Re: wip-libcephfs rebased and pulled up to v.65

From: Sage Weil <sage@inktank.com>
To: "Matt W. Benjamin" <matt@linuxbox.com>
Cc: ceph-devel <ceph-devel@vger.kernel.org>,
	aemerson <aemerson@linuxbox.com>,
	David Zafman <david.zafman@inktank.com>
Subject: Re: wip-libcephfs rebased and pulled up to v.65
Date: Wed, 3 Jul 2013 14:48:40 -0700 (PDT)	[thread overview]
Message-ID: <alpine.DEB.2.00.1307031439270.3375@cobra.newdream.net> (raw)
In-Reply-To: <515941372.160.1372801066160.JavaMail.root@thunderbeast.private.linuxbox.com>

Hi Matt,

This hit a problem in QA.. running 
ceph.git/qa/workunits/kernel_untar_build.sh on ceph-fuse crashes with

2013-07-03 12:51:14.176096 7fae7ee93780 10 client.4106 _lookup 10000004031.head(ref=4 cap_refs={} open={} mode=40775 size=0 mtime=2012-02-29 16:32:49.000000 caps=pAsLsXsFsx(0=pAsLsXsFsx) dirty_caps=Fx parents=0x9be18c0 0x9b4a000) generated = 1000000a29d.head(ref=3 cap_refs={} open={} mode=40775 size=0 mtime=2013-07-03 12:43:25.748972 caps=pAsLsXsFsx(0=pAsLsXsFsx) parents=0x8e909a0 0xa3b3d80)
2013-07-03 12:51:14.176126 7fae7ee93780 10 client.4106 fill_stat on 1000000a29d snap/devhead mode 040775 mtime 2013-07-03 12:43:25.748972 ctime 2013-07-03 12:43:25.748972
2013-07-03 12:51:14.176135 7fae7ee93780  3 client.4106 ll_lookup 0x9b4a000 generated -> 0 (1000000a29d)
2013-07-03 12:51:14.176148 7fae7ee93780  3 client.4106 ll_forget 10000004031 1
2013-07-03 12:51:14.191582 7fae7ee93780 -1 *** Caught signal (Segmentation fault) **
 in thread 7fae7ee93780

 ceph version 0.65-250-gfcf46bc (fcf46bc56dc37a90f69f6609e2ff3216d9195d2f)
 1: ceph-fuse() [0x7937fa]
 2: (()+0xfcb0) [0x7fae7e843cb0]
 3: (Client::_get_inodeno(Inode*)+0) [0x529a70]
 4: (Client::ll_forget(Inode*, int)+0x4a) [0x534a6a]
 5: ceph-fuse() [0x5261f5]
 6: (fuse_session_loop()+0x75) [0x7fae7ea61d65]
 7: (main()+0x848) [0x5228f8]
 8: (__libc_start_main()+0xed) [0x7fae7cd0676d]
 9: ceph-fuse() [0x523df9]
 NOTE: a copy of the executable, or `objdump -rdS <executable>` is needed to interpret this.

It wasn't immediately obvious what the problem is except that it looked 
like a (ll_)refcounting issue.  Also, the special-case of 'forget on root' 
looks suspicous; I'd like to pull in the patch from Ilya that takes a ref 
in the get_root_inode() method and make the refcounting uniform.

Thanks!
sage

On Tue, 2 Jul 2013, Matt W. Benjamin wrote:

> Hi Sage (et al),
> 
> We have rebased the former wip-libcephfs branch, on the model of the
> rebased example branch, as planned, and also pulled it up to Ceph's
> v65 tag/master, also as planned.
> 
> In addition to cross checking this, Adam has updated our Ganesha client
> driver to use the ll v2 API, and this checks out.
> 
> We've pushed wip-libcephfs-rebased-v65 to our public git repository,
> https://github.com/linuxbox2/linuxbox-ceph, for review.
> 
> Thanks,
> 
> Matt
> 
> -- 
> Matt Benjamin
> The Linux Box
> 206 South Fifth Ave. Suite 150
> Ann Arbor, MI  48104
> 
> http://linuxbox.com
> 
> tel.  734-761-4689 
> fax.  734-769-8938 
> cel.  734-216-5309 
> --
> To unsubscribe from this list: send the line "unsubscribe ceph-devel" in
> the body of a message to majordomo@vger.kernel.org
> More majordomo info at  http://vger.kernel.org/majordomo-info.html
> 
> 