Tag Info

Hot answers tagged

4

DNS has proper built in replication, so why bother with something like that? You're introducing an unnecessary a single point of failure. Especially when you'll still need to issue a reload of any updated zone files on both DNS servers, as simply updating the zone file does not make it active on either of the servers.


4

ZFS with lz4 compression is pretty quick, depending on the compressibility of the data. But the rest of the items you're comparing aren't actually comparable. You've listed cluster filesystems, network filesystems, caching layers, and traditional filesystems.


3

No contest, tmpfs ! tmpfs is designed as a performance enhancement ...


2

If you want to limit the amount of disk space each user (or group) is using, then you can set up disk quotas on the NFS server (as mentioned in another answer already). You can easily find tutorials about how to do that. Once disk quotas are enabled, you can set the maximum disk space each user is able to use (the default is not to have any limit, so you ...


2

If you set up Quotas in the shared file system (on the server side), the clients can't use more space than the quota set.


2

In that case first match wins since the two expressions are of the same type. (A single host expression would win even if it came after the IP subnet expression because it has precedence.) See exports(5) for details.


2

Couple of tools for you: lsof shows you the open file handles iotop shows the process-wise I/O statistics in the top manner nethogs shows you the per-process network traffic strace allows you to see what a process is doing


2

Yes, that is possible and safe. Yes, that should be possible. But I recommend you to enable Amazon Load Balancer Sticky Session feature Yes, it should be possible with some caveats: you should be careful about network protocol used for accessing shared data. Subversion relies on file locking and atomic file move operations. SMB and SMB2 are known to be ...


1

You should be able to mount an NFS volume on the CoreOS host. If there is no userland tools to export a filesystem from CoreOS, you can use Fedora toolbox container provided by the toolbox command.


1

In this kind of situation I often found very useful to capture the NFS traffic (e.g., with tcpdump or Wireshark) and have a look at it to see if there is a specific reason for the high load. For example, you can use something like: tcpdump -w filename.cap "port 2049" to save only NFS traffic (being on port 2049) to a capture file, then you can open that ...


1

In general NFS servers don't give root on the client root access. You need to explicitly share the filesytem to the client with root= You don't say which version of Solaris you're using or file system type so can't give detailed instructions. Look at the documentation for share (or just google for "solaris nfs share root")


1

This is caused by directories not being automounted when the container is run. I had thought that /usr/groups/thing was the automount point, but evidently the sub-directories are auto-mounted individually. The solution is to make sure each one is mounted before entering the container: $ (cd /usr/groups/thing/foo; cd /usr/groups/thing/bar) $ docker run -v ...


1

Exists different options to cache file attribute on a NFS mount (client side): NOAC Use the noac mount option to achieve attribute cache coherence among multiple clients. Almost every file system operation checks file attribute information. The client keeps this information cached for a period of time to reduce network and server load. When noac ...



Only top voted, non community-wiki answers of a minimum length are eligible