- Original Message -
From: "Matthew Ahrens via illumos-zfs"
With regards to buffers being allocated by arc_get_data_buf() I can't see
a path by which ARC will prevent a new buffer being allocated even when
arc_evict_needed().
It won't, but it will evict an existing buffer, thus free
On Wed, Sep 17, 2014 at 1:08 PM, Andriy Gapon wrote:
>
> It seems that dead code removing is a hot topic :-)
> I think that the following commit has introduced a little bit of such code:
> commit 5d7b4d438c4a51eccc95e77a83a437b4d48380eb
> Author: Matthew Ahrens
> Date: Thu Jun 5 13:19:08 2014
It seems that dead code removing is a hot topic :-)
I think that the following commit has introduced a little bit of such code:
commit 5d7b4d438c4a51eccc95e77a83a437b4d48380eb
Author: Matthew Ahrens
Date: Thu Jun 5 13:19:08 2014 -0800
4757 ZFS embedded-data block pointers ("zero block comp
- Original Message -
From: "Matthew Ahrens"
To: "illumos-zfs" ; "Steven Hartland"
Cc: "developer"
Sent: Sunday, September 14, 2014 6:31 PM
Subject: Re: [zfs] ZFS Write Throttle Dirty Data Limit Interation with Free
Memory
On Sun, Sep 14, 2014 at 10:00 AM, Steven Hartland via ill
---
This is an automatically generated e-mail. To reply, visit:
https://reviews.csiden.org/r/104/#review257
---
Ship it!
Ship It!
- Richard Elling
On Sept. 17, 2014, 3:11 a.m., Mat
---
This is an automatically generated e-mail. To reply, visit:
https://reviews.csiden.org/r/101/#review255
---
Ship it!
Ship It!
- Richard Elling
On Sept. 17, 2014, 2:45 a.m., Mat
Thanks for debugging this, Steven, and for the workaround!
--matt
On Wed, Sep 17, 2014 at 4:47 AM, Steven Hartland
wrote:
> After much debugging I can finally send mails to the list :)
>
> The problem is that the open-zfs.org mail server
> (AwesomeProxy?) has a broken implementation of ESMTP's
We've been investigating a problem with stalls on FreeBSD when
using ZFS and one of the current theories which is producing some
promising results is within the new IO scheduler, specifically
around how the dirty data limit being static limit.
The stalls occur when memory is close to the low wate
> Was it not just me then Jorgen?
Nope, I had the very same problem, appreciate you even included the fix!
Cheers,
Lund
___
developer mailing list
developer@open-zfs.org
http://lists.open-zfs.org/mailman/listinfo/developer
Was it not just me then Jorgen?
- Original Message -
From: "Jorgen Lundman"
To:
Sent: Wednesday, September 17, 2014 1:03 PM
Subject: [OpenZFS Developer] Testing Hartland's fix
testing test.
___
developer mailing list
developer@open-zfs.
LGTM
- Original Message -
From: "Matthew Ahrens"
To: "Christopher Siden"
Cc: "OpenZFS Developer Mailing List"
Sent: Wednesday, September 17, 2014 3:45 AM
Subject: [OpenZFS Developer] Review Request 101: 5174 add sdt probe forblocked
read in dbuf_read()
--
testing test.
___
developer mailing list
developer@open-zfs.org
http://lists.open-zfs.org/mailman/listinfo/developer
After much debugging I can finally send mails to the list :)
The problem is that the open-zfs.org mail server
(AwesomeProxy?) has a broken implementation of ESMTP's
pipelining, specifically if the end of DATA "." is
sent together with the "QUIT" (valid for pipelining)
the send will timeout and ne
Seems like open-zfs.org mail server has broken pipelining
support, test to see if local postfix config option disables
this and fixes the problem...
___
developer mailing list
developer@open-zfs.org
http://lists.open-zfs.org/mailman/listinfo/developer
This is a test to debug mail failures to developer@open-zfs.org
Regards
Steve
___
developer mailing list
developer@open-zfs.org
http://lists.open-zfs.org/mailman/listinfo/developer
15 matches
Mail list logo