[SOLVED] TERRIBLE NFS lags with RHEL 7 client

Vanyel · 08-22-2017, 01:44 PM

I have a bunch of RHEL6 & 5 machines auto mounting shares from a couple of CentOS 6.9 servers. Everything works fine with them.

Upgraded my first client to RHEL 7.4 and there are terrible lags with the automounts. Whenever I first cd into an automount, there's a lag of about 30 - 40 seconds before it finally mounts and the shell comes back. Thereafter, moving around within the mount is fine.

With these mounts active, 'df -h' or 'mount' commands lag about 10 seconds the first time they are issued too, then respond instantly thereafter.

It's like RHEL 7 has some new kind of negotiation going on that I can't pin down.

Anyone have any ideas? in case it helps, here's the output of the mount command when the automounts are active:

Code:

$ mount
<snip>
server1:/path_to/folder on /path_to/local_mountpoint type nfs4 (rw,relatime,vers=4.0,rsize=524288,wsize=524288,namlen=255,hard,proto=tcp,port=0,timeo=600,retrans=2,sec=sys,clientaddr=192.168.1.232,local_lock=none,addr=192.168.1.205)

server2:/path_to/folder on /path_to/outher_mountpoint  type nfs4 (rw,relatime,vers=4.0,rsize=1048576,wsize=1048576,namlen=255,hard,proto=tcp,port=0,timeo=600,retrans=2,sec=sys,clientaddr=192.168.1.232,local_lock=none,addr=192.168.1.187)

smallpond · 08-23-2017, 06:11 AM

Check DNS. It could be due to A or AAAA lookups.

Vanyel · 08-23-2017, 11:26 AM

Sorry I didn't mention it, but this is on an internal network and the machines are defined in /etc/hosts .

TB0ne · 08-23-2017, 12:04 PM

Quote:

Originally Posted by Vanyel

Sorry I didn't mention it, but this is on an internal network and the machines are defined in /etc/hosts .

From the RHEL 7 Documentation:

Quote:

Originally Posted by RHEL Docs

The NFS protocol version used in Red Hat Enterprise Linux 7 is identified by the mount options nfsvers or vers. By default, mount will use NFSv4 with mount -t nfs. If the server does not support NFSv4, the client will automatically step down to a version supported by the server. If the nfsvers/vers option is used to pass a particular version not supported by the server, the mount will fail. The file system type nfs4 is also available for legacy reasons; this is equivalent to running mount -t nfs -o nfsvers=4 host:/remote/export /local/directory.

(bold for emphasis added). If you're upgrading, this 'stepping-down' may be the cause. Since RHEl7 uses NFS4 by default, and your other systems don't/aren't...could be it. Have you opened a support case with Red Hat yet??

https://access.redhat.com/documentat...entconfig.html

John VV · 08-23-2017, 01:43 PM

first HOW did you "upgrade" rhel6 to 7 ?

the only real way - a NEW install ?

ALSO BE AWARE!!!! rhel 7 uses "systemd "!!! rhel6 used systemV

Quote:

t's like RHEL 7 has some new kind of negotiation going on that I can't pin down.

it DOSE systemD

this WILL change things in the auto connect

contact redhat and or read the admin guide for systemd on the redhat site

Vanyel · 08-23-2017, 04:27 PM

My "upgrade" was a complete wipe and re-install and re-copying user data, re-entering settings where needed.

Vanyel · 08-23-2017, 04:54 PM

YES! TB0ne was on the right track. After a little googling to figure out just the right syntax, in my /etc/auto.direct file I added the boldfaced

Code:

-fstype=nfs,rw,nfsvers=3

to the auto mount def lines and now they use NFS3 and they're instantaneous again!