Bug 80018: NFSv4 on RHEL 6.2 over six times slower than 5.7

[mark]

On 07/10/12 20:41, Corey Kovacs wrote:
> Well, looking at the Bugzilla, it looks like they are asking you to contact
> your support rep. If you are working in the government, there will be a TAM

I'm waiting for my manager to tell me how to file. I have no idea what a 
TAM is - we do all this ourselves.

> assigned to handle these problems. If you are a contractor working on
> systems to be delivered to the government, then you need to be a paying
> customer to expect support from Red Hat. Especially when your original
> problem manifested on CentOS. Granted, it's the same code base, but that
> doesn't make your problem on CentOS, a problem for Red Hat.

I'll say it one more time: we found the problem on CentOS. We went to 
our test RHEL system. Updated it. Exported a directory *from* the RHEL 
box to itself, to /mnt/foo, and ran the test, and got the same results.

In fact, I ran it twice today, updating the kernel in between, and with 
6.3, it's taking a consistent 7.5 min, instead of the 6.5 we were 
getting with 6.2
<snip>
> Now, all that said and done, here are some questions for you which might
> help us figure what would help.
>
> 1. What options are present on the mount? (cat /proc/mounts, thinks like
> sync can be a problem)

I"m not at work. I'll have to answer that in the morning. I will tell 
you that when we were first trying to figure it out, two months ago, I 
did try no sync.

> 2. What does your /etc/exports config look like on your server node (cat
> /etc/exports)

/scratch/foo <servername>: options

> 3. You are using NFSv4, are you using Kerberos with it?

I don't believe we have kerborous set with NFS. We do use it for other 
things.

>      3.a. If so, what mode are you using for your gss/krb flag? (krb5,
> krb5i, krb5p)
> 4. What's your network speed? Are you sure? (ethtool ethX to make sure)

Gigabit.

> 5. Selinux?
Permissive.

> 6. Auditing?

Do you mean selinux auditing? As I said, doing it on the local drive 
takes seconds. Doing it from a 5.x NFS server takes about 1.5 min. 
Therefore, there's nothing that could affect it on the one server.

> 7. How many clients are hitting your server and how many nfsd threads are
> you running on it?

No other clients. This is a test system.
>
> This is by no means an exhaustive list of things to look at.
>
> Anyway, in order to get any real help, you cannot just shout out, "My stuff
> is broke, it's Red Hat's fault, no one will listen to me!"
>
> Give us something to work with.

Try reading the damn bug.

>
> By the way, in looking at the responses to your bugzilla, it doesn't look
> to me like they were shamed into responding. It looks like are telling you
> to go through proper channels if they exist. full stop.

No, they gave *ZERO* responses until today. The one and only response 
before today was, "oh, we're up to 6.3, we'll not even look at it".

And my manager, who's a fed, and I, a contractor, along with the other 
admin under him, who is also a contractor, handle the licenses, etc, so 
there's no one else to wait for.

	mark