Re: Ceph on Raspberry Pi

[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]

 



Alfredo,

Thank you for your response.  I simply did apt-get install ceph on the nodes.

My /etc/apt/sources.list.d/ceph.list contains:

deb http://ceph.com/debian-emperor/ wheezy main

and the versions I received are what I got.

  Shlomo

-----------------------------
Shlomo Dubrowin

The Solution to the water crisis in Israel:

# According to WikiPedia, the Kinneret can hold
# 4 km^3, so FULL here is in cubit meters
FULL="4000"
while [ "$LEVEL" -lt "$FULL" ]; do
  cat /sea/med /sea/red |\
  grep -vi "salt" |\
  tee /sea/dead /lake/kinneret
  LEVEL=`du -c /sea/dead /lake/kinneret | grep total | awk '{print $1}'`
done


On Tue, Dec 3, 2013 at 4:15 PM, Alfredo Deza <alfredo.deza@xxxxxxxxxxx> wrote:
On Tue, Dec 3, 2013 at 8:55 AM, Shlomo Dubrowin <shlomo@xxxxxxxxxxxx> wrote:
> I'm trying to deploy Ceph on a group of Raspberry Pis using the procedure
> documented in: http://ceph.com/docs/master/start/quick-ceph-deploy/
>
> There used to be a site: http://ceph.com/docs/master/start/quick-start/ but
> that page is no longer valid.
>
> The first thing I noticed is that the command lsb_release -sc specified in
> the Ceph Deploy setup provides n/a even though the answer should be wheezy.
> I manually changed the /etc/apt/sources.list.d/ceph.list to specify wheezy.
>
> I've installed ceph-deploy and setup the keys to communicate between the
> deploy Raspberry Pi and all the Raspberry Pis including the system reunning
> ceph-deploy.
>
> When I tried to run ceph-deploy install <node1> the installation failed.
> I went to each node and installed ceph manually and I see the versions are:
>
> $ ceph-deploy --version
> 1.3.3
>
> $ ceph --version
> ceph version 0.43 (commit:9fa8781c0147d66fcef7c2dd0e09cd3c69747d37)
>
>
> All the nodes have the same ceph version.

That looks like a *very* old ceph version. Is there any reason you are
using 0.43 as opposed to the latest one?

How did you installed Ceph on those nodes?
>
> When I try to run a command, I get errors:
>
> $ ceph-deploy mon create baxter
> [ceph_deploy.cli][INFO  ] Invoked (1.3.3): /usr/bin/ceph-deploy mon create
> baxter
> [ceph_deploy.mon][DEBUG ] Deploying mon, cluster ceph hosts baxter
> [ceph_deploy.mon][DEBUG ] detecting platform for host baxter ...
> [baxter][DEBUG ] connected to host: baxter
> [baxter][DEBUG ] detect platform information from remote host
> [baxter][DEBUG ] detect machine type
> [ceph_deploy.mon][INFO  ] distro info: debian 7.0 wheezy
> [baxter][DEBUG ] determining if provided host has same hostname in remote
> [baxter][DEBUG ] get remote short hostname
> [baxter][DEBUG ] deploying mon to baxter
> [baxter][DEBUG ] get remote short hostname
> [baxter][DEBUG ] remote hostname: baxter
> [baxter][DEBUG ] write cluster configuration to /etc/ceph/{cluster}.conf
> [baxter][DEBUG ] create the mon path if it does not exist
> [baxter][DEBUG ] checking for done path: /var/lib/ceph/mon/ceph-baxter/done
> [baxter][DEBUG ] done path does not exist:
> /var/lib/ceph/mon/ceph-baxter/done
> [baxter][INFO  ] creating keyring file:
> /var/lib/ceph/tmp/ceph-baxter.mon.keyring
> [baxter][DEBUG ] create the monitor keyring file
> [baxter][INFO  ] Running command: sudo ceph-mon --cluster ceph --mkfs -i
> baxter --keyring /var/lib/ceph/tmp/ceph-baxter.mon.keyring
> [baxter][WARNIN] too many arguments: [--cluster,ceph]
> [baxter][WARNIN] usage: ceph-mon -i monid [--mon-data="" [flags]
> [baxter][WARNIN]   --debug_mon n
> [baxter][WARNIN]         debug monitor level (e.g. 10)
> [baxter][WARNIN]   --mkfs
> [baxter][WARNIN]         build fresh monitor fs
> [baxter][DEBUG ] --conf/-c        Read configuration from the given
> configuration file
> [baxter][DEBUG ] -d               Run in foreground, log to stderr.
> [baxter][DEBUG ] -f               Run in foreground, log to usual location.
> [baxter][DEBUG ] --id/-i          set ID portion of my name
> [baxter][DEBUG ] --name/-n        set name (TYPE.ID)
> [baxter][DEBUG ] --version        show version and quit
> [baxter][DEBUG ]
> [baxter][DEBUG ]    --debug_ms N
> [baxter][DEBUG ]         set message debug level (e.g. 1)
> [baxter][ERROR ] RuntimeError: command returned non-zero exit status: 1
> [ceph_deploy.mon][ERROR ] Failed to execute command: ceph-mon --cluster ceph
> --mkfs -i baxter --keyring /var/lib/ceph/tmp/ceph-baxter.mon.keyring
> [ceph_deploy][ERROR ] GenericError: Failed to create 1 monitors
>
>
> If I try to run the same command as the user ceph on the local machine, I
> get the same error:
>
> $ ceph-mon --cluster ceph --mkfs -i baxter --keyring
> /var/lib/ceph/tmp/ceph-baxter.mon.keyring
> failed to open log file '/var/log/ceph/mon.baxter.log': (13) Permission
> denied
> too many arguments: [--cluster,ceph]
> usage: ceph-mon -i monid [--mon-data="" [flags]
>   --debug_mon n
>         debug monitor level (e.g. 10)
>   --mkfs
>         build fresh monitor fs
> --conf/-c        Read configuration from the given configuration file
> -d               Run in foreground, log to stderr.
> -f               Run in foreground, log to usual location.
> --id/-i          set ID portion of my name
> --name/-n        set name (TYPE.ID)
> --version        show version and quit
>
>    --debug_ms N
>         set message debug level (e.g. 1)
>
>
>
> This makes me think that the ceph-deploy installed doesn't work with the
> ceph installed on each node.
>
> So I guess I have 2 questions:
>
> How can I determine which ceph-deploy goes with which ceph and how can I
> correct this apparent mismatch?

I don't think there was a ceph-deploy per-se that would've worked for
that Ceph version.

I believe that ceph-deploy is very backwards compatible for a few Ceph
versions. At the very least we should
fully support the latest 3 major versions and we special case a few
commands depending on the version installed.

But like I mentioned, the Ceph version you have installed seems like a
very very old one.

> Should I be looking at just using local ceph commands to configure this by
> hand? Are there instructions on how to do this by hand?

If you must have that version I don't think there is any option here
but to do it by hand and avoid ceph-deploy.

For docs, I think that might be even more difficult to do as you would
need to go to the tag for that version and look at the docs
for that tag in the repository.

Then again, what is the reason for using that version?

>
> Thank you.
>
>   Shlomo
>
> -----------------------------
> Shlomo Dubrowin
>
> The Solution to the water crisis in Israel:
>
> # According to WikiPedia, the Kinneret can hold
> # 4 km^3, so FULL here is in cubit meters
> FULL="4000"
> while [ "$LEVEL" -lt "$FULL" ]; do
>   cat /sea/med /sea/red |\
>   grep -vi "salt" |\
>   tee /sea/dead /lake/kinneret
>   LEVEL=`du -c /sea/dead /lake/kinneret | grep total | awk '{print $1}'`
> done
>
> _______________________________________________
> ceph-users mailing list
> ceph-users@xxxxxxxxxxxxxx
> http://lists.ceph.com/listinfo.cgi/ceph-users-ceph.com
>

_______________________________________________
ceph-users mailing list
ceph-users@xxxxxxxxxxxxxx
http://lists.ceph.com/listinfo.cgi/ceph-users-ceph.com

[Index of Archives]     [Information on CEPH]     [Linux Filesystem Development]     [Ceph Development]     [Ceph Large]     [Linux USB Development]     [Video for Linux]     [Linux Audio Users]     [Yosemite News]     [Linux Kernel]     [Linux SCSI]     [xfs]


  Powered by Linux