[Date Prev][Date Next]   [Thread Prev][Thread Next]   [Thread Index] [Date Index] [Author Index]

Re: htree stabilitity and performance issues



On Fri, 2003-12-19 at 01:04, Theodore Ts'o wrote:
> On Thu, Dec 18, 2003 at 01:36:25PM +1100, Adam Cassar wrote:
> > What's your take on the nfs client load issues? It does run for 4-5
> > hours albeit at higher load (how explained by your post) however it does
> > eventually die with the load going stupid (180 odd). It seems that the
> > patch still has some nfs interoperability problems.
> 
> Was this on the nfs *client* or the nfs *server*?  

This was on the client which is a MP box. The server was still
responsive (with a higher load however) and I could still cat files on
the filesystem.

There is also the separate issue of the filesystem corruption - and
e2fsck reporting everything was aok (e2fsprogs v1.34).

> I'd really, really like to see a ps listing on the machine involved;
> the output of "ps alxww" and "ps auxww" would be useful.  The question
> is what processes are hung in wait, and what they're waiting on....

It's going to be a little difficult for me to do that as I am going away
on vacation for a month starting from today. I should have really tried
to gather more state information before posting - sorry. When I get back
I can play with it, it should be too difficult to duplicate in a test
set up.

> It would also be interesting to see if the LD_PRELOAD hack which I
> sent you helped alleviate the load on the server?  With the LD_PRELOAD
> hack, the access pattern on stat's and open's should be restored to
> the original workload, so if that makes the problem go away, then
> the problem was merely that NFS doesn't degrade gracefully under load.

I am happy to give that a go as well but it is going to have to wait
unfortunately.

> I believe, although I am not sure, that there are some NFS
> improvements that went into 2.6 that didn't get back-ported to 2.4.
> So it might be that running 2.6.0 on the clients and/or servers might
> actually help.  That would be a pretty daring move, though....

So was trying the htree patch :)

> Finally, can you give me a little bit more detail of exactly what is
> running on the clients and server, and the rationale of why you are
> trying to apparently run incoming mail processes over NFS?  (Is that
> what you're doing?  If so, it sounds rather scary...)

The clients generally run exim and courier-imap/courier-pop. Mail is
delivered and read over NFS via the maildir format which is NFS safe. As
maildir stores each email in a single file I imagined htree would
improve performance on the servers. 

What appears to be the issue is that the performance just degrades over
time to an unusable state. I only really noticed the issue on the pop
server as the load on the machine was ridiculous and mail was not
retrievable. 




[Date Prev][Date Next]   [Thread Prev][Thread Next]   [Thread Index] [Date Index] [Author Index]