Tag Info

New answers tagged

0

Thanks for the help @Fox and @Sven, i have now solved the problem. It was a nfs-kernel-server setting, /etc/defaults/nfs-kernel-server contained the option --manage-gids that breaks the usage of secondary groups. So while the mysql user had the right permissions through a secondary group the permissions on the nfs-server's side were wrong. Hope someone ...


1

You don't state what engine are you using, but let me presume it's InnoDB (as it's pretty much standard these days) then in MySQL Docs (Using actual symbolic links has never been supported for InnoDB tables.) and The DATA DIRECTORY clause is a supported alternative to using symbolic links, which has always been problematic and was never supported ...


2

Try to use this Git setting, which makes a huge performance difference on NFS shares: git config core.preloadindex true


0

patched within ZFS on Linux: https://github.com/zfsonlinux/zfs/pull/3404 get the git version and compile it: git clone https://github.com/zfsonlinux/zfs.git


0

This is meant to be a comment on janneb's answer above, but I don't have the !"#$% reputation: The manual page was out of date for a long time, but has since been updated to say (emphasis mine): In Linux kernels up to 2.6.11, flock() does not lock files over NFS (i.e., the scope of locks was limited to the local system). Instead, one could use ...


4

From a clients point of view, if the failower is implemented good enough, it will only seem as if the NFS server were gone for a very short time and then came back. However, to avoid loosing data you might want to mount with the "sync" option. One way to look if a process is waiting for io is to do "ps aux" and look for "D" in the "STAT" column. Some common ...


0

Following your approach you need to take in mind to stop the nfs service when you perform the failover, so the steps should be: Stop NFS service on node2 Umount the NFS drive on node2 Float the service IP address from node2 to node1 Mount the NFS drive on node1 Start the NFS service on node1 You have other alternatives, for example you can use a cluster ...


1

Looks like a job for NFS. You can make server01 an NFS server that exports the /total_app directory. Then you can allow all servers to mount it. You must only be sure that server01 is always up, running, and accessible. Also be careful that the servers are not 'destroying' each other configurations for the applications in the directory.


3

What you're likely looking for is called a Distributed File System. These require special handling and are not part of normal distro installations. There is a Wikipedia article containing a non-exhaustive list of distributed file system platforms. The main selling point of distributed file systems is redundancy, scalability, and access transparency. Like ...


5

What you want do do can be achieved precisely with overlayfs. With it you can take your /full dir and another /notfull dir and mount them at /third-dir. Reads will seek through the /notfull into the /full and writes will go to a third dir. Have a look at the docs, remount the full dir away from the target name and then use overlayfs to recreate the ...


2

You won't be able to expand the space by mounting them both on /folder if that's what you want. If you want to create another directory like /folder/newshare like you said, you should be able to mount another NFS share there to make it seem like there's more space under /folder. I just confirmed that I could create a mount point for a NFS share inside ...


1

First I would use sshfs. Second up to NFSv3 on Linux NFS is insecure. You can either build a private NFS network or use secure NFSv4.


0

Given the complexity of setting up a replicated NFS server, we're opting to go with S3. The performance of s3fs-fuze was terrible (doing an ls on a directory with over 1,000 files would take close to a minute due to it needing to query metadata for each file, and caching didn't seem to help). However, I then tried out RioFS, which provided me with instant ...


0

An alternative would be to configure the webservers to transmit their access logs by means of syslog to your awstats server. That would mean zero configuration change to the awstats configuration when you bring webservers up and down. On an Apache webserver you might now have something like: ErrorLog "/var/log/www/error.log" CustomLog ...


1

When I do similar things in AWS or similar clouds, I use autofs to mount the log directories. That means that the log directories are mounted on a demand-driven basis, ie not until needed and unmounted shortly thereafter. It also means that access to a directory that maps to a down server involves a short wait before an empty directory is returned, instead ...


2

Take a look at nfstrace https://github.com/epam/nfstrace. It captures network traffic and perform deep packet inspection. It finds NFSv3/v4 procedures (READ, WRITE etc.) and tracks its statistics. Also, you can develop new pluggable modules for nfstrace.


1

In general, if you experience problems with idmapping you need: check that client can talk to LDAP/NIS server clean idmap cache by running 'nfsidmap -c' start rpc.idmapd in in a debug mode (rpc.idmapd -f -v) and watch the requests.


3

It is possible, though not trivial, to setup NFS Clusters in Amazon EC2 using DRBD for synchronous replication and Pacemaker + Corosync for automating failover of the NFS service and exports between nodes (without interrupting client access). If you're planning on replicating synchronously ("real-time"), you'll need both your EC2 instances to be in the ...


0

NOTE: 192.168.1.9 is client and 14 is server and is work fine in my two server .... Client pf.conf : block all pass in on em0 inet proto { tcp udp } from 192.168.1.14 to em0 port { sunrpc nfsd-status nfsd-keepalive nfsd lockd } pass out on em0 inet proto { tcp udp } from em0 to 192.168.1.14 port { sunrpc nfsd-status nfsd-keepalive nfsd lockd } Server ...


1

Decide what you want to do: application-independent and more "general" network security done by iptables on the NFS server application-specific security done by nfsd on the NFS server or both. In NFS, you share a directory read-only to a specific client as follows: echo "/mynfs clt.xmpl.com(sync)" >> /etc/exports.d/my-ro.exports exportfs -r


0

Since this question hasn't been marked closed or answered, I'd like to offer some input. Services for Unix (SFU) Applications is available from Microsoft, and appears to support all versions of MS Windows - however the installer does warn that there are compatibility issues with Vista. My efforts to use SFU (explicitly for NFS services) have repeatedly ...


2

You cannot specify the private IP in an Auto Scaling Group, but there is a way to accomplish the same goal. Use a userdata script to attach the same EIP to the instance upon launch. You can then use the public DNS of the EIP as the remote host for your NFS clients. When the NFS clients use DNS to perform a lookup on the public DNS of the NFS server, they ...



Top 50 recent answers are included