Re: [PATCH 1/2] libceph: request a new osdmap if lingering request maps to no osd

On Wed, 13 May 2015, Ilya Dryomov wrote: This commit does two things. First, if there are any homeless lingering requests, we now request a new osdmap even if the osdmap that is being processed brought no changes, i.e. if a given lingering request turned homeless in one of the previous epochs

ceph-osd: microbenchmark for 4k read operations

Hi, as discussed on todays performance meeting: find attached the spreadsheet which I had shown during the meeting. @Sam: your are right concerning the make_blist part: this probably is expensive due to message signing being turned on. For the osd daemon which I took the benchmarks, I see

Re: Slow file creating and deleting using bonnie ++ on Hammer

On Wed, May 20, 2015 at 11:15 AM, Barclay Jameson almightybe...@gmail.com wrote: I am trying to find out why boniee++ is choking at the creating files sequentially and deleting sequentially on cephfs. I enabled mds debug for about 30 seconds and I find a bunch of lines like the following:

firewall questions

It would be really convenient to have human-readable firewalld service definitions for Ceph, so that users could do things like: firewall-cmd --add-service=ceph-mon or firewall-cmd --add-service=ceph ... instead of having to know specific port numbers to open. In order to submit service

Re: RBD mirroring design draft

On 05/18/2015 09:22 AM, Chris H wrote: I am actually working on something very similar to this for another project. Writing very small sequential IO groups with flushes to the cloud is very slow. The structure I am working on is nearly identical as well (I originally did padding, but might be

Re: [PATCH 1/2] libceph: request a new osdmap if lingering request maps to no osd

On Wed, 20 May 2015, Ilya Dryomov wrote: On Wed, May 20, 2015 at 8:29 PM, Sage Weil s...@newdream.net wrote: On Wed, 13 May 2015, Ilya Dryomov wrote: This commit does two things. First, if there are any homeless lingering requests, we now request a new osdmap even if the osdmap that is

Slow file creating and deleting using bonnie ++ on Hammer

I am trying to find out why boniee++ is choking at the creating files sequentially and deleting sequentially on cephfs. I enabled mds debug for about 30 seconds and I find a bunch of lines like the following: 2015-05-19 15:54:12.930829 7f6de4ecf700 1 -- 192.168.40.3:6800/461756 --

Re: [PATCH 2/2] Revert libceph: clear r_req_lru_item in __unregister_linger_request()

On Wed, 13 May 2015, Ilya Dryomov wrote: This reverts commit ba9d114ec5578e6e99a4dfa37ff8ae688040fd64. .. which introduced a regression that prevented all lingering requests requeued in kick_requests() from ever being sent to the OSDs, resulting in a lot of missed notifies. In retrospect

Re: [PATCH 1/2] libceph: request a new osdmap if lingering request maps to no osd

On Wed, May 20, 2015 at 8:29 PM, Sage Weil s...@newdream.net wrote: On Wed, 13 May 2015, Ilya Dryomov wrote: This commit does two things. First, if there are any homeless lingering requests, we now request a new osdmap even if the osdmap that is being processed brought no changes, i.e. if a

RE: journal writes when running rados bench seq on ec pool

Hi -- I never saw an explanation for the writes that are occurring during ecpool reads. -- Tom -Original Message- From: Somnath Roy [mailto:somnath@sandisk.com] Sent: Wednesday, May 13, 2015 6:48 PM To: Deneau, Tom; ceph-devel Subject: RE: journal writes when running rados

Re: Slow file creating and deleting using bonnie ++ on Hammer

I dumped it out into my github wiki under the pages ceph MDS Log and Ceph s output: https://github.com/almightybeeij/Debug/wiki Let me know if you would like anything else or if you can't get to it. Thanks Greg! On Wed, May 20, 2015 at 1:22 PM, Gregory Farnum g...@gregs42.com wrote: On Wed,

Re: firewall questions

On Wed, 20 May 2015, Ken Dreyer wrote: It would be really convenient to have human-readable firewalld service definitions for Ceph, so that users could do things like: firewall-cmd --add-service=ceph-mon or firewall-cmd --add-service=ceph ... instead of having to know specific

MDS auth caps for cephfs

Looking at the MDSAuthCaps again, I think there are a few things we might need to clean up first. The way it is currently structured, the idea is that you have an array of grants (MDSCapGrant). For any operation, you'd look at each grant until one that says what you're trying to do is okay.

excluding /var/lib/ceph in updatedb.conf

Hi Dan, I watched your OpenStack Summit conference where you talked about excluding /var/lib/ceph from /etc/updatedb.conf. Did you open any bugs for that? That change seems like something we should get into the distros. - Ken -- To unsubscribe from this list: send the line unsubscribe

Re: excluding /var/lib/ceph in updatedb.conf

Woah, and Ken Dreyer even commented on that Redmine ticket there, long ago :) I will take that ticket and run this down with Fedora now. https://bugzilla.redhat.com/1223582 - Ken On 05/20/2015 04:43 PM, Dan van der Ster wrote: Hi Ken, Not sure, here's what I could find: The original

Re: Package build from master busted on Ubuntu Vivid

On 16/05/15 19:53, Mark Kirkwood wrote: Hi, While a standard build (configure, make etc) on $platform works fine, attempting to build packages gets: $ dpkg-buildpackage -j4 dpkg-buildpackage: source package ceph dpkg-buildpackage: source version 9.0.0-1 dpkg-buildpackage: source

[PATCH] Rbd: copy-on-read support for kernel rbd client

This is a new feature of rbd layering, when reading an object from child, if not exist, the kernel rbd client will not only request parent for the object, but also write it to child, and the jobs are done in an asynchronous way. Therefore, the subsequent accesses on this object will hit child