Re: [Nfs-ganesha-devel] [NFS-Ganesha-Support] Nfs-ganesha ntirpc crash

This list has been deprecated. Please subscribe to the new devel list at lists.nfs-ganesha.org. I haven't seen anything like this before. The problem is that the resarray in the compound is corrupted somehow (so it's actually a Ganesha bug). Is this reproducible? Do you have logs? Do you kn

Re: [Nfs-ganesha-devel] DSESS9002 and DSESS9003 test cases result in NFS4ERR_GRACE

This list has been deprecated. Please subscribe to the new devel list at lists.nfs-ganesha.org. Jeff revamped Grace about a year ago as part of his work on clustered FSAL_CEPH. This may have fixed the issue. Looking through, the only code that doesn't directly check the grace state is the NF

Re: [Nfs-ganesha-devel] [NFS-Ganesha-Support] Error messages

This list has been deprecated. Please subscribe to the new devel list at lists.nfs-ganesha.org. Hi, David. Answers inline. On Fri, May 17, 2019 at 11:42 AM David C wrote: > > Hi All > > I recently put an nfs-ganesha CEPH_FSAL deployment into production, so far so > good but I'm seeing some err

Re: [Nfs-ganesha-devel] libntirpc not available anymore?

This list has been deprecated. Please subscribe to the new devel list at lists.nfs-ganesha.org. Over the weekend, the ntirpc repo was deleted. I've restored it, and adjusted permissions so it shouldn't happen again. Daniel On 3/4/19 4:41 AM, Malahal Naineni wrote: This list has been depreca

Re: [Nfs-ganesha-devel] crash in opr_rbtree_insert with nfs-ganesha 2.6.3

This list has been deprecated. Please subscribe to the new devel list at lists.nfs-ganesha.org. This is not a known case (and, in fact, none of that code has changed in about a year). The issue is that nsm_clnt has been destroyed, but is being used. I don't see how this can happen, since all

Re: [Nfs-ganesha-devel] Ganesha 2.6.3 Segfault

This list has been deprecated. Please subscribe to the new devel list at lists.nfs-ganesha.org. I'm not seeing any easy way that cmpf could be corrupted. The structure before it is fairly complex, with it's last element being an integer, so it's unlikely that something wrote off the end of th

Re: [Nfs-ganesha-devel] 2.6.3 Health status is unhealthy

I need to do to clear the unhealthy status? Would you expect there to be a message to say the server has returned to a healthy state? Thanks David On Mon, 24 Sep 2018, 18:09 Daniel Gryniewicz, <mailto:d...@redhat.com>> wrote: This list has been deprecated. Please subscribe to t

Re: [Nfs-ganesha-devel] 2.6.3 Health status is unhealthy

This list has been deprecated. Please subscribe to the new devel list at lists.nfs-ganesha.org. I think this is due to the low traffic. What that check says is that we got new ops enqueued (1, in this case) but no ops dequeued. However, since there was only 1 op enqueued, I suspect that the i

Re: [Nfs-ganesha-devel] Issue with file locks after upgrade from 2.5.4 to 2.6.2

statd. Daniel On 09/11/2018 11:31 PM, Naresh Babu wrote: Thanks for the response, Daniel. Ping to ":::10.0.0.7" works fine. Do you suspect anything else? Thanks, Naresh On Tue, Sep 11, 2018 at 5:48 AM Daniel Gryniewicz <mailto:d...@redhat.com>> wrote: My guess is

Re: [Nfs-ganesha-devel] Issue with file locks after upgrade from 2.5.4 to 2.6.2

This list has been deprecated. Please subscribe to the new devel list at lists.nfs-ganesha.org. My guess is that this is related to IPv6. IPv6 support in 2.5 was spotty, but that's been fixed since. It's clearly using the 4-in-6 address (pretty common on a v6 enabled machine), and I don't bel

Re: [Nfs-ganesha-devel] 2.5 Ganesha with extended API crash

This list has been deprecated. Please subscribe to the new devel list at lists.nfs-ganesha.org. Looking at the code, I believe it isn't possible for fh_desc->addr to be null. Here's why: It comes from mdcache_locate_host(), which gets it from mdc_lookup(). This get's it from mdache_copy_fh()

Re: [Nfs-ganesha-devel] Connection to V4 port gets closed soon after it is established.

This list has been deprecated. Please subscribe to the new devel list at lists.nfs-ganesha.org. So, the issue appears to be that the xprt is created, but no packets are ready to be be read on it, so the end of the epoll loop cleans up unused xprts. Since the new xprt has not been used, it's la

Re: [Nfs-ganesha-devel] Umask Syntax

This list has been deprecated. Please subscribe to the new devel list at lists.nfs-ganesha.org. It looks like umask is just an integer. As such, it should accept octal, hex, or decimal input. It looks like there is residual code indicating that this once had special parsing code, but none of

Re: [Nfs-ganesha-devel] File create failing, after mounting from Mac OS

This list has been deprecated. Please subscribe to the new devel list at lists.nfs-ganesha.org. It's trying to set the FATTR4_ARCHIVE attribute, which is a deprecated attribute that Ganesha doesn't support. Since Ganesha doesn't advertise support for it, the client should not be setting it. (

Re: [Nfs-ganesha-devel] How to disable MDCACHE FSAL

This list has been deprecated. Please subscribe to the new devel list at lists.nfs-ganesha.org. Hi. The short answer is that you can't. MDCACHE provides the handle cache, which ensures that obj_handle pointers are stable across calls, and implements refcounting. Without it, Ganesha would cra

Re: [Nfs-ganesha-devel] cb_program or cb_callback_ident always the same

o find something to simulate the PID. Can you advice? Thank you On Fri, Jun 15, 2018 at 3:09 PM, Daniel Gryniewicz <mailto:d...@redhat.com>> wrote: I don't believe there's any necessity for a client to send different client_id's for different processes, as l

Re: [Nfs-ganesha-devel] cb_program or cb_callback_ident always the same

for 2 different processes. So if the client_id and the opaque always have the same value (for 2 different processes), how can we distinguish the client? I've tried with this field so_owner.so_nfs4_owner.so_clientid Thank you. Viet On Mon, Apr 30, 2018 at 2:38 PM, Daniel Gryniewicz &

Re: [Nfs-ganesha-devel] ACE permission check

This list has been deprecated. Please subscribe to the new devel list at lists.nfs-ganesha.org. If you take a look at code protected by ENABLE_VFS_DEBUG_ACL, you can see a test implementation of ACL enforcement I added to VFS a few years ago. It should be as complete as can be done using mode.

Re: [Nfs-ganesha-devel] ACE permission check

This list has been deprecated. Please subscribe to the new devel list at lists.nfs-ganesha.org. So, the access check is, of course, advisory to the client. It doesn't have to make one at all, but can just issue the rename, and expect it to succeed or fail based on permissions. I'm not sure wh

Re: [Nfs-ganesha-devel] Selinux denials with version 2.6.1-0.1.el7

This list has been deprecated. Please subscribe to the new devel list at lists.nfs-ganesha.org. Hi, Muminul Ganesha needed selinux policy added to allow it access to it's log and recovery directories. This was added in Fedora, but I don't know if it was added in Centos, or in what versions.

Re: [Nfs-ganesha-devel] cb_program or cb_callback_ident always the same

This list has been deprecated. Please subscribe to the new devel list at lists.nfs-ganesha.org. Hi. The client program ID in a lock owner is an opaque. That is, it's not defined in the spec, and the server can't use it for anything other than a byte string. The concatenation of the client-ID

Re: [Nfs-ganesha-devel] Is the mailing list dead? Or just quiet?

This list has been deprecated. Please subscribe to the new devel list at lists.nfs-ganesha.org. On 04/20/2018 09:40 AM, TomK wrote: On 4/20/2018 9:27 AM, Frank Filz wrote: This list has been deprecated. Please subscribe to the new devel list at lists.nfs-ganesha.org. This list has been depreca

[Nfs-ganesha-devel] Is the mailing list dead? Or just quiet?

This list has been deprecated. Please subscribe to the new devel list at lists.nfs-ganesha.org. All I've gotten this week has been gerrit notifications. Is everyone just quiet? Or did something happen to the list (or my subscription)? Daniel -

Re: [Nfs-ganesha-devel] Ganesh 2.3 : NFSv4 client gets error NFS4ERR_OLD_STATEID

So, NFS4ERR_OLD_STATEID can only happen in one circumstance: when the State presented by the client doesn't match the State that Ganesha expects. In this case, it's the sequence number that's off-by-one. This could be the result of a replay, but the code checks for this, and the owners must be

Re: [Nfs-ganesha-devel] mdcache growing beyond limits.

ill have an active entry. Thanks, Pradeep On 4/3/18, Daniel Gryniewicz mailto:d...@redhat.com>> wrote: > So, the way this is supposed to work is that getting a ref when the ref > is 1 is always an LRU_REQ_INITIAL ref, so that moves it to the MRU.  At >

Re: [Nfs-ganesha-devel] mdcache growing beyond limits.

So, the way this is supposed to work is that getting a ref when the ref is 1 is always an LRU_REQ_INITIAL ref, so that moves it to the MRU. At that point, further refs don't move it around in the queue, just increment the refcount. This should be the case, because mdcache_new_entry() and mdca

Re: [Nfs-ganesha-devel] NFSv4 recovery directory path

If you can handle PID and Cache being moved, you can override SYSSTATEDIR on the cmake line. Other than that, not currently. There's no good reason why it's not configurable, just that no one has needed it before. Daniel On 04/02/2018 05:33 AM, Sriram Patil wrote: Hi, Currently the NFSv4

Re: [Nfs-ganesha-devel] ganesha 2.5 - specifying network ID for allowing specific clients

So, the issue here appears to be that you're clients are showing up as IPv6 addresses (specifically as IPv4 compatibility addresses) and the CIDR network support in 2.5 doesn't work with IPv6 at all, so it's not matching. This is fixed in 2.6, by transitioning to using libcidr for client handl

Re: [Nfs-ganesha-devel] nss_getpwnam: name 't...@my.dom@localdomain' does not map into domain 'nix.my.dom'

On 03/23/2018 09:58 AM, William Allen Simpson wrote: On 3/23/18 7:59 AM, Daniel Gryniewicz wrote: Thanks, Tomk.  PR is here: https://review.gerrithub.io/404945 Actually, it seems fairly elegant. ntirpc and rdma also have the USE_ and _USE_ convention.  Both require libraries, and would

Re: [Nfs-ganesha-devel] nss_getpwnam: name 't...@my.dom@localdomain' does not map into domain 'nix.my.dom'

Thanks, Tomk. PR is here: https://review.gerrithub.io/404945 Daniel On 03/22/2018 05:39 PM, TomK wrote: On 3/22/2018 12:50 PM, Daniel Gryniewicz wrote: A side note, happy to test for you guy's once you have this done. No, we can't.  I'm working on a set of macros that make thi

Re: [Nfs-ganesha-devel] nss_getpwnam: name 't...@my.dom@localdomain' does not map into domain 'nix.my.dom'

e. For example, USE_GPFS_FSAL could be used at cmake command line and GPFS_FSAL could be used in the option(). Can't we use this option() inside conditionals? regards, malahal. On Thu, Mar 22, 2018 at 8:43 PM, Daniel Gryniewicz <mailto:d...@redhat.com>> wrote: I don't think thi

Re: [Nfs-ganesha-devel] nss_getpwnam: name 't...@my.dom@localdomain' does not map into domain 'nix.my.dom'

e this feature if required packages are installed.          case D:  We don't care, just disable } I don't know if DEFINED keyword works the way I want it though. Note that case A is the only one that fails here. Regards, Malahal. On Thu, Mar 22, 2018 at 5:33 PM, Daniel Gryni

Re: [Nfs-ganesha-devel] nss_getpwnam: name 't...@my.dom@localdomain' does not map into domain 'nix.my.dom'

elopers don't always use; a good extra sanity check ahead of packaging for Fedora and Ubuntu (to name two.) My 2¢. And having said all that, using STRICT_PACKAGE sounds like a good idea to me. -- Kaleb On 03/22/2018 08:03 AM, Daniel Gryniewicz wrote: So, there is an option STRICT_PACKAG

Re: [Nfs-ganesha-devel] nss_getpwnam: name 't...@my.dom@localdomain' does not map into domain 'nix.my.dom'

tion's requirements. Can some cmake guru fix this first? On Tue, Mar 20, 2018 at 8:38 PM, Daniel Gryniewicz <mailto:d...@redhat.com>> wrote: It's probably a good idea to add the build options to --version output, or something.  That way we can ask for it in these types o

Re: [Nfs-ganesha-devel] nss_getpwnam: name 't...@my.dom@localdomain' does not map into domain 'nix.my.dom'

It's probably a good idea to add the build options to --version output, or something. That way we can ask for it in these types of situations. I've added a card to the wishlist for this. Daniel On Tue, Mar 20, 2018 at 9:39 AM, TomK wrote: > On 3/19/2018 9:54 AM, Frank Filz wrote: >>> >>> Solve

[Nfs-ganesha-devel] Release V2.6.1

I'd like to announce the release of the latest stable version of Ganesha: 2.6.1 This is a bugfix release, and includes the following commits: e2664ee61 (HEAD -> V2.6-stable, tag: V2.6.1) V2.6.1 cf0b642d0 Pullup to ntirpc 1.6.2 1a6daaf34 PROXY: add sample config file 9f4ef2c96 MDCACHE - Initialize

Re: [Nfs-ganesha-devel] Change in ffilz/nfs-ganesha[next]: Pullup NTIRPC through #124

Yeah, except for drastic changes, we cannot take much away from variations in Jenkins runs, I think. Even if we know what hardware it's running on, we can't know how loaded it is. Daniel On Tue, Mar 20, 2018 at 4:59 AM, Niels de Vos wrote: > On Tue, Mar 20, 2018 at 08:59:21AM +0100, Dominique M

Re: [Nfs-ganesha-devel] rpcping

100k is a much more accurate measurement. I haven't gotten any crashes since the fixes from yesterday, but I can keep trying. On Thu, Mar 15, 2018 at 12:10 PM, William Allen Simpson wrote: > On 3/15/18 10:23 AM, Daniel Gryniewicz wrote: >> >> Can you try again with a lar

Re: [Nfs-ganesha-devel] rpcping

Can you try again with a larger count, like 100k? 500 is still quite small for a loop benchmark like this. Daniel On Thu, Mar 15, 2018 at 9:02 AM, William Allen Simpson wrote: > On 3/14/18 3:33 AM, William Allen Simpson wrote: >> >> rpcping tcp localhost threads=1 count=500 (port=2049 program=1

Re: [Nfs-ganesha-devel] rpcping

rpcping was not thread safe. I have fixes for it incoming. Daniel On 03/13/2018 12:13 PM, William Allen Simpson wrote: On 3/13/18 2:38 AM, William Allen Simpson wrote: In my measurements, using the new CLNT_CALL_BACK(), the client thread starts sending a stream of pings.  In every case, it pe

Re: [Nfs-ganesha-devel] Better late than never - US Daylight Savings Time has started and that means weekly conference call is an hour earlier

An hour later... Daniel On 03/13/2018 10:02 AM, Frank Filz wrote: -- Check out the vibrant tech community on one of the world's most engaging tech sites, Slashdot.org! http://sdm.link/slashdot _

Re: [Nfs-ganesha-devel] nfsv3 client writing file gets Invalid argument on glusterfs with quota on

On 03/07/2018 10:21 PM, Kinglong Mee wrote: On 2018/3/7 21:10, Daniel Gryniewicz wrote: On 03/06/2018 10:10 PM, Kinglong Mee wrote: On 2018/3/7 10:59, Kinglong Mee wrote: When using nfsv3 on glusterfs-3.13.1-1.el7.x86_64 and nfs-ganesha-2.6.0-0.2rc3.el7.centos.x86_64, I gets strange "In

Re: [Nfs-ganesha-devel] nfsv3 client writing file gets Invalid argument on glusterfs with quota on

On 03/06/2018 10:10 PM, Kinglong Mee wrote: On 2018/3/7 10:59, Kinglong Mee wrote: When using nfsv3 on glusterfs-3.13.1-1.el7.x86_64 and nfs-ganesha-2.6.0-0.2rc3.el7.centos.x86_64, I gets strange "Invalid argument" when writing file. 1. With quota disabled; nfs client mount nfs-ganesha share,

Re: [Nfs-ganesha-devel] Multiprotocol support in ganesha

/master/source3/smbd/oplock_linux.c Thanks, On Tue, Mar 6, 2018 at 9:29 AM, Daniel Gryniewicz <mailto:d...@redhat.com>> wrote: Ganesha has multi-protocol (NFS3, NFS4, and 9P).  There are no plans to add CIFS, since that is an insanely complicated protocol, and has a userspace d

Re: [Nfs-ganesha-devel] Multiprotocol support in ganesha

Ganesha has multi-protocol (NFS3, NFS4, and 9P). There are no plans to add CIFS, since that is an insanely complicated protocol, and has a userspace daemon implementation already (in the form of Samba). I personally wouldn't reject such support if it was offered, but as far as I know, no one

Re: [Nfs-ganesha-devel] nss_getpwnam: name 't...@my.dom@localdomain' does not map into domain 'nix.my.dom'

Based on the error messages, you client is not sending t...@nix.my.dom but is sending t...@my.dom@localdomain. Something is mis-configured on the client. Have you tried having identical (including case) idmapd.conf files on both the client and server? Idmap configuration has historically be

Re: [Nfs-ganesha-devel] Issue with large directories containing 100, 000 files

Hi. What version of Ganesha is this? Daniel On 03/05/2018 10:35 PM, Varghese Devassy via Nfs-ganesha-devel wrote: Hello, I am testing our own version of FSAL and I am observing an issue with directories containing 100,000 files. When I do ls on the directory, it only lists about 21 files. T

Re: [Nfs-ganesha-devel] inconsistent '*' spacing

We check this because consistent, readable style is important for maintainability. It's being checked now because commit hooks only operate on changed code. Daniel On 02/16/2018 01:01 PM, William Allen Simpson wrote: As I'm trying to update nfs41.h, I've run into the problem that the commit

Re: [Nfs-ganesha-devel] ls returns NOENT

Mdcache is working correctly here. It tried to find it in the cache, and failed. It got the write lock, and tried again, and failed. Then it called mdc_lookup_uncached(), which called into the FSAL, and got back ENOENT. So the underlying FSAL is to blame here, I believe. Daniel On 02/17/2

Re: [Nfs-ganesha-devel] nfs ganesha vs nfs kernel performance

ha CPU usage ~250% ( ~50K IOPS, latency ~2ms). -Deepak -------- *From:* Daniel Gryniewicz *Sent:* Tuesday, February 13, 2018 6:15:47 AM *To:* nfs-ganesha-devel@lists.sourceforge.net *Subject:* Re: [Nfs-ganesha-devel] nfs ganesha

Re: [Nfs-ganesha-devel] nfs ganesha vs nfs kernel performance

Also keep in mind that FSAL VFS can never, by it's very nature, beat knfsd, since it has to do everything knfsd does, but also has userspace <-> kernespace transitions. Ganesha's strength is exporting userspace-based cluster filesystems. That said, we're always working to make Ganesha faster,

Re: [Nfs-ganesha-devel] How do I compile nfs-ganesha-utils RPM

The Ceph project also maintains RPMs compatible with RHEL, if you'd prefer a Ceph-flavored version (also works without Ceph, of course, just like the Gluster versions). http://download.ceph.com/nfs-ganesha/rpm-V2.5-stable/luminous/x86_64/ Daniel On 02/09/2018 02:53 PM, Daniel Gryni

Re: [Nfs-ganesha-devel] How do I compile nfs-ganesha-utils RPM

rked. > > How do I build libntirpc RPM? > > On Fri, Feb 9, 2018 at 10:42 AM, Daniel Gryniewicz wrote: >> >> pass -DUSE_ADMIN_TOOLS=YES to your cmake command. I think you can >> also pass --with utils to rpmbuild. >> >> Daniel >> >> On Fri, F

Re: [Nfs-ganesha-devel] How do I compile nfs-ganesha-utils RPM

pass -DUSE_ADMIN_TOOLS=YES to your cmake command. I think you can also pass --with utils to rpmbuild. Daniel On Fri, Feb 9, 2018 at 10:28 AM, You Me wrote: > I compiled ganesha 2.5.1 from sources. 'make rpm' gave me all RPMs except > ganesha-utils. > How do I build that one? > > Thank you > > >

Re: [Nfs-ganesha-devel] Using root privileges when using kerberos exports with Ganesha.

I haven't done it, but I think it works if you have idmapping set up correctly. That is, if the idmapper domain is correct, then the client will send "root@DOMAIN" and the idmapd on the ganesha server will convert that to UID 0. Daniel On Thu, Feb 8, 2018 at 6:45 PM, Pradeep wrote: > Hello, > >

Re: [Nfs-ganesha-devel] owner and group issue.

This is not a known issue, as far as I'm aware. I'm assuming you never changed 4.txt at all, it just fixed itself on it's own? This means, to me, that the correct value is being set on the file, but an incorrect one is returned to the user. That's a getattr() issue, and it resolves itself be

Re: [Nfs-ganesha-devel] WIP example API for async/vector FSAL ops

On 02/07/2018 09:05 AM, William Allen Simpson wrote: On 2/6/18 10:40 AM, Daniel Gryniewicz wrote: On 02/06/2018 10:26 AM, William Allen Simpson wrote: On 2/6/18 8:25 AM, Daniel Gryniewicz wrote: Hi, all. I've worked up a sample API for async/vector for FSAL ops.  The example op is

Re: [Nfs-ganesha-devel] WIP example API for async/vector FSAL ops

On 02/06/2018 10:26 AM, William Allen Simpson wrote: On 2/6/18 8:25 AM, Daniel Gryniewicz wrote: Hi, all. I've worked up a sample API for async/vector for FSAL ops.  The example op is read(), and I've "implemented" it for all FSALs, so that I can verify that it does, in

Re: [Nfs-ganesha-devel] reg. unbounded readdir cache

It's the dirent cache. There's a bounded version in 2.5 and later. Daniel On 02/06/2018 07:18 AM, Suresh kosuru wrote: Hi, I am working on NFS Ganesha v2.3. From what I have heard, readdir cache is unbounded in Ganesha. Can you please elaborate on which cache is unbounded. Is it the inode c

[Nfs-ganesha-devel] WIP example API for async/vector FSAL ops

Hi, all. I've worked up a sample API for async/vector for FSAL ops. The example op is read(), and I've "implemented" it for all FSALs, so that I can verify that it does, in fact, work for some definition of work. The commit is the top of this branch: https://github.com/dang/nfs-ganesha/tre

Re: [Nfs-ganesha-devel] Crash in graceful shutdown.

ganesha.nfsd) will reproduce it. But you might hit a double-free problem before that - here is a patch that fixes it. https://review.gerrithub.io/#/c/398092/ Feel free to rework the patch above. On Thu, Feb 1, 2018 at 6:59 AM, Daniel Gryniewicz <mailto:d...@redhat.com>> wrote:

Re: [Nfs-ganesha-devel] Crash in graceful shutdown.

I've been actually deliberately leaving that crash in. It indicates a refcount leak (and is usually the only indicator of a refcount leak, other than slowly growing memory over a long time). Can you get me a reproducer for this? If so, I can track down the leak. Daniel On 02/01/2018 09:51 A

Re: [Nfs-ganesha-devel] Correct initialization sequence

On 01/31/2018 10:27 AM, William Allen Simpson wrote: On 1/31/18 8:44 AM, Daniel Gryniewicz wrote: Agreed. Daniel On 01/30/2018 11:46 PM, Malahal Naineni wrote: Looking at the code, dupreq2_pkginit() only depends on Ganesha config processing to initialize few things, so it should be OK to

Re: [Nfs-ganesha-devel] Features board list

On 01/31/2018 10:21 AM, William Allen Simpson wrote: On 1/30/18 12:03 PM, Supriti Singh wrote: Hello all, As discussed in community call, I am sharing the feature board list: https://github.com/nfs-ganesha/nfs-ganesha/projects for nfs-ganesha 2.6 and 2.7. The aim is to use these boards to tra

Re: [Nfs-ganesha-devel] Correct initialization sequence

Agreed. Daniel On 01/30/2018 11:46 PM, Malahal Naineni wrote: Looking at the code, dupreq2_pkginit() only depends on Ganesha config processing to initialize few things, so it should be OK to call anytime after Ganesha config processing. Regards, Malahal. On Wed, Jan 31, 2018 at 8:00 AM, Pra

Re: [Nfs-ganesha-devel] Change in ffilz/nfs-ganesha[next]: Pullup ntirpc 1.6

On 01/30/2018 09:11 AM, William Allen Simpson wrote: On 1/29/18 2:27 PM, Daniel Gryniewicz wrote: On 01/29/2018 02:09 PM, William Allen Simpson wrote: On 1/29/18 1:13 PM, GerritHub wrote: Daniel Gryniewicz has uploaded this change for *review*. View Change <https://review.gerrithub.io/397

Re: [Nfs-ganesha-devel] 'missed' recv with 2.6-rc2?

It's already in ntirpc, and we'll submit a pullup for this week. Daniel On 01/29/2018 12:53 PM, Pradeep wrote: Hi Bill, Are you planning to pull this into the next ganesha RC? Thanks, Pradeep On Sun, Jan 28, 2018 at 7:13 AM, William Allen Simpson > wr

Re: [Nfs-ganesha-devel] 'missed' recv with 2.6-rc2?

I don't think you re-arm a FD in epoll. You arm it once, and it fires until you disarm it, as far as I know. You just call epoll_wait() to get new events. The thread model is a bit odd; When the epoll fires, all the events are found, and a thread is submitted for each one except one. That

Re: [Nfs-ganesha-devel] ganesha crash when stopping with gluster handles exist

How about this as a band-aid, since non-chunked readdir is going away: https://review.gerrithub.io/396194 Daniel On 01/24/2018 05:29 AM, Kinglong Mee wrote: With the latest codes (nfs-ganesha-2.6.0-0.2rc3.el7.centos.x86_64), set ganesha.conf includes, CACHEINODE { Dir_Max = 50; Dir_

Re: [Nfs-ganesha-devel] NULL pointer deref in clnt_ncreate_timed()

Hi, Pradeep. Can you try with this patch on ntirpc? https://github.com/nfs-ganesha/ntirpc/pull/105 Thanks, Daniel On 01/22/2018 08:08 PM, Pradeep wrote: Hello, I'm running into a crash in libntirpc with rc2: #2  #3  0x7f9004de31f4 in clnt_ncreate_timed (hostname=0x57592e "localhost",

Re: [Nfs-ganesha-devel] NULL pointer deref in clnt_ncreate_timed()

On 01/23/2018 09:04 AM, William Allen Simpson wrote: On 1/22/18 8:08 PM, Pradeep wrote: Hello, I'm running into a crash in libntirpc with rc2: #2  #3  0x7f9004de31f4 in clnt_ncreate_timed (hostname=0x57592e "localhost", prog=100024, vers=1,      netclass=0x57592a "tcp", tp=0x0) at /usr/

Re: [Nfs-ganesha-devel] Segfault in dupreq cache.

Hi, Pradeep. There's been 6 sets of code changes to dupreq since dev-22, including refcount bugfixes. Can you try with 2.6-rc2? Daniel On 01/13/2018 08:50 PM, Pradeep wrote: Hello, I'm seeing a segfault in nfs_dupreq_finish() with 2.6.dev.22. This is when using NFSv4.0 clients. The dupreq_

Re: [Nfs-ganesha-devel] Possibility of converting FSAL_VFS and FSAL_XFS to use stackable FSALS

On 01/12/2018 10:08 AM, Frank Filz wrote: So, the linux/freebsd split was not something I was considering for stacking. I agree, that doesn't match well. But the VFS subfsal API looks like it can just go away and be a stack instead. And I certainly would not like to see it extended. Y

Re: [Nfs-ganesha-devel] NFS-Ganesha and compatible ceph version

raham Norton, HRB 21284 (AG Nürnberg) Daniel Gryniewicz  01/09/18 3:45 PM >>> Yes, 2.7 will be compatible with mimic, and likely with luminous as well, since features for Ganesha tend to be backported in Ceph. The only major work I'm aware of coming for Ceph/Ganesha is HA on OpenSh

Re: [Nfs-ganesha-devel] Possibility of converting FSAL_VFS and FSAL_XFS to use stackable FSALS

So, the linux/freebsd split was not something I was considering for stacking. I agree, that doesn't match well. But the VFS subfsal API looks like it can just go away and be a stack instead. And I certainly would not like to see it extended. Daniel On 01/09/2018 06:19 PM, Frank Filz wrote:

Re: [Nfs-ganesha-devel] NFS-Ganesha and compatible ceph version

Yes, 2.7 will be compatible with mimic, and likely with luminous as well, since features for Ganesha tend to be backported in Ceph. The only major work I'm aware of coming for Ceph/Ganesha is HA on OpenShift/Kuberneties for FSAL_CEPH, and full in-place read/write for FSAL_RGW. I'm sure lots of li

Re: [Nfs-ganesha-devel] [FSAL_VFS] Probable memory leak when deallocating handles

I would actually like to see the sub-FSAL removed, and xfs converted to a stacked FSAL at some point, now that patfs is dead. But it's not high on my list (especially with testing xfs...) Daniel On Mon, Jan 8, 2018 at 3:15 PM, Frank Filz wrote: >> The problem is that the sub-FSAL interface was

Re: [Nfs-ganesha-devel] [FSAL_VFS] Probable memory leak when deallocating handles

The problem is that the sub-FSAL interface was never intended to be a general interface. It was added specifically to allow certain features to be added to panfs, so it only has the entrypoints needed for that work. Realistically, if I were implementing it now, I'd use a stacked FSAL, but at

Re: [Nfs-ganesha-devel] Implement a FSAL for S3-compatible storage

On 12/15/2017 12:48 PM, Frank Filz wrote: From: Aurelien RAINONE [mailto:aurelien.rain...@gmail.com] Sent: Friday, December 15, 2017 9:30 AM To: Nfs-ganesha-devel@lists.sourceforge.net Subject: [Nfs-ganesha-devel] Implement a FSAL for S3-compatible storage Hi, I'm new to nfs-ganesha and on this

Re: [Nfs-ganesha-devel] dev.20 segfault on shutdown

This is why I've never bothered to fix the crashes when desrtoy_fsals() finds things; it's an indication to me that something is leaking refs. On Tue, Dec 12, 2017 at 7:55 PM, Frank Filz wrote: >> I was testing code I'd written over the weekend, but it segfaulted on >> shutdown after running pynf

Re: [Nfs-ganesha-devel] test over FSAl_NULL

Okay, with this fix, stacking NULL works for me: https://review.gerrithub.io/391463 Daniel On 12/12/2017 11:52 AM, Daniel Gryniewicz wrote: Okay, I'm able to reproduce. I'm looking at this, but the problem is that the export being set before mdcache is called is NULL

Re: [Nfs-ganesha-devel] test over FSAl_NULL

aniel On 12/08/2017 10:59 AM, Daniel Gryniewicz wrote: I run NULL semi-regularly. The last time I ran it was a couple of months ago, so something may have crept in. I'll try again. That said, the code in that callpath looks correct. Daniel On 12/08/2017 05:46 AM, LUCAS Patrice wrote: H

Re: [Nfs-ganesha-devel] Stacked FSALs and fsal_export parameters and op_ctx

fsal_export should probably be set anywhere gsh_export is set. Daniel On 12/07/2017 07:54 PM, Frank Filz wrote: Stacked FSALs often depend on op_ctx->fsal_export being set. We also have lots of FSAL methods that take the fsal_export as a parameter. I wonder if we would be better off removing

Re: [Nfs-ganesha-devel] Stacked FSALs and fsal_export parameters and op_ctx

I'm on the opposite end. I'm not in favor of passing op_ctx needlessly to every function in Ganesha. Any useful threading system must have some form of TLS. Daniel On 12/08/2017 10:13 AM, Matt Benjamin wrote: I'd like to see this use of TLS as a "hidden parameter" replaced regardless. It ha

Re: [Nfs-ganesha-devel] test over FSAl_NULL

I run NULL semi-regularly. The last time I ran it was a couple of months ago, so something may have crept in. I'll try again. That said, the code in that callpath looks correct. Daniel On 12/08/2017 05:46 AM, LUCAS Patrice wrote: Hi, Does anyone recently test the FSAL_NULL stackable FSAL

Re: [Nfs-ganesha-devel] Proposal: new scripts repo

On 11/22/2017 08:04 AM, LUCAS Patrice wrote: On 11/21/17 15:20, Daniel Gryniewicz wrote: Hi, All. I would like to propose a new repo on our github for helpful scripts. I'd like to have testing/performance scripts there, as a start, but maybe other useful things can go there too in the f

[Nfs-ganesha-devel] Proposal: new scripts repo

Hi, All. I would like to propose a new repo on our github for helpful scripts. I'd like to have testing/performance scripts there, as a start, but maybe other useful things can go there too in the future. The idea is that this would be a place to share things like setups for running dbench/i

Re: [Nfs-ganesha-devel] Exporting several cephfs filesystems

Hi, Alessandro Currently, there's no way to address more than one ceph cluster in a single Ganesha. The ceph configuration is in the global CEPH block, which is a singleton, shared between all cephfs exports. Typically, the way you would do this would be to run a ganesha server per ceph clu

Re: [Nfs-ganesha-devel] Subfsal export

king the implementation > > opensource does not violate any license terms. > > > > Thanks, > > Sriram > > > > On 08-Nov-2017 8:19 PM, "Daniel Gryniewicz" mailto:d...@redhat.com>

Re: [Nfs-ganesha-devel] crash in jemalloc leading to a deadlock.

Allocating in a backtrace seems like a very bad idea. If there's ever a crash during an allocation, it is guaranteed to deadlock. Daniel On 11/08/2017 01:43 PM, Pradeep wrote: I'm using Ganesha 2.6 dev.12 with jemalloc-3.6.0 and hitting a case where jemalloc seem to be holding a lock and cras

Re: [Nfs-ganesha-devel] Subfsal export

so, I hope creating a sub fsal and not making the implementation opensource does not violate any license terms. Thanks, Sriram On 08-Nov-2017 8:19 PM, "Daniel Gryniewicz" <mailto:d...@redhat.com>> wrote: On 11/08/2017 02:41 AM, sriram patil wrote: Hi, In the

Re: [Nfs-ganesha-devel] Subfsal export

On 11/08/2017 02:41 AM, sriram patil wrote: Hi, In the subfsal framework, I see that subfsals can have their own fsal_obj_handles by implementing vfs_sub_alloc_handle and then use subfsal specific variables using container_of. It does not provide same functionality for fsal_export however. T

Re: [Nfs-ganesha-devel] Backport list for 2.5.4

ed for the recovery code (aka IP failover code) reorganization patches to go into V2.6 alone. Do they now have enough runtime to get merged into V2.5 stable branch? Regards, Malahal. On Tue, Oct 31, 2017 at 11:37 PM, Daniel Gryniewicz <mailto:d...@redhat.com>> wrote: Here's

Re: [Nfs-ganesha-devel] CI failures

On 11/02/2017 11:46 AM, Frank Filz wrote: Ok, so this patch: https://review.gerrithub.io/#/c/385433/ has a real failure visible, however, it clearly has nothing to do with the patch at hand. How do we want to handle that for merge? The patch clearly is ready for merge, but with a -1 Verify, if w

[Nfs-ganesha-devel] Backport list for 2.5.4

DOS_URLS block. Change-Id: I1e81eae784de09abb3d1fbf715bb05eac6e8333e Signed-off-by: Matt Benjamin commit c4a05dedc13de3bae73b17b040be147ca520 Author: Daniel Gryniewicz AuthorDate: Tue Oct 24 13:07:40 2017 -0400 Commit: Daniel Gryniewicz CommitDate: Tue Oct 24 13:07:40 2017 -040

Re: [Nfs-ganesha-devel] Ganesha 2.3 and 2.5 - crash in free_nfs_request

This sounds like a use-after-free. That memory is either poisoned by the allocator, or re-allocated and used by something that memset() it. You can try running under valgrind, if it's fast enough, or you can try backporting the ASAN additions from these commits: 1b48c0237c10c48d33840ab278d5c

Re: [Nfs-ganesha-devel] ABBA deadlock in 2.5 (likely in 2.6 as well)

Maybe something like this: https://paste.fedoraproject.org/paste/CptGkmoRutBKYjno5FiSjg/ Daniel On 10/23/2017 10:13 AM, Malahal Naineni wrote: Let us say we have X/Y path in the file system. We have attr_lock and content_lock on each object. The locks we are interested here are attr_lock on X

Re: [Nfs-ganesha-devel] Crash in mdcache_alloc_handle() during unexport

This looks like a legit issue. I'm working on a fix for it. Daniel On 10/05/2017 01:07 PM, Pradeep wrote: Hello, This issue is with 2.6-dev.11 (don't think it is specific to this version). It appears that there is a race between mdcache_unexport() and mdcache_alloc_handle(). If an request c

Re: [Nfs-ganesha-devel] Multi instance Ganesha with DBus

running inside containers. Thanks, Sriram On 03-Oct-2017 5:35 PM, "Daniel Gryniewicz" <mailto:d...@redhat.com>> wrote: On 10/03/2017 02:31 AM, sriram patil wrote: Hi, AFAIK we can run only single instance of nfs-ganesha on a given machine which s

Re: [Nfs-ganesha-devel] Pull up NTIRPC #80 & #81

On 10/03/2017 12:47 PM, William Allen Simpson wrote: https://review.gerrithub.io/#/c/380970/ Begging for a mid-week dev release. These patches attempt to fix a crash found early in Bake-a-thon. DanG and I couldn't reproduce, so we need this to determine whether it has fixed the QE crash -- so

  1   2   3   4   >