in Cygwin 1.7.8. That's expected to be released some time in January, but you could try the latest snapshots at http://www.cygwin.com/snapshots.
Also, you might want to avoid symlinks in the path, as they cause additional overhead. You could mount the directories in question ...
2
You can try using umount -f to forcibly unmount. Depending on disk activity, it hasn't always worked for me; sometimes I've had to cycle the machine (unless I could restore the service).
For the future, you can look into whether mount_nfs arguments -i (interruptible), -s (soft mount), or -Rn (maximum retry count) are appropriate for your environment.
I ...
2
I found the answer to this here: http://sophiedogg.com/?p=141
/var/lib/nfs/statd/sm/ - directory containing statd monitor list
/var/lib/nfs/statd/sm.bak/ - directory containing statd notify list
Before removing these files, you should stop the rpcbind, statd, and lockd services. Below is a list of commands to run to fix this issue on a RPM based ...
2
Large MSDOS filesystems are not supported as NFS exports on FreeBSD due to some implementation limitations. You could try if you can successfully mount the FS without the -o large option, but if that fails, you will be out of luck with NFS (but maybe you could try Samba instead).
2
No, this is not possible with a single NFS export on kernel based nfs server.
As exports(5) manual says:
The file /etc/exports contains a table of local physical file systems
on an NFS server that are accessible to NFS clients.
However you can export parts of the tree and reassemble them on the clients with bind.
Symbolic links
Another palliative ...
1
The spammy log messages you see are coming from the ocf:heartbeat:exportfs resource agent. They appear every 30 seconds, which corresponds to the monitoring interval you specified in the exportfs primitive definitions. The resource agent is a bit too verbose, IMHO, but this should not be a problem. Just make sure you logrotate often enough that the logs ...
1
It's clear from debugging a bit that Ubuntu's using rpc.idmap as shipped (up to 12.04, at least).
The archaeology: You need a kernel with the nfsidmap mechanism compiled in (the option is 'new' idmap - so I assume it's the replacement of rpc.idmap, therefore) to get the nfsidmap calls to work. Also, ...
1
That document is rather old (2006!). In the absence of better security mechanisms (i.e. NFSv4 + GSS) adding a host to exports means you implicitly trust that host, its users and processes.
SSH port-forwarding isn't your only problem, you can disallow it (sshd's AllowTcpForwarding no), but as sshd_config(5) says
Note that disabling TCP forwarding does ...
1
You need to use dtrace, for that level of introspection. Here's basically the nfsv3fileio.d example from https://wikis.oracle.com/display/DTrace/nfsv3+Provider, but updated to run live (I think, my testing was minimal). That same page has a few more examples. You might also Google for 'nfssvrtop'.
#!/usr/sbin/dtrace -s
#pragma D option quiet
...
1
That's quite normal, you are running a user mode NFS server on top of
a Unix layer on top of another operating system.
Using a GNU/Linux Server would be much better.
Take a look in the event viewer to see if there are any permissions issues and in the case set the permissions of the mount accordingly, like:
/disk3 (rw,all_squash,anonuid=xxx,anongid=yyy)
1
the common placement of NFS exports is outside any os controlled areas ( /usr /etc /var ) usually in another area called /exports
If you are automounting home directories under /home this will change your placement.
Usage of the showmount -e to verify is very helpful.
Enjoy.
fe007
1
the only thing I can tell you is that we have some NFS servers running lenny without any problems at all. The hardware is old (proliants g4, if I recall correctly, I am not at work now) but the disks are fast scsi 15krpm, so that's what basically matters.
Debugging nfs problems can be frustrating. I found that this helps you get more info from the logs:
* ...
1
Your permission errors may stem from the fact that the machine you were on mounted the NFS filesystem as your userid rather than root, meaning that you weren't seeing the entire filesystem. Or, possibly the NFS mounts are changing the UID on purpose for different groups.
rsync -aplxo --delete /nfsmount/ /newmountlocation/
Is probably the best way to do ...
1
Check your /proc/mounts or /etc/mtab; most likely, you're mounting it ro (as opposed to rw) on the client side. Ah, wait, no! Files are being created read-only? That's a umask issue, either on the server or client ends. Check shell umask and mount umask on the client end, and shell, daemon, mount, and exported umask on the server. If you're creating ...
1
On your server, you are remounting your /home directory inside /share/home? That's...unusual. I'm betting NFS has shared your local /shared/home directory, instead of the remounted one from /home. You can probably test this by unmounting /share/home on the server, and see if the files you've been creating on the client are there.
Restarting the NFS ...
1
rtacconi's answer is dated and would apply for 2003, not 2008.
Windows 2008 R2 has a stock NFS server. NFS Basically just acts as another protocol. If you install the role, NFS is a tab under SHARING.
We use it to serve up over 500K files to Centos 5.5 boxes in our web cluster.
Seems to work pretty well.
1
On scaling up the ebs volume and it's data will not be "cloned" , to have this behavior you'd want to automate it at boot.
1) grab the latest snapshot of WS-1 ebs volume
2) create and attache the volume.
Another method, depending on how much data is on the ebs is to pull it down from s3.
With the security group, you can allow any server in the ...
1
lookup(program) means automount thinks your map is an executable, does it have the executable bit set?
lookup(file) is what you should be seeing
See http://www.squarebox.co.uk/cgi-squarebox/manServer/automount.8.
1
As for the "hanging" of the Monit process during NFS server faults, this can be circumvented by two methods.
You change the NFS mount options from hard to soft, which causes the NFS layer to issue an I/O error to the accessing application after retrans retries. As this can introduce other problems with respect to data integrity (your writing applications ...
1
The general approach would be (assuming none of the Monit built-in rules are applicable)
Find out how you would do the checks manually
Write shell scripts performing these checks, returning 0 for 'success' and 1 for 'failure'
Let Monit test those scripts (example is from official documentation):
check program myscript with path ...
1
so it seems it was the wildcard exports, which, if you read the man page, are not recommended. I'd read that before but for some reason didn't fix it. I still think this is a "bug" and it should work in theory but in practice, it doesn't.
Hope this helps others.
Example of my new exports file:
/nfs server1.a2hosting.com(fsid=0,ro,nohide,no_root_squash)
...
1
Great questions, highlights a bigger point with the documentation IMO. Here is an attempt at a complete answer:
What does "subsequent exports on that line only" mean?
An example's probably easiest here:
/export/stuff -rw 10.0.0.54 10.0.0.55
is equivalent to:
/export/stuff 10.0.0.54(rw) 10.0.0.55(rw)
Is fsid=0 not required anymore?
This depends on ...
1
Read the error message again. It says "euid is not dir owner" which means "the user with the id 5000 is not the owner of the directory".
Correct that. Authentication is not the problem here. Authorization is.
On the other hand: Don't do NFS that way with Dovecot! See http://wiki2.dovecot.org/NFS for details. And then NFS is no native service on Windows, ...
Only top voted, non community-wiki answers of a minimum length are eligible
