Re: [ceph-users] centos6.4 + libvirt + qemu + rbd/ceph
On 12/05/2013 10:44 PM, Chris C wrote: I've been working on getting this setup working. I have virtual machines working using rbd based images by editing the domain directly. Is there any way to make the creation process better? We are hoping to be able to use a virsh pool using the rbd driver but it appears that Redhat has not compiled libvirt with rbd support. Thought? Recompile libvirt? Since RedHat hasn't enabled the RBD support in libvirt that's your problem. Might be that they'll do it in RHEL 7 where librbd is available natively? Thanks, /Chris C ___ ceph-users mailing list ceph-users@lists.ceph.com http://lists.ceph.com/listinfo.cgi/ceph-users-ceph.com -- Wido den Hollander 42on B.V. Phone: +31 (0)20 700 9902 Skype: contact42on ___ ceph-users mailing list ceph-users@lists.ceph.com http://lists.ceph.com/listinfo.cgi/ceph-users-ceph.com
[ceph-users] REST API issue for getting bucket policy
Hi all, I am working on the ceph radosgw(v0.72.1) and when I call the rest api to read the bucket policy, I got an internal server error(request URL is: /admin/bucket?policyformat=jsonbucket=test.). However, when I call this: /admin/bucket?policyformat=jsonbucket=testobject=obj, I got the policy of the object returned. Besides, I do have the right permission(buckets=*). Any idea? Thanks! Wei smime.p7s Description: S/MIME cryptographic signature ___ ceph-users mailing list ceph-users@lists.ceph.com http://lists.ceph.com/listinfo.cgi/ceph-users-ceph.com
Re: [ceph-users] centos6.4 + libvirt + qemu + rbd/ceph
See thread a couple days ago [ceph-users] qemu-kvm packages for centos On Thu, Dec 5, 2013 at 10:44 PM, Chris C mazzy...@gmail.com wrote: I've been working on getting this setup working. I have virtual machines working using rbd based images by editing the domain directly. Is there any way to make the creation process better? We are hoping to be able to use a virsh pool using the rbd driver but it appears that Redhat has not compiled libvirt with rbd support. Thought? Thanks, /Chris C ___ ceph-users mailing list ceph-users@lists.ceph.com http://lists.ceph.com/listinfo.cgi/ceph-users-ceph.com ___ ceph-users mailing list ceph-users@lists.ceph.com http://lists.ceph.com/listinfo.cgi/ceph-users-ceph.com
Re: [ceph-users] Journal, SSD and OS
Arf forgot to mention that I’ll do a software mdadm RAID 1 with both sda1 and sdb1 and put the OS on this. The rest (sda2 and sdb2) will go for the journals. @James: I think that Gandalf’s main idea was to save some costs/space on the servers so having dedicated disks is not an option. (that what I understand from your comment “have the OS somewhere else” but I could be wrong) Sébastien Han Cloud Engineer Always give 100%. Unless you're giving blood.” Phone: +33 (0)1 49 70 99 72 Mail: sebastien@enovance.com Address : 10, rue de la Victoire - 75009 Paris Web : www.enovance.com - Twitter : @enovance On 05 Dec 2013, at 16:02, James Pearce ja...@peacon.co.uk wrote: Another option is to run journals on individually presented SSDs, in a 5:1 ratio (spinning-disk:ssd) and have the OS somewhere else. Then the failure domain is smaller. Ideally implement some way to monitor SSD write life SMART data - at least it gives a guide as to device condition compared to its rated life. That can be done with smartmontools, but it would be nice to have it on the InkTank dashboard for example. On 2013-12-05 14:26, Sebastien Han wrote: Hi guys, I won’t do a RAID 1 with SSDs since they both write the same data. Thus, they are more likely to “almost” die at the same time. What I will try to do instead is to use both disk in JBOD mode or (degraded RAID0). Then I will create a tiny root partition for the OS. Then I’ll still have something like /dev/sda2 and /dev/sdb2 and then I can take advantage of the 2 disks independently. The good thing with that is that you can balance your journals across both SSDs. From a performance perspective this is really good. The bad thing as always is that if you loose a SSD you loose all the journals attached to it. Cheers. Sébastien Han Cloud Engineer Always give 100%. Unless you're giving blood.” Phone: +33 (0)1 49 70 99 72 Mail: sebastien@enovance.com Address : 10, rue de la Victoire - 75009 Paris Web : www.enovance.com - Twitter : @enovance On 05 Dec 2013, at 10:53, Gandalf Corvotempesta gandalf.corvotempe...@gmail.com wrote: 2013/12/4 Simon Leinen simon.lei...@switch.ch: I think this is a fine configuration - you won't be writing to the root partition too much, outside journals. We also put journals on the same SSDs as root partitions (not that we're very ambitious about performance...). Do you suggest a RAID1 for the OS partitions on SSDs ? Is this safe or a RAID1 will decrease SSD life? ___ ceph-users mailing list ceph-users@lists.ceph.com http://lists.ceph.com/listinfo.cgi/ceph-users-ceph.com ___ ceph-users mailing list ceph-users@lists.ceph.com http://lists.ceph.com/listinfo.cgi/ceph-users-ceph.com ___ ceph-users mailing list ceph-users@lists.ceph.com http://lists.ceph.com/listinfo.cgi/ceph-users-ceph.com signature.asc Description: Message signed with OpenPGP using GPGMail ___ ceph-users mailing list ceph-users@lists.ceph.com http://lists.ceph.com/listinfo.cgi/ceph-users-ceph.com
[ceph-users] how to set up disks in the same host
Hi all, What will be the fastest disks setup between those 2: - 1 OSD build from 6 disks in raid 10 and one ssd for journal - 3 OSDs, each with 2 disks in raid 1 and a common ssd for all journals (or more ssds if ssd performance will be an issue) Mainly, will 1 OSD raid 10 be faster or slower then independent OSDs? Best regards, Cristian Falcas ___ ceph-users mailing list ceph-users@lists.ceph.com http://lists.ceph.com/listinfo.cgi/ceph-users-ceph.com
Re: [ceph-users] Basic cephx configuration
Hi every one, I did not get any answer to my basic cephx question last week, so let me ask it one more time here, before I completely give up on Ceph and move on. So, my issue is: When all authentication settings are none: * The cluster works fine * The file /etc/ceph/ceph.client.admin.keyring exists Then I set auth_cluster_required to cephx. When I try to connect to the cluster, it detects client.admin and denies access with operation not supported, even for commands like ceph health. Finally, after I explicitly set the keyring parameter in the config (to the default value, because the keyring file was already in the default location), the cluster works fine again. So the behavior changes when I add those 2 default lines to the config: [client.admin] keyring = /etc/ceph/ceph.client.admin.keyring From the ceph.com documentation [1], about this keyring parameter: Description:The path to the keyring file. Type: String Required: No Default:/etc/ceph/$cluster.$name.keyring ... so, I need help: * maybe this is a real bug? (was it already reported ?) * maybe I am deeply stupid, and I don't understand what required and default means? (can anyone send me a good dictionary ?) * maybe obi-wan kenobi? Thanks to anyone who will respond anything (at that point, even a three-letter e-mail reading ACK would make me feel better). Best wishes for the future of Ceph, and best regards. Nicolas Canceill Scalable Storage Systems SURFsara (Amsterdam, NL) [1] http://ceph.com/docs/master/rados/configuration/auth-config-ref/#keys On 11/29/2013 03:09 PM, nicolasc wrote: An update on this issue: Explicitly setting the keyring parameter to its default value, in the client section, like this: [client.admin] keyring = /etc/ceph/ceph.client.admin.keyring solves the problem in the particular case when ONLY auth_cluster_required is set to cephx, and the two remaining auth parameters are set to none. The documentation clearly states that /etc/ceph/ceph.client.admin.keyring is the default value of the keyring setting [1], so this looks like a bug. Should I report it on the tracker? (BTW, all of this is on v0.72.1.) Also, does anyone have any idea about why this is not enough to enable the auth_service_required setting? That one still gives me the error: client.admin authentication error (95) Operation not supported Best regards, Nicolas Canceill Scalable Storage Systems SURFsara (Amsterdam, NL) [1] http://ceph.com/docs/master/rados/configuration/auth-config-ref/#keys On 11/29/2013 10:22 AM, nicolasc wrote: Hello every one, Just ran a fresh install of version Emperor on an empty cluster, and I am left clueless, trying to troubleshoot cephx. After ceph-deploy created the keys, I used ceph-authtool to generate the client.admin keyring and the monitor keyring, as indicated in the doc. The configuration is really out-of-the-box: 3 monitors, each with the keyring in /var/lib/ceph/mon/ceph-???/keyring, all keyrings have umask 644 and are owned by ceph. However, no matter which combination of auth_cluster_, auth_service_, or auth_client_required, is set to cephx; no matter either the keyring options like -k and --id on the command line. Authentication fails every time with: client.admin authentication error (95) Operation not supported Error connecting to cluster: Error A big thanks to any one who gives me a hint about what it means. (This message carries so little information, I feel it could be simply replaced by the ! character.) I have looked in every ceph and system log file, nothing more. Best regards, Nicolas Canceill Scalable Storage Systems SURFsara (Amsterdam, NL) ___ ceph-users mailing list ceph-users@lists.ceph.com http://lists.ceph.com/listinfo.cgi/ceph-users-ceph.com ___ ceph-users mailing list ceph-users@lists.ceph.com http://lists.ceph.com/listinfo.cgi/ceph-users-ceph.com
Re: [ceph-users] how to set up disks in the same host
On 12/06/2013 11:00 AM, Cristian Falcas wrote: Hi all, What will be the fastest disks setup between those 2: - 1 OSD build from 6 disks in raid 10 and one ssd for journal - 3 OSDs, each with 2 disks in raid 1 and a common ssd for all journals (or more ssds if ssd performance will be an issue) Mainly, will 1 OSD raid 10 be faster or slower then independent OSDs? Simply run 6 OSDs without any RAID and one SSD for the journaling. Danger is though, if you loose the journal, you loose all OSDs. So better place two SSDs and then 3 OSDs per journal and make sure that using crush objects go two OSDs on the different journals. You shouldn't use RAID underneath a OSD, let the replication handle all that. Best regards, Cristian Falcas ___ ceph-users mailing list ceph-users@lists.ceph.com http://lists.ceph.com/listinfo.cgi/ceph-users-ceph.com -- Wido den Hollander 42on B.V. Phone: +31 (0)20 700 9902 Skype: contact42on ___ ceph-users mailing list ceph-users@lists.ceph.com http://lists.ceph.com/listinfo.cgi/ceph-users-ceph.com
Re: [ceph-users] Journal, SSD and OS
2013/12/6 Sebastien Han sebastien@enovance.com: @James: I think that Gandalf’s main idea was to save some costs/space on the servers so having dedicated disks is not an option. (that what I understand from your comment “have the OS somewhere else” but I could be wrong) You are right. I don't have space for one or two disks to be used as OS. I was talking about RAID1 just for OS partition (stored on SSD) and not for the journal. Actually I'm testing OS stored on a USB pen drive with /tmp /var/run /var/lock placed on a tmpfs (ram) and log stored externally via rsyslog. This should avoid USB read/writes (that are slow) as much as possible. It's also very easy to restore to a new pen drive from a failure: dd if=/backup.img of=/dev/sdX bs=4M ___ ceph-users mailing list ceph-users@lists.ceph.com http://lists.ceph.com/listinfo.cgi/ceph-users-ceph.com
Re: [ceph-users] Journal, SSD and OS
Most servers also have internal SD card slots. There are SD cards advertising 90MB/s, though I haven't tried them as OS boot personally. On 2013-12-06 11:14, Gandalf Corvotempesta wrote: 2013/12/6 Sebastien Han sebastien@enovance.com: @James: I think that Gandalf’s main idea was to save some costs/space on the servers so having dedicated disks is not an option. (that what I understand from your comment “have the OS somewhere else” but I could be wrong) You are right. I don't have space for one or two disks to be used as OS. I was talking about RAID1 just for OS partition (stored on SSD) and not for the journal. Actually I'm testing OS stored on a USB pen drive with /tmp /var/run /var/lock placed on a tmpfs (ram) and log stored externally via rsyslog. This should avoid USB read/writes (that are slow) as much as possible. It's also very easy to restore to a new pen drive from a failure: dd if=/backup.img of=/dev/sdX bs=4M ___ ceph-users mailing list ceph-users@lists.ceph.com http://lists.ceph.com/listinfo.cgi/ceph-users-ceph.com
Re: [ceph-users] Openstack--instance-boot-from-ceph-volume:: error could not open disk image rbd
Le 05/12/2013 14:01, Karan Singh a écrit : Hello Everyone Trying to boot from ceph volume using bolg http://www.sebastien-han.fr/blog/2012/06/10/introducing-ceph-to-openstack/ and http://docs.openstack.org/user-guide/content/boot_from_volume.html Need help for this error. = *Logs from /var/log/libvirt/qemu ::* ** = qemu-kvm: -drive file=rbd:ceph-volumes/volume-dd315dda-b22a-4cf8-8b77-7c2b2f163155:id=volumes:key=AQC804xS8HzFJxAAD/zzQ8LMzq9wDLq/5a472g==:auth_supported=cephx\;none:mon_host=192.168.1.31\:6789\;192.168.1.33\:6789\;192.168.1.38\:6789,if=none,id=drive-virtio-disk0,format=raw,serial=dd315dda-b22a-4cf8-8b77-7c2b2f163155,cache=none: *could not open disk image* rbd:ceph-volumes/volume-dd315dda-b22a-4cf8-8b77-7c2b2f163155:id=volumes:key=AQC804xS8HzFJxAAD/zzQ8LMzq9wDLq/5a472g==:auth_supported=cephx\;none:mon_host=192.168.1.31\:6789\;192.168.1.33\:6789\;192.168.1.38\:6789: No such file or directory 2013-12-05 12:42:29.544+: shutting down Hello, Does your qemu supports RBD images ? See in the last line of, supported formats : qemu-img -h I had the same thing recently, using Debian wheezy, where qemu does not supports RBD. [...] ___ ceph-users mailing list ceph-users@lists.ceph.com http://lists.ceph.com/listinfo.cgi/ceph-users-ceph.com
Re: [ceph-users] Journal, SSD and OS
Most servers also have internal SD card slots. There are SD cards advertising 90MB/s, though I haven't tried them as OS boot personally. We did this with some servers 2 1/2 years ago with some Blade hardware: did not work out so well. High level of failures on the SD card's even with all stuff properly on tempfs. For Ceph we have the OS on one of the two SSDs where the journals live. We are graphing and alerting on the SSD wear-level SMART info. Up to now it has worked out quite well. (We have a very similar setup for Openstack Swift for a few years now where we use the SSD for flashcache instead of journals) Cheers, Robert van Leeuwen ___ ceph-users mailing list ceph-users@lists.ceph.com http://lists.ceph.com/listinfo.cgi/ceph-users-ceph.com
Re: [ceph-users] Mounting Ceph on Linux/Windows
Out of curiosity I tried the 'ceph' command from windows too. I had to rename librados.dll to librados.so.2, install a readline replacement (https://pypi.python.org/pypi/pyreadline/2.0), and even then it completely ignored anything I put on the command line, but from the ceph shell I could do thinks like 'health', 'status', 'osd tree', 'osd lspools', 'auth list', etc. I thought that was pretty neat. James ___ ceph-users mailing list ceph-users@lists.ceph.com http://lists.ceph.com/listinfo.cgi/ceph-users-ceph.com
[ceph-users] recreate bucket error
Hi, In version dumpling upgraded from bobtail working create the same bucket. root@vm-1:/etc/apache2/sites-enabled# s3 -u create testcreate Bucket successfully created. root@vm-1:/etc/apache2/sites-enabled# s3 -u create testcreate Bucket successfully created. I installed new dumpling cluster and: root@s1:/var/log/radosgw# s3 -u create test1 Bucket successfully created. root@s1:/var/log/radosgw# s3 -u create test1 ERROR: ErrorUnknown In radosgw logs: 2013-12-06 13:59:56.083109 7f162d7c2700 1 == starting new request req=0xb7d480 = 2013-12-06 13:59:56.083227 7f162d7c2700 2 req 5:0.000119::PUT /test1/::initializing 2013-12-06 13:59:56.083261 7f162d7c2700 10 meta HTTP_X_AMZ_DATE 2013-12-06 13:59:56.083274 7f162d7c2700 10 x x-amz-date:Fri, 06 Dec 2013 12:59:56 GMT 2013-12-06 13:59:56.083298 7f162d7c2700 10 s-object=NULL s-bucket=test1 2013-12-06 13:59:56.083307 7f162d7c2700 2 req 5:0.000199:s3:PUT /test1/::getting op 2013-12-06 13:59:56.083315 7f162d7c2700 2 req 5:0.000207:s3:PUT /test1/:create_bucket:authorizing 2013-12-06 13:59:56.091724 7f162d7c2700 10 get_canon_resource(): dest= 2013-12-06 13:59:56.091742 7f162d7c2700 10 auth_hdr: PUT x-amz-date:Fri, 06 Dec 2013 12:59:56 GMT /test1/ 2013-12-06 13:59:56.091836 7f162d7c2700 2 req 5:0.008728:s3:PUT /test1/:create_bucket:reading permissions 2013-12-06 13:59:56.091848 7f162d7c2700 2 req 5:0.008740:s3:PUT /test1/:create_bucket:verifying op mask 2013-12-06 13:59:56.091852 7f162d7c2700 2 req 5:0.008744:s3:PUT /test1/:create_bucket:verifying op permissions 2013-12-06 13:59:56.093858 7f162d7c2700 2 req 5:0.010750:s3:PUT /test1/:create_bucket:verifying op params 2013-12-06 13:59:56.093882 7f162d7c2700 2 req 5:0.010773:s3:PUT /test1/:create_bucket:executing 2013-12-06 13:59:56.104819 7f162d7c2700 0 WARNING: couldn't find acl header for object, generating default 2013-12-06 13:59:56.132625 7f162d7c2700 0 get_bucket_info returned -125 2013-12-06 13:59:56.132656 7f162d7c2700 0 WARNING: set_req_state_err err_no=125 resorting to 500 2013-12-06 13:59:56.132693 7f162d7c2700 2 req 5:0.049584:s3:PUT /test1/:create_bucket:http status=500 2013-12-06 13:59:56.132890 7f162d7c2700 1 == req done req=0xb7d480 http_status=500 == -- Regards Dominik ___ ceph-users mailing list ceph-users@lists.ceph.com http://lists.ceph.com/listinfo.cgi/ceph-users-ceph.com
Re: [ceph-users] centos6.4 + libvirt + qemu + rbd/ceph
Wido, we were thinking along that line as well. We're trying to figure out which path will cause the least amount of pain ;) /C On Fri, Dec 6, 2013 at 3:27 AM, Wido den Hollander w...@42on.com wrote: On 12/05/2013 10:44 PM, Chris C wrote: I've been working on getting this setup working. I have virtual machines working using rbd based images by editing the domain directly. Is there any way to make the creation process better? We are hoping to be able to use a virsh pool using the rbd driver but it appears that Redhat has not compiled libvirt with rbd support. Thought? Recompile libvirt? Since RedHat hasn't enabled the RBD support in libvirt that's your problem. Might be that they'll do it in RHEL 7 where librbd is available natively? Thanks, /Chris C ___ ceph-users mailing list ceph-users@lists.ceph.com http://lists.ceph.com/listinfo.cgi/ceph-users-ceph.com -- Wido den Hollander 42on B.V. Phone: +31 (0)20 700 9902 Skype: contact42on ___ ceph-users mailing list ceph-users@lists.ceph.com http://lists.ceph.com/listinfo.cgi/ceph-users-ceph.com ___ ceph-users mailing list ceph-users@lists.ceph.com http://lists.ceph.com/listinfo.cgi/ceph-users-ceph.com
Re: [ceph-users] installation help
On Fri, Dec 6, 2013 at 5:13 AM, Wojciech Giel wojciech.g...@cimr.cam.ac.uk wrote: Hello, I trying to install ceph but can't get it working documentation is not clear and confusing how to do it. I have cloned 3 machines with ubuntu 12.04 minimal system. I'm trying to follow docs http://ceph.com/docs/master/start/quick-start-preflight/ but got some questions: step 4. Configure your ceph-deploy admin node with password-less SSH access This should be done on ceph account, its't it? If you are following the quickstart to the letter, yes. If you are using the latest version of ceph-deploy (1.3.3) you will benefit from automatically getting these set up for you when you do `ceph-deploy new {hosts}` It will make sure it can SSH without a password prompt to the hosts you are setting up. next: http://ceph.com/docs/master/start/quick-ceph-deploy/ creating directories for maintaining the configuration for ceph-deploy is on ceph account or root? ceph-deploy will create the ceph.conf and other files from wherever it runs from owned by the user executing ceph-deploy. So it depends here on what user you are calling ceph-deploy. do all following step in docs are on ceph account or root? All steps in the quickstart assume that you have created a ceph user and you are connecting to remote hosts with a ceph user. In the end it doesn't matter. But if you want to get things right from the get-go I would try and match what the quickstart uses so you can troubleshoot easier. if on ceph account step 3. creating mon on 3 machines gives on two remote machines these errors: [ceph1][DEBUG ] locating the `service` executable... [ceph1][INFO ] Running command: sudo initctl emit ceph-mon cluster=ceph id=ceph1 [ceph1][INFO ] Running command: sudo ceph --cluster=ceph --admin-daemon /var/run/ceph/ceph-mon.ceph1.asok mon_status [ceph1][ERROR ] admin_socket: exception getting command descriptions: [Errno 2] No such file or directory [ceph1][WARNIN] monitor: mon.ceph1, might not be running yet [ceph1][INFO ] Running command: sudo ceph --cluster=ceph --admin-daemon /var/run/ceph/ceph-mon.ceph1.asok mon_status [ceph1][ERROR ] admin_socket: exception getting command descriptions: [Errno 2] No such file or directory [ceph1][WARNIN] ceph1 is not defined in `mon initial members` [ceph1][WARNIN] monitor ceph1 does not exist in monmap [ceph1][WARNIN] neither `public_addr` nor `public_network` keys are defined for monitors [ceph1][WARNIN] monitors may not be able to form quorum It looks like you've tried a few things in that server and you've ended in a broken state. If you are deploying the `ceph1` mon, that should've been defined in your ceph.conf and it should've been done automatically for you when you called `ceph-deploy new ceph1`. This is an example of creating a new config file for a server I have called `node1`: $ ceph-deploy new node1 [ceph_deploy.cli][INFO ] Invoked (1.3.3): /Users/alfredo/.virtualenvs/ceph-deploy/bin/ceph-deploy new node1 [ceph_deploy.new][DEBUG ] Creating new cluster named ceph [ceph_deploy.new][DEBUG ] Resolving host node1 [ceph_deploy.new][DEBUG ] Monitor node1 at 192.168.111.100 [ceph_deploy.new][INFO ] making sure passwordless SSH succeeds [node1][DEBUG ] connected to host: papaya.local [node1][INFO ] Running command: ssh -CT -o BatchMode=yes node1 [ceph_deploy.new][DEBUG ] Monitor initial members are ['node1'] [ceph_deploy.new][DEBUG ] Monitor addrs are ['192.168.111.100'] [ceph_deploy.new][DEBUG ] Creating a random mon key... [ceph_deploy.new][DEBUG ] Writing initial config to ceph.conf... [ceph_deploy.new][DEBUG ] Writing monitor keyring to ceph.mon.keyring... $ cat ceph.conf [global] fsid = 4e04aeaf-7025-4d33-bbcb-b27e75749b97 mon_initial_members = node1 mon_host = 192.168.111.100 auth_supported = cephx osd_journal_size = 1024 filestore_xattr_use_omap = true See how `mon_initial_members` has `node1` in it? step 4. gathering keys should it be from all mon servers? checking status on ceph account gives: $ ceph health 2013-12-06 09:48:41.550270 7f16b6eea700 -1 monclient(hunting): ERROR: missing keyring, cannot use cephx for authentication 2013-12-06 09:48:41.550278 7f16b6eea700 0 librados: client.admin initialization error (2) No such file or directory Error connecting to cluster: ObjectNotFound That happens because you need to call `ceph` with sudo (something that ceph-deploy takes care for you) on root account: # ceph status cluster 5ee9b196-ef36-46dd-870e-6ef1824b1cd0 health HEALTH_ERR 192 pgs stuck inactive; 192 pgs stuck unclean; no osds monmap e1: 1 mons at {ceph0=192.168.45.222:6789/0}, election epoch 2, quorum 0 ceph0 osdmap e1: 0 osds: 0 up, 0 in pgmap v2: 192 pgs, 3 pools, 0 bytes data, 0 objects 0 kB used, 0 kB / 0 kB avail 192 creating Ceph management after installation should it be done on root or ceph account? It doesn't matter, just as long as you have
Re: [ceph-users] installation help
On Fri, Dec 6, 2013 at 5:13 AM, Wojciech Giel wojciech.g...@cimr.cam.ac.uk wrote: Hello, I trying to install ceph but can't get it working documentation is not clear and confusing how to do it. I have cloned 3 machines with ubuntu 12.04 minimal system. I'm trying to follow docs http://ceph.com/docs/master/start/quick-start-preflight/ but got some questions: step 4. Configure your ceph-deploy admin node with password-less SSH access This should be done on ceph account, its't it? next: http://ceph.com/docs/master/start/quick-ceph-deploy/ creating directories for maintaining the configuration for ceph-deploy is on ceph account or root? do all following step in docs are on ceph account or root? if on ceph account step 3. creating mon on 3 machines gives on two remote machines these errors: [ceph1][DEBUG ] locating the `service` executable... [ceph1][INFO ] Running command: sudo initctl emit ceph-mon cluster=ceph id=ceph1 [ceph1][INFO ] Running command: sudo ceph --cluster=ceph --admin-daemon /var/run/ceph/ceph-mon.ceph1.asok mon_status [ceph1][ERROR ] admin_socket: exception getting command descriptions: [Errno 2] No such file or directory [ceph1][WARNIN] monitor: mon.ceph1, might not be running yet [ceph1][INFO ] Running command: sudo ceph --cluster=ceph --admin-daemon /var/run/ceph/ceph-mon.ceph1.asok mon_status [ceph1][ERROR ] admin_socket: exception getting command descriptions: [Errno 2] No such file or directory [ceph1][WARNIN] ceph1 is not defined in `mon initial members` [ceph1][WARNIN] monitor ceph1 does not exist in monmap [ceph1][WARNIN] neither `public_addr` nor `public_network` keys are defined for monitors [ceph1][WARNIN] monitors may not be able to form quorum step 4. gathering keys should it be from all mon servers? checking status on ceph account gives: $ ceph health 2013-12-06 09:48:41.550270 7f16b6eea700 -1 monclient(hunting): ERROR: missing keyring, cannot use cephx for authentication 2013-12-06 09:48:41.550278 7f16b6eea700 0 librados: client.admin initialization error (2) No such file or directory Error connecting to cluster: ObjectNotFound on root account: # ceph status cluster 5ee9b196-ef36-46dd-870e-6ef1824b1cd0 health HEALTH_ERR 192 pgs stuck inactive; 192 pgs stuck unclean; no osds monmap e1: 1 mons at {ceph0=192.168.45.222:6789/0}, election epoch 2, quorum 0 ceph0 osdmap e1: 0 osds: 0 up, 0 in pgmap v2: 192 pgs, 3 pools, 0 bytes data, 0 objects 0 kB used, 0 kB / 0 kB avail 192 creating Ceph management after installation should it be done on root or ceph account? I've attached typescrpi from what I have done. I just went through your typescript and you did call `new` but not for all the hosts you are using: ceph-deploy new ceph0 And then you did: ceph-deploy install ceph0 ceph1 ceph2 And finally (reason why you have issues) you tried to deploy a mon to those hosts that were not defined in your ceph.conf: ceph-deploy mon create ceph0 ceph1 ceph2 You see, you need to define those as well when calling `new` like: ceph-deploy new ceph0 ceph1 ceph2 thanks Wojciech ___ ceph-users mailing list ceph-users@lists.ceph.com http://lists.ceph.com/listinfo.cgi/ceph-users-ceph.com ___ ceph-users mailing list ceph-users@lists.ceph.com http://lists.ceph.com/listinfo.cgi/ceph-users-ceph.com
Re: [ceph-users] optimal setup with 4 x ethernet ports
looking at tcpdump all the traffic is going exactly where it is supposed to go, in particular an osd on the 192.168.228.x network appears to talk to an osd on the 192.168.229.x network without anything strange happening. I was just wondering if there was anything about ceph that could make this non-optimal, assuming traffic was reasonably balanced between all the osd's (eg all the same weights). I think the only time it would suffer is if writes to other osds result in a replica write to a single osd, and even then a single OSD is still limited to 7200RPM disk speed anyway so the loss isn't going to be that great. Should be fine given you only have a 1:1 ratio of link to disk. I think I'll be moving over to bonded setup anyway, although I'm not sure if rr or lacp is best... rr will give the best potential throughput, but lacp should give similar aggregate throughput if there are plenty of connections going on, and less cpu load as no need to reassemble fragments. One of the DreamHost clusters is using a pair of bonded 1GbE links on the public network and another pair for the cluster network, we configured each to use mode 802.3ad. -- Kyle ___ ceph-users mailing list ceph-users@lists.ceph.com http://lists.ceph.com/listinfo.cgi/ceph-users-ceph.com
Re: [ceph-users] centos6.4 + libvirt + qemu + rbd/ceph
Dan, I found the thread but it looks like another dead end :( /Chris C On Fri, Dec 6, 2013 at 4:46 AM, Dan van der Ster d...@vanderster.com wrote: See thread a couple days ago [ceph-users] qemu-kvm packages for centos On Thu, Dec 5, 2013 at 10:44 PM, Chris C mazzy...@gmail.com wrote: I've been working on getting this setup working. I have virtual machines working using rbd based images by editing the domain directly. Is there any way to make the creation process better? We are hoping to be able to use a virsh pool using the rbd driver but it appears that Redhat has not compiled libvirt with rbd support. Thought? Thanks, /Chris C ___ ceph-users mailing list ceph-users@lists.ceph.com http://lists.ceph.com/listinfo.cgi/ceph-users-ceph.com ___ ceph-users mailing list ceph-users@lists.ceph.com http://lists.ceph.com/listinfo.cgi/ceph-users-ceph.com
Re: [ceph-users] centos6.4 + libvirt + qemu + rbd/ceph
I think the version of Libvirt included with RHEL/CentOS supports RBD storage (but not pools), so outside of compiling a newer version not sure there can be anything else done aside from waiting for repo additions/newer versions of the distro. Not sure what your scenario is, but this is the exact reason we switched our underlying virtualization infrastructure to Ubuntu. Their cloud archive PPA has updated packages for QEMU/KVM, Libvirt, Open vSwitch, etc. that are backported for LTS releases, and is something I personally think RHEL is WAY behind the curve on (getting better with their RDO initiative though). We didn't like consuming resources validating that updated builds of QEMU/Libvirt were going to cause problems and just allocated those resources to learning the Ubuntu environment. As far as streamlining management on top of that, you have some options (outside of virt-manager, which has no native support for RBD IIRC) like Proxmox (which is an entire solution like ESXi/Hyper-V using KVM) or something like OpenStack or OpenNebula (we use OpenNebula). Beats having to edit domains by hand. ;-) - Original Message - From: Chris C mazzy...@gmail.com To: Dan van der Ster d...@vanderster.com Cc: ceph-users@lists.ceph.com Sent: Friday, December 6, 2013 10:37:03 AM Subject: Re: [ceph-users] centos6.4 + libvirt + qemu + rbd/ceph Dan, I found the thread but it looks like another dead end :( /Chris C On Fri, Dec 6, 2013 at 4:46 AM, Dan van der Ster d...@vanderster.com wrote: See thread a couple days ago [ceph-users] qemu-kvm packages for centos On Thu, Dec 5, 2013 at 10:44 PM, Chris C mazzy...@gmail.com wrote: I've been working on getting this setup working. I have virtual machines working using rbd based images by editing the domain directly. Is there any way to make the creation process better? We are hoping to be able to use a virsh pool using the rbd driver but it appears that Redhat has not compiled libvirt with rbd support. Thought? Thanks, /Chris C ___ ceph-users mailing list ceph-users@lists.ceph.com http://lists.ceph.com/listinfo.cgi/ceph-users-ceph.com ___ ceph-users mailing list ceph-users@lists.ceph.com http://lists.ceph.com/listinfo.cgi/ceph-users-ceph.com NOTICE: Protect the information in this message in accordance with the company's security policies. If you received this message in error, immediately notify the sender and destroy all copies.___ ceph-users mailing list ceph-users@lists.ceph.com http://lists.ceph.com/listinfo.cgi/ceph-users-ceph.com
Re: [ceph-users] Mounting Ceph on Linux/Windows
[Moving this thread to ceph-devel] ___ ceph-users mailing list ceph-users@lists.ceph.com http://lists.ceph.com/listinfo.cgi/ceph-users-ceph.com
Re: [ceph-users] Impact of fancy striping
Hi James, Thank you for this clarification. I am quite aware of that, which is why the journals are on SAS disks in RAID0 (SSDs out of scope). I still have trouble believing that fast-but-not-super-fast journals is the main reason for the poor performances observed. Maybe I am mistaken? Best regards, Nicolas Canceill Scalable Storage Systems SURFsara (Amsterdam, NL) On 12/03/2013 03:01 PM, James Pearce wrote: I would really appreciate it if someone could: - explain why the journal setup is way more important than striping settings; I'm not sure if it's what you're asking, but any write must be physically written to the journal before the operation is acknowledged. So the overall cluster performance (or rather write latency) is always governed by the speed of those journals. Data is then gathered up into (hopefully) larger blocks and committed to OSDs later. ___ ceph-users mailing list ceph-users@lists.ceph.com http://lists.ceph.com/listinfo.cgi/ceph-users-ceph.com ___ ceph-users mailing list ceph-users@lists.ceph.com http://lists.ceph.com/listinfo.cgi/ceph-users-ceph.com
Re: [ceph-users] Basic cephx configuration
Hi Dan, Thank you for the advice and indications. We have the exact same configuration, except I am only enabling auth cluster, and I am using ceph.client.admin.keyring instead of simply keyring. Both locations /etc/ceph/ceph.client.admin.keyring and /etc/ceph/keyring are presented as default values for the keyring configuration setting. I will try /etc/ceph/keyring, but I doubt this changes much. I am curious of whether your setup still works if you remove the keyring = /etc/ceph/keyring setting (you are also using a default location, so you could remove that line safely, right?). Thank you very much for answering. Best regards, Nicolas Canceill Scalable Storage Systems SURFsara (Amsterdam, NL) On 12/06/2013 11:42 AM, Dan Van Der Ster wrote: Hi, All of our clusters have this in ceph.conf: [global] auth cluster required = cephx auth service required = cephx auth client required = cephx keyring = /etc/ceph/keyring and the client.admin secret in /etc/ceph/keyring: # cat /etc/ceph/keyring [client.admin] key = ... With that you should be able to do ceph health without passing —id or —keyring args. (this is with dumpling, not emperor, but I guess it didn’t change.) If it still doesn’t work, check the capabilities that client.admin has (with ceph auth list). Should be caps: [mds] allow caps: [mon] allow * caps: [osd] allow * Cheers, Dan On 06 Dec 2013, at 11:06, nicolasc nicolas.cance...@surfsara.nl wrote: Hi every one, I did not get any answer to my basic cephx question last week, so let me ask it one more time here, before I completely give up on Ceph and move on. So, my issue is: When all authentication settings are none: * The cluster works fine * The file /etc/ceph/ceph.client.admin.keyring exists Then I set auth_cluster_required to cephx. When I try to connect to the cluster, it detects client.admin and denies access with operation not supported, even for commands like ceph health. Finally, after I explicitly set the keyring parameter in the config (to the default value, because the keyring file was already in the default location), the cluster works fine again. So the behavior changes when I add those 2 default lines to the config: [client.admin] keyring = /etc/ceph/ceph.client.admin.keyring From the ceph.com documentation [1], about this keyring parameter: Description:The path to the keyring file. Type: String Required: No Default:/etc/ceph/$cluster.$name.keyring ... so, I need help: * maybe this is a real bug? (was it already reported ?) * maybe I am deeply stupid, and I don't understand what required and default means? (can anyone send me a good dictionary ?) * maybe obi-wan kenobi? Thanks to anyone who will respond anything (at that point, even a three-letter e-mail reading ACK would make me feel better). Best wishes for the future of Ceph, and best regards. Nicolas Canceill Scalable Storage Systems SURFsara (Amsterdam, NL) [1] http://ceph.com/docs/master/rados/configuration/auth-config-ref/#keys On 11/29/2013 03:09 PM, nicolasc wrote: An update on this issue: Explicitly setting the keyring parameter to its default value, in the client section, like this: [client.admin] keyring = /etc/ceph/ceph.client.admin.keyring solves the problem in the particular case when ONLY auth_cluster_required is set to cephx, and the two remaining auth parameters are set to none. The documentation clearly states that /etc/ceph/ceph.client.admin.keyring is the default value of the keyring setting [1], so this looks like a bug. Should I report it on the tracker? (BTW, all of this is on v0.72.1.) Also, does anyone have any idea about why this is not enough to enable the auth_service_required setting? That one still gives me the error: client.admin authentication error (95) Operation not supported Best regards, Nicolas Canceill Scalable Storage Systems SURFsara (Amsterdam, NL) [1] http://ceph.com/docs/master/rados/configuration/auth-config-ref/#keys On 11/29/2013 10:22 AM, nicolasc wrote: Hello every one, Just ran a fresh install of version Emperor on an empty cluster, and I am left clueless, trying to troubleshoot cephx. After ceph-deploy created the keys, I used ceph-authtool to generate the client.admin keyring and the monitor keyring, as indicated in the doc. The configuration is really out-of-the-box: 3 monitors, each with the keyring in /var/lib/ceph/mon/ceph-???/keyring, all keyrings have umask 644 and are owned by ceph. However, no matter which combination of auth_cluster_, auth_service_, or auth_client_required, is set to cephx; no matter either the keyring options like -k and --id on the command line. Authentication fails every time with: client.admin authentication error (95) Operation not supported Error connecting to cluster: Error A big thanks to any one who gives me a hint about what it means. (This message carries so little information, I feel it could be
Re: [ceph-users] Impact of fancy striping
Hopefully a Ceph developer will be able to clarify how small writes are journaled? The write-through 'bug' seems to explain small-block performance I've measured in various configurations (I find similar results to you). I've not still tested the patch cited, but it would be *very* interesting to know the impact on clusters running with spinning-disk journals in particular. On 2013-12-06 16:05, nicolasc wrote: Hi James, Thank you for this clarification. I am quite aware of that, which is why the journals are on SAS disks in RAID0 (SSDs out of scope). I still have trouble believing that fast-but-not-super-fast journals is the main reason for the poor performances observed. Maybe I am mistaken? Best regards, Nicolas Canceill Scalable Storage Systems SURFsara (Amsterdam, NL) On 12/03/2013 03:01 PM, James Pearce wrote: I would really appreciate it if someone could: - explain why the journal setup is way more important than striping settings; I'm not sure if it's what you're asking, but any write must be physically written to the journal before the operation is acknowledged. So the overall cluster performance (or rather write latency) is always governed by the speed of those journals. Data is then gathered up into (hopefully) larger blocks and committed to OSDs later. ___ ceph-users mailing list ceph-users@lists.ceph.com http://lists.ceph.com/listinfo.cgi/ceph-users-ceph.com ___ ceph-users mailing list ceph-users@lists.ceph.com http://lists.ceph.com/listinfo.cgi/ceph-users-ceph.com
Re: [ceph-users] centos6.4 + libvirt + qemu + rbd/ceph
Will throw my US $0.02 in here... We’re running CentOS 6.4[1] + modern Ceph, VMs are managed by CloudStack. We use distro packages whenever possible - most times folks suggest building something from source, I have to be dragged kicking and screaming to agreement. Ceph support is one of the very few exceptions. We build/package/sign modern kernels (3.6.11), modern libvirt (0.9.13), modern qemu (1.2.0) and probably a few other bits. The improved functionality is easily worth the effort. John 1:Due to RH Cloud agreements we’ll have to run RHEL hypervisors soon, will be happy to share thoughts on getting Ceph working with that once we do it On Dec 6, 2013, at 5:25 AM, Chris C mazzy...@gmail.commailto:mazzy...@gmail.com wrote: Wido, we were thinking along that line as well. We're trying to figure out which path will cause the least amount of pain ;) /C On Fri, Dec 6, 2013 at 3:27 AM, Wido den Hollander w...@42on.commailto:w...@42on.com wrote: On 12/05/2013 10:44 PM, Chris C wrote: I've been working on getting this setup working. I have virtual machines working using rbd based images by editing the domain directly. Is there any way to make the creation process better? We are hoping to be able to use a virsh pool using the rbd driver but it appears that Redhat has not compiled libvirt with rbd support. Thought? Recompile libvirt? Since RedHat hasn't enabled the RBD support in libvirt that's your problem. Might be that they'll do it in RHEL 7 where librbd is available natively? Thanks, /Chris C ___ ceph-users mailing list ceph-users@lists.ceph.commailto:ceph-users@lists.ceph.com http://lists.ceph.com/listinfo.cgi/ceph-users-ceph.com -- Wido den Hollander 42on B.V. Phone: +31 (0)20 700 9902tel:%2B31%20%280%2920%20700%209902 Skype: contact42on ___ ceph-users mailing list ceph-users@lists.ceph.commailto:ceph-users@lists.ceph.com http://lists.ceph.com/listinfo.cgi/ceph-users-ceph.com ___ ceph-users mailing list ceph-users@lists.ceph.commailto:ceph-users@lists.ceph.com http://lists.ceph.com/listinfo.cgi/ceph-users-ceph.com Stratosechttp://stratosec.co/ - Compliance as a Service o: 415.315.9385 @johnlkinsellahttp://twitter.com/johnlkinsella ___ ceph-users mailing list ceph-users@lists.ceph.com http://lists.ceph.com/listinfo.cgi/ceph-users-ceph.com
Re: [ceph-users] recreate bucket error
I'm having trouble reproducing this one. Are you running on latest dumpling? Does it happen with any newly created bucket, or just with buckets that existed before? Yehuda On Fri, Dec 6, 2013 at 5:07 AM, Dominik Mostowiec dominikmostow...@gmail.com wrote: Hi, In version dumpling upgraded from bobtail working create the same bucket. root@vm-1:/etc/apache2/sites-enabled# s3 -u create testcreate Bucket successfully created. root@vm-1:/etc/apache2/sites-enabled# s3 -u create testcreate Bucket successfully created. I installed new dumpling cluster and: root@s1:/var/log/radosgw# s3 -u create test1 Bucket successfully created. root@s1:/var/log/radosgw# s3 -u create test1 ERROR: ErrorUnknown In radosgw logs: 2013-12-06 13:59:56.083109 7f162d7c2700 1 == starting new request req=0xb7d480 = 2013-12-06 13:59:56.083227 7f162d7c2700 2 req 5:0.000119::PUT /test1/::initializing 2013-12-06 13:59:56.083261 7f162d7c2700 10 meta HTTP_X_AMZ_DATE 2013-12-06 13:59:56.083274 7f162d7c2700 10 x x-amz-date:Fri, 06 Dec 2013 12:59:56 GMT 2013-12-06 13:59:56.083298 7f162d7c2700 10 s-object=NULL s-bucket=test1 2013-12-06 13:59:56.083307 7f162d7c2700 2 req 5:0.000199:s3:PUT /test1/::getting op 2013-12-06 13:59:56.083315 7f162d7c2700 2 req 5:0.000207:s3:PUT /test1/:create_bucket:authorizing 2013-12-06 13:59:56.091724 7f162d7c2700 10 get_canon_resource(): dest= 2013-12-06 13:59:56.091742 7f162d7c2700 10 auth_hdr: PUT x-amz-date:Fri, 06 Dec 2013 12:59:56 GMT /test1/ 2013-12-06 13:59:56.091836 7f162d7c2700 2 req 5:0.008728:s3:PUT /test1/:create_bucket:reading permissions 2013-12-06 13:59:56.091848 7f162d7c2700 2 req 5:0.008740:s3:PUT /test1/:create_bucket:verifying op mask 2013-12-06 13:59:56.091852 7f162d7c2700 2 req 5:0.008744:s3:PUT /test1/:create_bucket:verifying op permissions 2013-12-06 13:59:56.093858 7f162d7c2700 2 req 5:0.010750:s3:PUT /test1/:create_bucket:verifying op params 2013-12-06 13:59:56.093882 7f162d7c2700 2 req 5:0.010773:s3:PUT /test1/:create_bucket:executing 2013-12-06 13:59:56.104819 7f162d7c2700 0 WARNING: couldn't find acl header for object, generating default 2013-12-06 13:59:56.132625 7f162d7c2700 0 get_bucket_info returned -125 2013-12-06 13:59:56.132656 7f162d7c2700 0 WARNING: set_req_state_err err_no=125 resorting to 500 2013-12-06 13:59:56.132693 7f162d7c2700 2 req 5:0.049584:s3:PUT /test1/:create_bucket:http status=500 2013-12-06 13:59:56.132890 7f162d7c2700 1 == req done req=0xb7d480 http_status=500 == -- Regards Dominik ___ ceph-users mailing list ceph-users@lists.ceph.com http://lists.ceph.com/listinfo.cgi/ceph-users-ceph.com ___ ceph-users mailing list ceph-users@lists.ceph.com http://lists.ceph.com/listinfo.cgi/ceph-users-ceph.com
Re: [ceph-users] Impact of fancy striping
If I understand correctly you have one sas disk as a journal for multiple OSDs. If you do small synchronous writes it will become a IO bottleneck pretty quickly: Due to multiple journals on the same disk it will no longer be sequential writes writes to one journal but 4k writes to x journals making it fully random. I would expect a performance of 100 to 200 IOPS max. Doing an iostat -x or atop should show this bottleneck immediately. This is also the reason to go with SSDs: they have reasonable random IO performance. Cheers, Robert van Leeuwen Sent from my iPad On 6 dec. 2013, at 17:05, nicolasc nicolas.cance...@surfsara.nl wrote: Hi James, Thank you for this clarification. I am quite aware of that, which is why the journals are on SAS disks in RAID0 (SSDs out of scope). I still have trouble believing that fast-but-not-super-fast journals is the main reason for the poor performances observed. Maybe I am mistaken? Best regards, Nicolas Canceill Scalable Storage Systems SURFsara (Amsterdam, NL) On 12/03/2013 03:01 PM, James Pearce wrote: I would really appreciate it if someone could: - explain why the journal setup is way more important than striping settings; I'm not sure if it's what you're asking, but any write must be physically written to the journal before the operation is acknowledged. So the overall cluster performance (or rather write latency) is always governed by the speed of those journals. Data is then gathered up into (hopefully) larger blocks and committed to OSDs later. ___ ceph-users mailing list ceph-users@lists.ceph.com http://lists.ceph.com/listinfo.cgi/ceph-users-ceph.com ___ ceph-users mailing list ceph-users@lists.ceph.com http://lists.ceph.com/listinfo.cgi/ceph-users-ceph.com ___ ceph-users mailing list ceph-users@lists.ceph.com http://lists.ceph.com/listinfo.cgi/ceph-users-ceph.com
Re: [ceph-users] Basic cephx configuration
Ncolasc, You said: Just ran a fresh install of version Emperor on an empty cluster, and I am left clueless, trying to troubleshoot cephx. *After ceph-deploy created the keys, I used ceph-authtool to generate the client.admin keyring and the monitor keyring, as indicated in the doc.* The configuration is really out-of-the-box: 3 monitors, each with the keyring in /var/lib/ceph/mon/ceph-???/keyring, all keyrings have umask 644 and are owned by ceph. The ceph-deploy utility already generates the monitor keyring and the ceph.client.admin.keyring for you. I have a wip-doc branch for a manual deployment procedure without ceph-deploy, which I don't recommend for a first time user. However, it does detail what is going on. http://ceph.com/docs/wip-doc-build-cluster/install/manual-deployment/ Referring to steps 8-11, you'll notice that the manual process involves creating the monitor secret on step 8. Then, we generate a ceph.client.admin.keyring on step 9. See what happens on step 10 and 11? I add the ceph.client.admin.keyring contents to the monitor secret keyring; then, I feed that to step 11 for creating the monmap. Since ceph-deploy already creates the ceph.client.admin.keyring for you and populates it in the monmap, the fact that you are creating one after ceph-deploy has done this for you probably implies that you have overwritten the ceph.client.admin.keyring that was generated by ceph-deploy. The one you generated probably isn't in your monmap, so you are passing it the wrong key. If you had the right key or if you turn off cephx, you could execute ceph auth list to see the client.admin key contents. It's likely different from what you have in your ceph.client.admin.keyring file. ceph-deploy new generates a mon key. Then you deploy one or more monitors. Then, you use ceph-deploy gatherkeys. At that point, you should have a ceph.client.admin.keyring in the local/current directory after you executed ceph-deploy gatherkeys. Let me know if this helps. On Fri, Dec 6, 2013 at 7:59 AM, nicolasc nicolas.cance...@surfsara.nlwrote: Hi Dan, Thank you for the advice and indications. We have the exact same configuration, except I am only enabling auth cluster, and I am using ceph.client.admin.keyring instead of simply keyring. Both locations /etc/ceph/ceph.client.admin.keyring and /etc/ceph/keyring are presented as default values for the keyring configuration setting. I will try /etc/ceph/keyring, but I doubt this changes much. I am curious of whether your setup still works if you remove the keyring = /etc/ceph/keyring setting (you are also using a default location, so you could remove that line safely, right?). Thank you very much for answering. Best regards, Nicolas Canceill Scalable Storage Systems SURFsara (Amsterdam, NL) On 12/06/2013 11:42 AM, Dan Van Der Ster wrote: Hi, All of our clusters have this in ceph.conf: [global] auth cluster required = cephx auth service required = cephx auth client required = cephx keyring = /etc/ceph/keyring and the client.admin secret in /etc/ceph/keyring: # cat /etc/ceph/keyring [client.admin] key = ... With that you should be able to do ceph health without passing —id or —keyring args. (this is with dumpling, not emperor, but I guess it didn’t change.) If it still doesn’t work, check the capabilities that client.admin has (with ceph auth list). Should be caps: [mds] allow caps: [mon] allow * caps: [osd] allow * Cheers, Dan On 06 Dec 2013, at 11:06, nicolasc nicolas.cance...@surfsara.nl wrote: Hi every one, I did not get any answer to my basic cephx question last week, so let me ask it one more time here, before I completely give up on Ceph and move on. So, my issue is: When all authentication settings are none: * The cluster works fine * The file /etc/ceph/ceph.client.admin.keyring exists Then I set auth_cluster_required to cephx. When I try to connect to the cluster, it detects client.admin and denies access with operation not supported, even for commands like ceph health. Finally, after I explicitly set the keyring parameter in the config (to the default value, because the keyring file was already in the default location), the cluster works fine again. So the behavior changes when I add those 2 default lines to the config: [client.admin] keyring = /etc/ceph/ceph.client.admin.keyring From the ceph.com documentation [1], about this keyring parameter: Description:The path to the keyring file. Type: String Required: No Default:/etc/ceph/$cluster.$name.keyring ... so, I need help: * maybe this is a real bug? (was it already reported ?) * maybe I am deeply stupid, and I don't understand what required and default means? (can anyone send me a good dictionary ?) * maybe obi-wan kenobi? Thanks to anyone who will respond anything (at that point, even a three-letter e-mail reading ACK would make me feel better).
Re: [ceph-users] REST API issue for getting bucket policy
On Fri, Dec 6, 2013 at 1:45 AM, Gao, Wei M wei.m@intel.com wrote: Hi all, I am working on the ceph radosgw(v0.72.1) and when I call the rest api to read the bucket policy, I got an internal server error(request URL is: /admin/bucket?policyformat=jsonbucket=test.). However, when I call this: /admin/bucket?policyformat=jsonbucket=testobject=obj, I got the policy of the object returned. Besides, I do have the right permission(buckets=*). Any idea? Thanks! That's a bug. I opened issue #6940. Thanks, Yehuda ___ ceph-users mailing list ceph-users@lists.ceph.com http://lists.ceph.com/listinfo.cgi/ceph-users-ceph.com
[ceph-users] My experience with ceph now documentted
Hello Cephers I would like to say a BIG THANKS to ceph community for helping me in setting up and learning ceph. I have created a small documentation http://karan-mj.blogspot.fi/ of my experience with ceph till now , i belive it would help beginners in installing ceph and integrating it with openstack. I would keep updating this blog. PS -- i recommend original ceph documentation http://ceph.com/docs/master/ and other original content published by Ceph community , INKTANK and other partners. My attempt http://karan-mj.blogspot.fi/ is just to contribute for a regular online content about ceph. Karan Singh CSC - IT Center for Science Ltd. P.O. Box 405, FI-02101 Espoo, FINLAND http://www.csc.fi/ | +358 (0) 503 812758 ___ ceph-users mailing list ceph-users@lists.ceph.com http://lists.ceph.com/listinfo.cgi/ceph-users-ceph.com
Re: [ceph-users] centos6.4 + libvirt + qemu + rbd/ceph
On 12/06/2013 04:03 PM, Alek Paunov wrote: We use only Fedora servers for everything, so I am curious, why you are excluded this option from your research? (CentOS is always problematic with the new bits of technology). 6 months lifecycle and having to os-upgrade your entire data center 3 times a year? (OK maybe it's 18 months and once every 9 months) -- Dimitri Maziuk Programmer/sysadmin BioMagResBank, UW-Madison -- http://www.bmrb.wisc.edu signature.asc Description: OpenPGP digital signature ___ ceph-users mailing list ceph-users@lists.ceph.com http://lists.ceph.com/listinfo.cgi/ceph-users-ceph.com
Re: [ceph-users] centos6.4 + libvirt + qemu + rbd/ceph
On 07.12.2013 00:11, Dimitri Maziuk wrote: On 12/06/2013 04:03 PM, Alek Paunov wrote: We use only Fedora servers for everything, so I am curious, why you are excluded this option from your research? (CentOS is always problematic with the new bits of technology). 6 months lifecycle and having to os-upgrade your entire data center 3 times a year? (OK maybe it's 18 months and once every 9 months) Most servers novadays are re-provisioned even more often, but every new Fedora release comes with more and more KVM/Libvirt features and resolved issues, so the net effect is positive anyway. Yes, we need some extra tests to follow the cadence, just like ceph upgrades and the all other components. ___ ceph-users mailing list ceph-users@lists.ceph.com http://lists.ceph.com/listinfo.cgi/ceph-users-ceph.com
Re: [ceph-users] centos6.4 + libvirt + qemu + rbd/ceph
On 12/06/2013 04:28 PM, Alek Paunov wrote: On 07.12.2013 00:11, Dimitri Maziuk wrote: 6 months lifecycle and having to os-upgrade your entire data center 3 times a year? (OK maybe it's 18 months and once every 9 months) Most servers novadays are re-provisioned even more often, Not where I work they aren't. Fedora release comes with more and more KVM/Libvirt features and resolved issues, so the net effect is positive anyway. Yes, that is the main argument for tracking ubuntu. ;) -- Dimitri Maziuk Programmer/sysadmin BioMagResBank, UW-Madison -- http://www.bmrb.wisc.edu signature.asc Description: OpenPGP digital signature ___ ceph-users mailing list ceph-users@lists.ceph.com http://lists.ceph.com/listinfo.cgi/ceph-users-ceph.com
Re: [ceph-users] centos6.4 + libvirt + qemu + rbd/ceph
We rely on the stability of rhel/centos as well. We have no patch/upgrade policy or regulatory directive to do so. Our servers are set and forget. We circle back for patch/upgrades only for break/fix. I tried F19 just for the fun of it. We ended up with conflicts trying to run qemu-kvm with ceph. I could get one or the other working but not both. Our architecture is calling for compute and storage to live on the same host to save in hardware costs. I also tried to recompile libvirt and qemu-kvm today. I didn't even see rbd libraries in the source code. /C On Fri, Dec 6, 2013 at 5:56 PM, Dimitri Maziuk dmaz...@bmrb.wisc.eduwrote: On 12/06/2013 04:28 PM, Alek Paunov wrote: On 07.12.2013 00:11, Dimitri Maziuk wrote: 6 months lifecycle and having to os-upgrade your entire data center 3 times a year? (OK maybe it's 18 months and once every 9 months) Most servers novadays are re-provisioned even more often, Not where I work they aren't. Fedora release comes with more and more KVM/Libvirt features and resolved issues, so the net effect is positive anyway. Yes, that is the main argument for tracking ubuntu. ;) -- Dimitri Maziuk Programmer/sysadmin BioMagResBank, UW-Madison -- http://www.bmrb.wisc.edu ___ ceph-users mailing list ceph-users@lists.ceph.com http://lists.ceph.com/listinfo.cgi/ceph-users-ceph.com ___ ceph-users mailing list ceph-users@lists.ceph.com http://lists.ceph.com/listinfo.cgi/ceph-users-ceph.com
Re: [ceph-users] centos6.4 + libvirt + qemu + rbd/ceph
On 07.12.2013 01:03, Chris C wrote: We rely on the stability of rhel/centos as well. We have no patch/upgrade policy or regulatory directive to do so. Our servers are set and forget. We circle back for patch/upgrades only for break/fix. Stability means keeping the ABIs (and in general all interfaces and conventions) stable. It is very important when e.g. you intent to deploy some old Sybase on these boxes. How this type of stability helps the Ceph/KVM node ... ? I tried F19 just for the fun of it. We ended up with conflicts trying to run qemu-kvm with ceph. I could get one or the other working but not both. Our architecture is calling for compute and storage to live on the same host to save in hardware costs. I also tried to recompile libvirt and qemu-kvm today. I didn't even see rbd libraries in the source code. OSD/libvirt-kvm dual role node should work just fine with F19/F20. If you are interested in Fedora deployments, we could try to resolve these issues. Alek ___ ceph-users mailing list ceph-users@lists.ceph.com http://lists.ceph.com/listinfo.cgi/ceph-users-ceph.com
Re: [ceph-users] Ephemeral RBD with Havana and Dumpling
On 12/05/2013 02:37 PM, Dmitry Borodaenko wrote: Josh, On Tue, Nov 19, 2013 at 4:24 PM, Josh Durgin josh.dur...@inktank.com wrote: I hope I can release or push commits to this branch contains live-migration, incorrect filesystem size fix and ceph-snapshort support in a few days. Can't wait to see this patch! Are you getting rid of the shared storage requirement for live-migration? Yes, that's what Haomai's patch will fix for rbd-based ephemeral volumes (bug https://bugs.launchpad.net/nova/+bug/1250751). We've got a version of a Nova patch that makes live migrations work for non volume-backed instances, and hopefully addresses the concerns raised in code review in https://review.openstack.org/56527, along with a bunch of small bugfixes, e.g. missing max_size parameter in direct_fetch, and a fix for http://tracker.ceph.com/issues/6693. I have submitted it as a pull request to your nova fork on GitHub: https://github.com/jdurgin/nova/pull/1 Thanks! Our changes depend on the rest of commits on your havana-ephemeral-rbd branch, and the whole patchset is now at 7 commits, which is going to be rather tedious to submit to the OpenStack Gerrit as a series of dependent changes. Do you think we should keep the current commit history in its current form, or would it be easier to squash it down to a more manageable number of patches? As discussed on irc yesterday, most of these are submitted to icehouse already in slightly different form, since this branch is based on stable/havana. I'd prefer to keep the commits small and self contained in this branch at least. If it takes too long to get them upstream, I'm fine with having them squashed for faster upstream review. Josh ___ ceph-users mailing list ceph-users@lists.ceph.com http://lists.ceph.com/listinfo.cgi/ceph-users-ceph.com