Hi, what version of rocky are you using. Maybe was in the middle of a backport which temporally broke ceph. <div><br></div><div>Could you try latest stable/rocky branch? <br><div dir="auto"><br></div><div dir="auto">It is now working properly. </div><div dir="auto"><br></div><div dir="auto">Regards<br><br><div class="gmail_quote"><div dir="ltr">On Wed, Sep 26, 2018, 2:32 PM Florian Engelmann <<a href="mailto:florian.engelmann@everyware.ch">florian.engelmann@everyware.ch</a>> wrote:<br></div><blockquote class="gmail_quote" style="margin:0 0 0 .8ex;border-left:1px #ccc solid;padding-left:1ex">Hi,<br>
<br>
I tried to deploy Rocky in a multinode setup but ceph-osd fails with:<br>
<br>
<br>
failed: [xxxxxxxxxxx-poc2] (item=[0, {u'fs_uuid': u'', u'bs_wal_label': <br>
u'', u'external_journal': False, u'bs_blk_label': u'', <br>
u'bs_db_partition_num': u'', u'journal_device': u'', u'journal': u'', <br>
u'partition': u'/dev/nvme0n1', u'bs_wal_partition_num': u'', <br>
u'fs_label': u'', u'journal_num': 0, u'bs_wal_device': u'', <br>
u'partition_num': u'1', u'bs_db_label': u'', u'bs_blk_partition_num': <br>
u'', u'device': u'/dev/nvme0n1', u'bs_db_device': u'', <br>
u'partition_label': u'KOLLA_CEPH_OSD_BOOTSTRAP_BS', u'bs_blk_device': <br>
u''}]) => {<br>
     "changed": true,<br>
     "item": [<br>
         0,<br>
         {<br>
             "bs_blk_device": "",<br>
             "bs_blk_label": "",<br>
             "bs_blk_partition_num": "",<br>
             "bs_db_device": "",<br>
             "bs_db_label": "",<br>
             "bs_db_partition_num": "",<br>
             "bs_wal_device": "",<br>
             "bs_wal_label": "",<br>
             "bs_wal_partition_num": "",<br>
             "device": "/dev/nvme0n1",<br>
             "external_journal": false,<br>
             "fs_label": "",<br>
             "fs_uuid": "",<br>
             "journal": "",<br>
             "journal_device": "",<br>
             "journal_num": 0,<br>
             "partition": "/dev/nvme0n1",<br>
             "partition_label": "KOLLA_CEPH_OSD_BOOTSTRAP_BS",<br>
             "partition_num": "1"<br>
         }<br>
     ]<br>
}<br>
<br>
MSG:<br>
<br>
Container exited with non-zero return code 2<br>
<br>
We tried to debug the error message by starting the container with a <br>
modified endpoint but we are stuck at the following point right now:<br>
<br>
<br>
docker run  -e "HOSTNAME=10.0.153.11" -e "JOURNAL_DEV=" -e <br>
"JOURNAL_PARTITION=" -e "JOURNAL_PARTITION_NUM=0" -e <br>
"KOLLA_BOOTSTRAP=null" -e "KOLLA_CONFIG_STRATEGY=COPY_ALWAYS" -e <br>
"KOLLA_SERVICE_NAME=bootstrap-osd-0" -e "OSD_BS_BLK_DEV=" -e <br>
"OSD_BS_BLK_LABEL=" -e "OSD_BS_BLK_PARTNUM=" -e "OSD_BS_DB_DEV=" -e <br>
"OSD_BS_DB_LABEL=" -e "OSD_BS_DB_PARTNUM=" -e "OSD_BS_DEV=/dev/nvme0n1" <br>
-e "OSD_BS_LABEL=KOLLA_CEPH_OSD_BOOTSTRAP_BS" -e "OSD_BS_PARTNUM=1" -e <br>
"OSD_BS_WAL_DEV=" -e "OSD_BS_WAL_LABEL=" -e "OSD_BS_WAL_PARTNUM=" -e <br>
"OSD_DEV=/dev/nvme0n1" -e "OSD_FILESYSTEM=xfs" -e "OSD_INITIAL_WEIGHT=1" <br>
-e "OSD_PARTITION=/dev/nvme0n1" -e "OSD_PARTITION_NUM=1" -e <br>
"OSD_STORETYPE=bluestore" -e "USE_EXTERNAL_JOURNAL=false"   -v <br>
"/etc/kolla//ceph-osd/:/var/lib/kolla/config_files/:ro" -v <br>
"/etc/localtime:/etc/localtime:ro" -v "/dev/:/dev/" -v <br>
"kolla_logs:/var/log/kolla/" -ti --privileged=true --entrypoint <br>
/bin/bash <br>
<a href="http://10.0.128.7:5000/openstack/openstack-kolla-cfg/ubuntu-source-ceph-osd:7.0.0.3" rel="noreferrer" target="_blank">10.0.128.7:5000/openstack/openstack-kolla-cfg/ubuntu-source-ceph-osd:7.0.0.3</a><br>
<br>
<br>
<br>
cat /var/lib/kolla/config_files/ceph.client.admin.keyring > <br>
/etc/ceph/ceph.client.admin.keyring<br>
<br>
<br>
cat /var/lib/kolla/config_files/ceph.conf > /etc/ceph/ceph.conf<br>
<br>
<br>
(bootstrap-osd-0)[root@985e2dee22bc /]# /usr/bin/ceph-osd -d <br>
--public-addr 10.0.153.11 --cluster-addr 10.0.153.11<br>
usage: ceph-osd -i <ID> [flags]<br>
   --osd-data PATH data directory<br>
   --osd-journal PATH<br>
                     journal file or block device<br>
   --mkfs            create a [new] data directory<br>
   --mkkey           generate a new secret key. This is normally used in <br>
combination with --mkfs<br>
   --convert-filestore<br>
                     run any pending upgrade operations<br>
   --flush-journal   flush all data out of journal<br>
   --mkjournal       initialize a new journal<br>
   --check-wants-journal<br>
                     check whether a journal is desired<br>
   --check-allows-journal<br>
                     check whether a journal is allowed<br>
   --check-needs-journal<br>
                     check whether a journal is required<br>
   --debug_osd <N>   set debug level (e.g. 10)<br>
   --get-device-fsid PATH<br>
                     get OSD fsid for the given block device<br>
<br>
   --conf/-c FILE    read configuration from the given configuration file<br>
   --id/-i ID        set ID portion of my name<br>
   --name/-n <a href="http://TYPE.ID" rel="noreferrer" target="_blank">TYPE.ID</a> set name<br>
   --cluster NAME    set cluster name (default: ceph)<br>
   --setuser USER    set uid to user or uid (and gid to user's gid)<br>
   --setgroup GROUP  set gid to group or gid<br>
   --version         show version and quit<br>
<br>
   -d                run in foreground, log to stderr.<br>
   -f                run in foreground, log to usual location.<br>
   --debug_ms N      set message debug level (e.g. 1)<br>
2018-09-26 12:28:07.801066 7fbda64b4e40  0 ceph version 12.2.4 <br>
(52085d5249a80c5f5121a76d6288429f35e4e77b) luminous (stable), process <br>
(unknown), pid 46<br>
2018-09-26 12:28:07.801078 7fbda64b4e40 -1 must specify '-i #' where # <br>
is the osd number<br>
<br>
<br>
But it looks like "-i" is not set anywere?<br>
<br>
grep command <br>
/opt/stack/kolla-ansible/ansible/roles/ceph/templates/ceph-osd.json.j2<br>
"command": "/usr/bin/ceph-osd -f --public-addr {{ <br>
hostvars[inventory_hostname]['ansible_' + <br>
storage_interface]['ipv4']['address'] }} --cluster-addr {{ <br>
hostvars[inventory_hostname]['ansible_' + <br>
cluster_interface]['ipv4']['address'] }}",<br>
<br>
What's wrong with our setup?<br>
<br>
All the best,<br>
Flo<br>
<br>
<br>
-- <br>
<br>
EveryWare AG<br>
Florian Engelmann<br>
Systems Engineer<br>
Zurlindenstrasse 52a<br>
CH-8003 Zürich<br>
<br>
tel: +41 44 466 60 00<br>
fax: +41 44 466 60 10<br>
mail: mailto:<a href="mailto:florian.engelmann@everyware.ch" target="_blank">florian.engelmann@everyware.ch</a><br>
web: <a href="http://www.everyware.ch" rel="noreferrer" target="_blank">http://www.everyware.ch</a><br>
__________________________________________________________________________<br>
OpenStack Development Mailing List (not for usage questions)<br>
Unsubscribe: <a href="http://OpenStack-dev-request@lists.openstack.org?subject:unsubscribe" rel="noreferrer" target="_blank">OpenStack-dev-request@lists.openstack.org?subject:unsubscribe</a><br>
<a href="http://lists.openstack.org/cgi-bin/mailman/listinfo/openstack-dev" rel="noreferrer" target="_blank">http://lists.openstack.org/cgi-bin/mailman/listinfo/openstack-dev</a><br>
</blockquote></div></div></div>