[Top] [All Lists]

Re: [Ksummit-2005-discuss] Summary of 2005 Kernel Summit Proposed Topics

To: open-iscsi@xxxxxxxxxxxxxxxx
Subject: Re: [Ksummit-2005-discuss] Summary of 2005 Kernel Summit Proposed Topics
From: Dmitry Yusupov <dmitry_yus@xxxxxxxxx>
Date: Wed, 30 Mar 2005 09:39:43 -0800
Cc: Andi Kleen <ak@xxxxxx>, jamal <hadi@xxxxxxxxxx>, James Bottomley <James.Bottomley@xxxxxxxxxxxxxxxxxxxxx>, Rik van Riel <riel@xxxxxxxxxx>, andrea@xxxxxxx, michaelc@xxxxxxxxxxx, ksummit-2005-discuss@xxxxxxxxx, netdev <netdev@xxxxxxxxxxx>
In-reply-to: <>
References: <> <1111905181.4753.15.camel@mylaptop> <> <> <> <1112027284.5531.27.camel@mulgrave> <> <1112116762.5088.65.camel@beastie> <1112130512.1077.107.camel@jzny.localdomain> <> <>
Sender: netdev-bounce@xxxxxxxxxxx
On Wed, 2005-03-30 at 09:24 -0800, Matt Mackall wrote: 
> I seem to recall this being fairly easy to trigger by simply pulling
> the network cable while there's heavy mmap + write load. The system
> will quickly spiral down into OOM and will remain wedged when you plug
> the network back in. With iSCSI, after some extended period all the
> I/Os will have SCSI timeouts and lose everything.

We've discussed that already. SCSI timeout logic just doesn't fit. (see
rfc3720). For iSCSI, SCSI timeout logic *must* be disabled until iSCSI
recovery is complete. host block/unblock logic in recent iSCSI transport
patch will help to implement that.

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