Pawel Jakub Dawidek wrote:
> On Wed, Sep 09, 2009 at 10:07:08AM -0700, Matthew Ahrens wrote:
>> Unfortunately, I don't think that the send stream would help us diagnose it
>> further. What we really need is a way to reproduce how to get into this
>> situation in the fi
Unfortunately, I don't think that the send stream would help us diagnose it
further. What we really need is a way to reproduce how to get into this
situation in the first place.
--matt
Pawel Jakub Dawidek wrote:
> Hi.
>
> There is a bug (6709336), which was closed with 'Not Reproducible' reas
Ty Newton wrote:
> Hi,
> I've read a few articles about the lack of 'simple' raidz pool expansion
> capability in ZFS. I am interested in having a go at developing this
> functionality. Is anyone working on this at the moment?
FYI, Adam Leventhal and I worked out how to do this a while back.
gary mazz wrote:
> Hi,
>
> I'd like to use ZFS in a highly scalable data application.
>
> My requirements are very loose at this point. but I know I need object
> locking and I'd like to use the DMU interface. I would like the DMU
> transactions to be distributed across machines ? The preferr
Marc Moreau wrote:
> * This comment is noted from reading
> *
> http://src.opensolaris.org/source/xref/onnv/onnv-gate/usr/src/uts/common/fs/zfs/arc.c
> * which may or may not be the current version.
>
> I was reading through the code and I noticed that arc_no_grow and arc_dead
> are declared as
Jeremy Archer wrote:
> Also, someone may think that those init's are necessary, and chase a
> non-existent bug when one of them is missing. :-)
We try to keep all the mutex_init() & friends there for the ZFS ports to
other operating systems. Some of them need it.
--matt
Mark Maybee wrote:
> Pawel Jakub Dawidek wrote:
>> On Sat, Apr 18, 2009 at 06:05:56PM -0700, Matthew.Ahrens at sun.com wrote:
>>> Author: Matthew Ahrens
>>> Repository: /hg/onnv/onnv-gate
>>> Latest revision: f41cf682d0d3e3cf5c4ec17669b903ae621ef882
>
Eugene Turkulevich wrote:
> Hi, I have an idea about improvements of ZFS from the one side and
> improvements for of the GUI interface from the other side. Let me describe my
> idea.
>
>>From the one side we have an ZFS on-th-fly snapshots, and starting from
>>OpenSolaris 2008.11 we have TimeSl
Eddie Edwards wrote:
> In principle, I think you're right, in that an objset could hold both
> versions of a file. It would contain physical pointers to each copy.
> They would start out the same and diverge when one or other copy was
> modified.
>
> For instance, zfs could support a fast "cp" wh
Chen Zheng wrote:
> Hi matt,
>
> I have some problems about understanding zfs COW implemention. Suppose
> b and c are both children dir of a, if c changes, there will be new
> versions of both a and c, namely c' and a'.
>
> a a'
> b c c'
>
Actually, there will not be a new version of A unl
Cyril ROUHASSIA wrote:
>
> Dear all,
> I have some problems with zfs implementation issues:
>
> * I do not really understand what is a DSL directory (I though is
> was kind of structure listing all datasets + properties..) and
> what is the unique "Active Dataset" within a DSL
yes, please do, and include a crash dump.
--matt
Nils Goroll wrote:
> Has happened again after I replaced the faulty cable, seems to be
> reproducible
> with my zfs send/receive script.
>
> Should I open a bug?
>
> panic[cpu1]/thread=ff0167f4b200:
> dirtying dbuf obj=3d199 lvl=1 blkid=1 b
How do we get 4% faster when running on the same endianness as the fs was
created on? (created on x86, run on amd64) Where are the bulk of the calls
to BSWAP_* happening in that case?
--matt
Dan Anderson wrote:
> Here's some performance results running " find . -exec ls -l" on separate ZFS
>
Andreas Dilger wrote:
> On Feb 07, 2008 17:14 +, Ricardo Correia wrote:
>> On Qui, 2008-02-07 at 09:44 -0700, Mark Shellenbaum wrote:
>>> Section 2.2.4
>>>
>>> Not sure what you mean by storing all xattrs in a single object? Are you
>>> saying all of the Lustre attributes would be stored in a
Ricardo M. Correia wrote:
> On Qui, 2008-02-07 at 22:51 -0700, Neil Perrin wrote:
>> I believe when a prototype using 1K dnodes was tested it showed an
>> unacceptable (30%?) hit on some benchmarks. So if can possibly
>> avoid increasing the dnode size (by default) then we should do so.
>
> Hmm, i
Ricardo M. Correia wrote:
>
> On Qui, 2008-02-07 at 12:48 -0800, Matthew Ahrens wrote:
>> First, note that the problem is quite constrained:
>>
>> There are a fixed number of attribute names (for any given software version)
>> since these are interpreted sy
Mark Shellenbaum wrote:
> Girish Shilamkar wrote:
>> Hi,
>> With reference to my previous mail ([zfs-code] Nvpair for storing EAs
>> in dnode) which had expressed our intent to use libnvpair for EAs in
>> dnode. I am now sending a high level design document for the same.
>>
>> Any suggestions/
jay xiong wrote:
> I have two questions for zfs code:
>
> In dbuf_hold_imple():
> In this piece of code, it will copy the data to the extra buffer if the
> current one is syncing into disks. My question is that why not deferring this
> data copy to dbuf_dirty, where we check the db_data_pending
Girish Shilamkar wrote:
> Hi,
> While using DMU for lustre we need to store some extra information in
> directory entries which is required for lustre operations. The size of
> typical entries to be stored in directory entries are 16 bytes, hence we
> can't use micro zap. And we would like to
Neil Perrin wrote:
> Matthew Ahrens wrote:
>> Neil Perrin wrote:
>>>
>>> Ricardo M. Correia wrote:
>>>> Ricardo M. Correia wrote:
>>>>> boolean_t need_lock = !RW_LOCK_HELD(&dp->dp_config_rwlock);
>>>>>
>>>>>
Neil Perrin wrote:
>
> Ricardo M. Correia wrote:
>> Ricardo M. Correia wrote:
>>> boolean_t need_lock = !RW_LOCK_HELD(&dp->dp_config_rwlock);
>>>
>>> if (need_lock)
>>> rw_enter(&dp->dp_config_rwlock, RW_READER);
>>>
>> Maybe I posted to soon.
>> Am I right that this specific code (in dsl_
Mark Furner wrote:
> Hi
>
> Am not sure who to address this query to, there is no list for ZFS forensics
> (yet). I'm looking at a DD dump of a VTOC partition containing a ZFS test
> file system, follow the Uberblock DVAs to 3 separate locations (ditto blocks,
> as expected) but don't find an o
Pawel Jakub Dawidek wrote:
> On Fri, Oct 12, 2007 at 11:10:31AM +0100, Darren J Moffat wrote:
>> Pawel Jakub Dawidek wrote:
>>> I understand it's not trivial, but beeing able to reference the same
>>> block from different datasets would be a really nice feature to have.
>>> The functionality discus
Pawel Jakub Dawidek wrote:
> On Thu, Oct 11, 2007 at 10:47:44AM +0100, Robert Milkowski wrote:
>> Hello Matthew,
>>
>> Thursday, October 11, 2007, 9:10:13 AM, you wrote:
>>
>> MA> Robert Milkowski wrote:
I haven't looked into details but in theory one should be
able to copy/move a file wi
Robert Milkowski wrote:
> I haven't looked into details but in theory one should be
> able to copy/move a file within the same datapool between datasets
> without having to actually copy data blocks... or maybe there's some
> detail which actually makes it hard to implement...
Once a block is refe
Darren J Moffat wrote:
> Andreas Dilger wrote:
>
>> I agree, but I suspect large dnodes could also be of use to ZFS at
>> some point, either for fast EAs and/or small files, so we wanted to
>> get some buy-in from the ZFS developers on an approach that would
>> be suitable for ZFS also. In partic
eric kustarz wrote:
> On Sep 12, 2007, at 4:31 AM, Darren J Moffat wrote:
>
>> The DMU layer will determine which objects in be encrypted when being
>> placed in a dataset that has the encryption property turned on.
>>
>> http://opensolaris.org/os/project/zfs-crypto/phase1/dmu_ot/
>>
>> Which of t
Pawel Jakub Dawidek wrote:
> On Tue, Aug 07, 2007 at 11:28:31PM +0100, James Blackburn wrote:
>> Well I read this email having just written a mammoth one in the other
>> thread, my thoughts:
>>
>> The main difficulty in this, as far as I see it, is you're
>> intentionally moving data on a checksumm
Duff,
You should resync with Lin Ling's putback for: PSARC 2007/342 "Enhanced ZFS
Pool Properties" from this morning. It substantially changes the way the
pool properties are implemented, and should make implementing a new pool
property much easier. (Also note, I fixed particular nastyness th
Darren J Moffat wrote:
> The DMU layer will determine which objects in be encrypted when being
> placed in a dataset that has the encryption property turned on.
>
> http://opensolaris.org/os/project/zfs-crypto/phase1/dmu_ot/
>
> Which of these marked as FALSE can be and should be *TRUE* ?
>
Ac
Atul Vidwansa wrote:
> ZFS Experts,
>
> Is it possible to use DMU as general purpose transaction engine? More
> specifically, in following order:
>
> 1. Create transaction:
> tx = dmu_tx_create(os);
> error = dmu_tx_assign(tx, TXG_WAIT)
>
> 2. Decide what to modify(say create new object):
> dmu_
Don Cragun wrote:
> Chris,
> If ZFS is ever going to pass the test suites so we can claim
> that ZFS meets POSIX and UNIX filesystem requirements, you'll find that
> the conformance test suites will be trying things similar to what I've
> been describing. The test suites certainly aren't per
Darren J Moffat wrote:
> George Wilson wrote:
>> New code should either update the SPA_VERSION whenever you change the
>> on-disk format and/or the new ZPL_VERSION whenever you change the Posix
>> layer.
>
> That is what I thought but as you say it is currently a little confusing.
>
> Given tha
Mark Furner wrote:
> Thanks for the clarification, Darren, and sorry for cross-posting.
>
> OK, physical device -> pool -> file-system(s)
>
> Some questions:
>
> 1) zpool import allows options similar to zfs. Can I set the same or similar
> read-only (RO) options for the whole pool
> (noexec,n
Darren J Moffat wrote:
> While doing a zio we have the information about what pool and dataset we
> are operating on encoded in a zbookmark_t.
>
> I found I can go from a object to a dataset name using
> dsl_dsobj_to_dsname() but that needs a pool name.
>
> The reason I need the dataset name is
George Wilson wrote:
> Darren J Moffat wrote:
>> Is it possible to have dataset properties that are managed using the
>> dsl_prop_set() / dsl_prop_get() interfaces that aren't made available
>> via zfs(1), in fact I probably don't want them in userland at all.
>>
> You can set the pd_visible f
Manoj Joseph wrote:
> Hi,
>
> Replying to myself again. :)
>
> I see this problem only if I attempt to use a zpool that already exists.
> If I create one (using files instead of devices, don't know if it
> matters) like ztest does, it works like a charm.
You should probably be posting on zfs-d
Darren J Moffat wrote:
> Matthew Ahrens wrote:
>> Darren J Moffat wrote:
>>> My initial thought was using *dsl_dataset_phys_t* however that
>>> doesn't have sufficient space so it may be better to store a
>>> "pointer" there to some external
Darren J Moffat wrote:
> My initial thought was using
> *dsl_dataset_phys_t* however that doesn't have sufficient space so it
> may be better to store a "pointer" there to some external object.
Yep, I would recommend that you store each key in an object in the MOS,
and keep that object's number
Pawel Jakub Dawidek wrote:
> On Tue, May 01, 2007 at 05:48:15PM -0700, Matthew Ahrens wrote:
>> Pawel Jakub Dawidek wrote:
>>> so I started to use it. The thing is that it broke some functionality.
>>> For example 'zpool status -v' doesn't show file n
Pawel Jakub Dawidek wrote:
> so I started to use it. The thing is that it broke some functionality.
> For example 'zpool status -v' doesn't show file names anymore.
>
> I tracked it down to the zap_value_search() function and changing:
>
> if (za->za_first_integer == value) {
> into:
>
Jeremy Teo wrote:
> Hello,
>
> I'm curious as to what the Dirty Time Log structures are for : namely
>
> 1) vdev_dtl_map
> 2) vdev_dtl_scrub
The DTL keeps track of times (txg ranges) when a device was offline.
For example, in a mirrored pair, one side of the mirror may go offline
for some time
Jason Banham wrote:
> Hello,
>
> I've been looking at the source code to the transaction engine / DMU
> and was wondering what the purpose of dn->dn_assigned_txg was
> for?
An object can be modified in many transactions (dmu_tx_t's)
simultaneously, but they must all be for the same transaction
No?l Dellofano wrote:
>> How to I tell the system not to mount certain zfs at boot?
> you can set the 'canmount' property to 'off' for whatever filesystem you
> do not want mounted:
Or if you don't want anything mounted, 'zfs set mountpoint=none
'. (the mountpoint property is inherited.)
--mat
Adhari C Mahendra wrote:
> Sorry being offline for moment.
>
>> --matt
>>
>> ps. do you have any snapshots? Please send the
>> output of 'du -hs
>> ', 'zfs list', 'zfs get -r all '.
> No I do not use any snapshot. Sorry, I have ripped out the zfs so I am unable
> to give you the output of those
Darren J Moffat wrote:
> Thanks again for trying. Now what do I do ;-(
The bug doesn't exist in nevada, right? So try removing your code until
you find what's broken.
--matt
Darren J Moffat wrote:
> eric kustarz wrote:
>> Darren J Moffat wrote:
>>> eric kustarz wrote:
>>>
Not sure if you've already posted this or not, but do you have a
webrev?
>>>
>>> Yes it was in the original message:
>>>
>>> http://cr.grommit.com/~darrenm/zfs-crypto/
>>>
>>
>> I didn't se
Adhari C Mahendra wrote:
> [Sol 10 6/6 x64]
>
> I am very newbie in ZFS.
>
> I have created 30GB storage pool and become a root_pool. If I run "du -hs"
> from root directory , it reports only used 5.4G. But when I run "df -h", it
> reports used 26G. Why it is happens? How to reclaim back to 5.4
Eric Schrock wrote:
> http://cr.grommit.com/~eschrock/zfs-prop/
Shouldn't dsl_prop_get* all take into account the inheritability of
properties? It looks like non-inheritable properties will be inherited
unless you are using dsl_prop_get_all(). It seems like the check for
inheritability needs
Pawel Jakub Dawidek wrote:
> Hi.
>
> I've almost all file system functions working.
>
> I started to run some heavy file system regression tests. They work. fsx
> wasn't able to break my port, but the test you can find here:
>
> http://people.freebsd.org/~kan/fsstress.tar.gz
>
> broke it.
50 matches
Mail list logo