[Top] [All Lists]

RE: fc10, some processes stuck in D state

To: <david@xxxxxxxxxxxxx>
Subject: RE: fc10, some processes stuck in D state
From: <yuji_touya@xxxxxxxxxxxxxxxxxxxx>
Date: Fri, 7 Jan 2011 20:00:38 +0900
Accept-language: ja-JP
Acceptlanguage: ja-JP
Cc: <xfs@xxxxxxxxxxx>
In-reply-to: <20110106050057.GF8322@dastard>
References: <8529A87D856C184491994079B5F87B68C1A8289FCC@xxxxxxxxxxxxxxxxxxxx> <20110106050057.GF8322@dastard>
Thread-index: AcutXr+au7BdH9lvSkG7QOSD1Vl6bAAFjkJQ
Thread-topic: fc10, some processes stuck in D state

Thank you for your replying.

> -----Original Message-----
> From: Dave Chinner [mailto:david@xxxxxxxxxxxxx] 
> Sent: Thursday, January 06, 2011 2:01 PM

> Everything is waiting for log space to be freed. Typically a sign
> that metadata has not been flushed or that IO completion has not occurred
> so the tail is not moving forward.

Nice to know it! It will help us.

> What did you change 3 months ago? Or did this always happen?

This always happens.

> So doing new read IOs starts stuff moving again? That sounds like an IO
> completion has not arrived from the lower layers until a new IO is
> issued and completes. Perhaps the hardware RAID is not issuing an
> interrupt when it should?

Yes, new read IOs seem to wake up them.
Are there any tools/ways to examine whether expected interrupt occurred or not?

> What type of RAID controller/storage hardware are you using? Is it
> all running the latest firmware, appropriate drivers, etc?

PCI Express adapter and an external RAID system, connected with fiber channel 
each other.
The BIOS, PCI Express adapter and RAID system's firmware are not up to date.
We will try to update these softwares and check if the same problem occur.
It would be nice to reproduce this problem as easy as posible.
If there is suitable application(benchmark or test program etc), please let me 


<Prev in Thread] Current Thread [Next in Thread>