Re: [ceph-users] RGW authentication fail with AWS S3 v4

2017-02-06 Thread Khang Nguyễn Nhật
Dear Daniel, I think it's bug. Because if i have a big file and with 15 minutes expires, I can't finished download it. 2017-02-06 15:12 GMT+07:00 Khang Nguyễn Nhật : > Dear Daniel, > I think it's bug. Because if i have a big file and with 15 minutes > expires, I can't finished download it. > > 2

Re: [ceph-users] ceph df : negative numbers

2017-02-06 Thread Wido den Hollander
> Op 6 februari 2017 om 11:10 schreef Florent B : > > > # ceph -v > ceph version 10.2.5 (c461ee19ecbc0c5c330aca20f7392c9a00730367) > > (officiel Ceph packages for Jessie) > > > Yes I recently adjusted pg_num, but all objects were correctly rebalanced. > > Then a manually deleted some objects

[ceph-users] Unsolved questions

2017-02-06 Thread Ranjan Ghosh
Hi everyone, I'm now running our two-node mini-cluster for some months. OSD, MDS and Monitor is running on both nodes. Additionally there is a very small third node which is only running a third monitor but no MDS/OSD. On both main servers, CephFS is mounted via FSTab/Kernel driver. The mounte

Re: [ceph-users] ceph df : negative numbers

2017-02-06 Thread Shinobu Kinjo
How about *pve01-rbd01*? * rados -p pve01-rbd01 ls | wc -l ? On Mon, Feb 6, 2017 at 9:40 PM, Florent B wrote: > On 02/06/2017 11:12 AM, Wido den Hollander wrote: >>> Op 6 februari 2017 om 11:10 schreef Florent B : >>> >>> >>> # ceph -v >>> ceph version 10.2.5 (c461ee19ecbc0c5c330aca20f7392c9a0

[ceph-users] "Numerical argument out of domain" error occurs during rbd export-diff | rbd import-diff

2017-02-06 Thread Bernhard J . M . Grün
Hello together, I got an error message during execution of the following command line: # rbd create backup/one-71-242-2 --object-size 32M --stripe-unit 4194304 --stripe-count 8 -s 1 # create an empty image # rbd export-diff --no-progress --whole-object one_ssd/one-71-242-2@backup-20170206-0829

Re: [ceph-users] ceph df : negative numbers

2017-02-06 Thread Vasu Kulkarni
worth filing an issue at http://tracker.ceph.com/ , looks like there are 2 different issues and should be easy to recreate. On Mon, Feb 6, 2017 at 9:01 AM, Florent B wrote: > On 02/06/2017 05:49 PM, Shinobu Kinjo wrote: > > How about *pve01-rbd01*? > > > > * rados -p pve01-rbd01 ls | wc -l > >

Re: [ceph-users] ceph df : negative numbers

2017-02-06 Thread Shinobu Kinjo
I've not been able to reproduce an issue with exactly same version of your cluster. ./ceph -v ceph version 10.2.5 (c461ee19ecbc0c5c330aca20f7392c9a00730367) ./rados df | grep cephfs cephfs_data_a 409600 10000 000 10

Re: [ceph-users] Fwd: lost power. monitors died. Cephx errors now

2017-02-06 Thread Sean Sullivan
So the cluster has been dead and down since around 8/10/2016. I have since rebooted the cluster in order to try and use the new ceph-monstore-tool rebuild functionality. I built the debian packages for the tools for hammer that were recently backported and installed it across all of the servers:

[ceph-users] virt-install into rbd hangs during Anaconda package installation

2017-02-06 Thread Tracy Reed
This is what I'm doing on my CentOS 7/KVM/virtlib server: rbd create --size 20G pool/vm.mydomain.com rbd map pool/vm.mydomain.com --name client.admin virt-install --name vm.mydomain.com --ram 2048 --disk path=/dev/rbd/pool/vm.mydomain.com --vcpus 1 --os-type linux --os-variant rhel6 --networ

Re: [ceph-users] Unsolved questions

2017-02-06 Thread Konstantin Shalygin
1) Every once in a while, some processes (PHP) accessing the filesystem get stuck in a D-state (Uninterruptable sleep). I wonder if this happens due to network fluctuations (both server are connected via a simple Gigabit crosslink cable) or how to diagnose this. Why exactly does this happen in the