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

Re: Oops, incomplete patch. (fwd)

>To: Russ Rew <address@hidden>
>From: "Stonie R. Cooper" <address@hidden>
>Subject: Re: 20020730: The significance of pbuf_flush messages.
>Organization: Planetary Data
>Keywords: pbuf_flush-problem, Linux file systems, ext2, ext3, reiserfs, fsck

------- Forwarded Message

Date:    Fri, 06 Sep 2002 01:23:25 -0000
From:    "Stonie R. Cooper" <address@hidden>
To:      Russ Rew <address@hidden>
cc:      address@hidden, address@hidden
Subject: Re: Oops, incomplete patch.

Russ,

On Thursday 05 September 2002 15:42, Russ Rew wrote:
> Thanks for the information on this problem.  Here's my summary of what
> you learned, let me know if I got it wrong:
>
>   - On a Linux platfrom, use a simple ext2 file system partition for
>     the LDM product queue, since reiserfs or ext3 will cause
>     performance problems, with symptom lots of pbuf_flush messages.

That is correct - and it as the level of the kernel, not in user space (i.e. 
not readily fixable); I haven't dug, and probably won't, but would guess it's 
tied to the way the queue is maintained versus the journaling logging that 
takes place with ext3 and reiserfs.

>
>   - For even better performance, turn off the periodic file system
>     checking with "tunefs -c -1".

This is only necessary if the "regularly" scheduled fsck's that occur with 
ext2 annoy you.  I prefer to only fsck when needed - and given file and 
purpose of this particular filesystem, it seems silly to waste the time.

>
> I'm still not sure if the patch we provided to LDM 5.2 is also needed,
> but we're including it in 5.2.1 anyway ...

Any speedup without loss of stability is good - I don't think you wasted your 
time.

>
> --Russ

-- 
Stonie R. Cooper
Planetary Data, Incorporated
ph. (402) 782-6611

------- End of Forwarded Message