Re: OSD suicide after being down/in for one day as it needs to search large amount of objects

2014-08-20 Thread Guang Yang
Thanks Greg. On Aug 20, 2014, at 6:09 AM, Gregory Farnum g...@inktank.com wrote: On Mon, Aug 18, 2014 at 11:30 PM, Guang Yang yguan...@outlook.com wrote: Hi ceph-devel, David (cc’ed) reported a bug (http://tracker.ceph.com/issues/9128) which we came across in our test cluster during our

Re: RadosGW crashing on copy for one specific object

2014-08-20 Thread Sylvain Munaut
Hi, What does 'radosgw-admin object stat --bucket=bucket --object=object' show? { name: 5ae1b8cb8a2bdc3c2d7e1868b60d76abea2536f4604d6d312df95b719470fb3b\/render-image, size: 239879, policy: { acl: { acl_user_map: [ { user: kp, acl: 15}],

Re: OSD suicide after being down/in for one day as it needs to search large amount of objects

2014-08-20 Thread Sage Weil
On Wed, 20 Aug 2014, Guang Yang wrote: Thanks Greg. On Aug 20, 2014, at 6:09 AM, Gregory Farnum g...@inktank.com wrote: On Mon, Aug 18, 2014 at 11:30 PM, Guang Yang yguan...@outlook.com wrote: Hi ceph-devel, David (cc?ed) reported a bug (http://tracker.ceph.com/issues/9128) which we

RE: Deadlock in ceph journal

2014-08-20 Thread Sage Weil
I suspect what is really needed is a drain_aio() function that will wait for all pending aio ops to complete on shutdown. What happens to those IOs if the process exists while they are in flight is probably undefined; we should just avoid doing that. sage On Wed, 20 Aug 2014, Somnath Roy

Re: RadosGW crashing on copy for one specific object

2014-08-20 Thread Yehuda Sadeh
On Wed, Aug 20, 2014 at 6:47 AM, Sylvain Munaut s.mun...@whatever-company.com wrote: Hi, What does 'radosgw-admin object stat --bucket=bucket --object=object' show? { name: 5ae1b8cb8a2bdc3c2d7e1868b60d76abea2536f4604d6d312df95b719470fb3b\/render-image, size: 239879, policy: { acl:

Re: Translating a RadosGW object name into a filename on disk

2014-08-20 Thread Craig Lewis
Looks like I need to upgrade to Firefly to get ceph-kvstore-tool before I can proceed. I am getting some hits just from grepping the LevelDB store, but so far nothing has panned out. Thanks for the help! On Tue, Aug 19, 2014 at 10:27 AM, Gregory Farnum g...@inktank.com wrote: It's been a while

Re: RadosGW crashing on copy for one specific object

2014-08-20 Thread Yehuda Sadeh
Looking at the code some more, that change isn't going to fix it for you. I'll send an updated fix soon. Yehuda On Wed, Aug 20, 2014 at 10:25 AM, Yehuda Sadeh yeh...@redhat.com wrote: On Wed, Aug 20, 2014 at 6:47 AM, Sylvain Munaut s.mun...@whatever-company.com wrote: Hi, What does

Re: Translating a RadosGW object name into a filename on disk

2014-08-20 Thread Sage Weil
On Wed, 20 Aug 2014, Craig Lewis wrote: Looks like I need to upgrade to Firefly to get ceph-kvstore-tool before I can proceed. I am getting some hits just from grepping the LevelDB store, but so far nothing has panned out. FWIW if you just need the tool, you can wget the .deb and 'dpkg -x

Re: RadosGW crashing on copy for one specific object

2014-08-20 Thread Yehuda Sadeh
If by any chance you applied the previous patch, revert it, as it's wrong. This might fix the issue: diff --git a/src/rgw/rgw_rados.h b/src/rgw/rgw_rados.h index d50fb59..0f13590 100644 --- a/src/rgw/rgw_rados.h +++ b/src/rgw/rgw_rados.h @@ -298,6 +298,9 @@ public: bool has_tail() { if

Re: [Ceph-maintainers] disabling updatedb

2014-08-20 Thread Dan Mick
Just adding a note in case you hadn't noticed that updatedb itself has a CLI for managing the .conf: --add-prune{fs,names,paths}. Sadly, there is no --remove, but at least it lets the conf file format be abstract. +1 on everything has a .d/ dir though. On 02/20/2014 10:47 AM, Sage Weil wrote:

Re: Ceph authentication/authorization paradignms

2014-08-20 Thread Christopher R. Hertel
[At end...] - Original Message - From: Gregory Farnum g...@inktank.com To: Christopher R. Hertel c...@redhat.com Cc: ceph-devel@vger.kernel.org Sent: Tuesday, August 19, 2014 4:57:59 PM Subject: Re: Ceph authentication/authorization paradignms On Thu, Aug 14, 2014 at 10:10 AM,

Re: OSD suicide after being down/in for one day as it needs to search large amount of objects

2014-08-20 Thread Guang Yang
Thanks Sage. We will provide a patch based on this. Thanks, Guang On Aug 20, 2014, at 11:19 PM, Sage Weil sw...@redhat.com wrote: On Wed, 20 Aug 2014, Guang Yang wrote: Thanks Greg. On Aug 20, 2014, at 6:09 AM, Gregory Farnum g...@inktank.com wrote: On Mon, Aug 18, 2014 at 11:30 PM, Guang

RE: Deadlock in ceph journal

2014-08-20 Thread Ma, Jianpeng
Yes, Maybe for io_submit, it must use io_getevent. Otherwise the result is undefined. If stop_write == true, we don't use aio. How about this way? Jianpeng -Original Message- From: Sage Weil [mailto:sw...@redhat.com] Sent: Wednesday, August 20, 2014 11:34 PM To: Somnath Roy Cc:

RE: Deadlock in ceph journal

2014-08-20 Thread Sage Weil
On Thu, 21 Aug 2014, Ma, Jianpeng wrote: Yes, Maybe for io_submit, it must use io_getevent. Otherwise the result is undefined. If stop_write == true, we don't use aio. How about this way? That seems reasonable, now that I understand why it doesn't work the other way. Do you mind resending