Re: ceph-deploy hang on CentOS 6.4

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

 




>-----Original Message-----
>From: Alfredo Deza [mailto:alfredo.deza@xxxxxxxxxxx]
>Sent: Thursday, October 24, 2013 5:24 AM
>To: Gruher, Joseph R
>Cc: ceph-users@xxxxxxxxxxxxxx
>Subject: Re:  ceph-deploy hang on CentOS 6.4
>
>On Wed, Oct 23, 2013 at 12:43 PM, Gruher, Joseph R
><joseph.r.gruher@xxxxxxxxx> wrote:
>>
>>
>>>-----Original Message-----
>>>From: Alfredo Deza [mailto:alfredo.deza@xxxxxxxxxxx]
>>>
>>>Did you tried working with the `--no-adjust-repos` flag in ceph-deploy
>>>? It will allow you to tell ceph-deploy to just go and install ceph
>>>without attempting to import keys or doing anything with your repos.
>>
>> I have tried this in the past but it caused problems further down the install
>process, I believe due to old or mismatched versions being installed.  That was
>on Ubuntu 12.04.2.  It was discussed a bit on this list at the time.  I would not
>recommend --no-adjust-repos based on my experience.
>
>It would be very useful to have logs or a reproducible scenario so we can
>improve this. Mismatched versions doesn't sound like something ceph-deploy
>would do specifically, but rather a problem with installing and removing
>packages and having issues there.

I've attached the previous thread where I ran into the issue with --no-adjust-repos.  Not sure if attachments are allowed by the list, if it doesn't get through let me know and I can forward it to you.

To summarize, it would let me work around proxy requirements during "ceph-deploy install", but then I would run into an issue during when attempting "ceph-deploy mon create".  After resolving the proxy issue and removing --no-adjust-repos the issue during "ceph-deploy mon create" disappeared.  The assumption is that they are related but that would require more detailed investigation to prove.  See attachment for details.

To reproduce (at the time) I was just running Ubuntu 12.04.2 with 3.6.10 kernel, then installing ceph-deploy and doing a ceph-deploy new, ceph-deploy install with --no-adjust-repos, and then ceph-deploy mon create where I would get hung up on what appears to be an invalid argument --cluster to ceph-mon.  After removing --no-adjust-repos same steps would complete successfully.  The problem may be that the --cluster flag wasn't supported by the version of ceph installed.

>>
>>>The documentation for this can be found here:
>>>https://github.com/ceph/ceph-deploy#proxy-or-firewall-installs
>>
>> This doc only mentions setting the wget proxy, I would suggest it be
>updated to include the curl and rpm proxies may need to be set as well.
>
>Yes, I will be updating those too, thanks for the examples!
>>
>> Thanks,
>> Joe
>>
--- Begin Message ---

>-----Original Message-----
>From: Alfredo Deza [mailto:alfredo.deza@xxxxxxxxxxx]
>Sent: Monday, September 23, 2013 5:45 AM
>To: Gruher, Joseph R
>Cc: ceph-users@xxxxxxxxxxxxxx
>Subject: Re:  monitor deployment during quick start
>
>On Fri, Sep 20, 2013 at 3:58 PM, Gruher, Joseph R
><joseph.r.gruher@xxxxxxxxx> wrote:
>> Sorry, not trying to repost or bump my thread, but I think I can restate my
>question here and for better clarity.  I am confused about the "--cluster"
>argument used when "ceph-deploy mon create" invokes "ceph-mon" on the
>target system.  I always get a failure at this point when running "ceph-deploy
>mon create" and this then halts the whole ceph quick start process.
>>
>> Here is the line where "ceph-deploy mon create" fails:
>> [cephtest02][INFO  ] Running command: ceph-mon --cluster ceph --mkfs
>> -i cephtest02 --keyring /var/lib/ceph/tmp/ceph-cephtest02.mon.keyring
>>
>> Running the same command manually on the target system gives an error.
>As far as I can tell from the man page and the built-in help and the website
>(http://ceph.com/docs/next/man/8/ceph-mon/) it seems "--cluster" is not a
>valid argument for ceph-mon?  Is this a problem in ceph-deploy?  Does this
>work for anyone else?
>>
>> ceph@cephtest02:~$ sudo ceph-mon --cluster ceph --mkfs -i cephtest02
>> --keyring /var/lib/ceph/tmp/ceph-cephtest02.mon.keyring
>> too many arguments: [--cluster,ceph]
>> usage: ceph-mon -i monid [--mon-data=pathtodata] [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) ceph@cephtest02:~$
>>
>> Can anyone clarify if "--cluster" is a supported argument for ceph-mon?
>
>This is a *weird* corner you've stumbled upon. The flag is indeed used by
>ceph-deploy and that hasn't changed in a while. However, as you point out,
>there is no trace of that flag anywhere! I can't find where is that defined at all.
>
>Running the latest version of ceph-deploy + ceph, that flag *does* work for
>me.
>
>What version of ceph are you using?

I think a version issue is likely.  I was running "ceph-deploy install" on Ubuntu 12.04.2 with the --no-adjust-repos switch because the wget it would execute was hanging for me.  Ultimately the wget problem seemed to be a proxy issue.  Even though I had proxy configured globally (/etc/environment), and could successfully execute my own wgets on the command line, I ultimately had to directly configure the proxy in the wget configuration (/etc/wgetrc) to avoid the hang in "ceph-deploy install".  Once that was resolved, I removed the --no-adjust-repos switch from "deph-deploy install", and that in turn seemed to resolve the later failure in "ceph-deploy mon create".  It all runs fine for me now.  Unfortunately I did not specifically track the resulting ceph versions when deploying with and without --no-adjust-repos.  I can tell you after resolving the issue I am currently running ceph-deploy 1.2.6 and I assume it is deploying Ceph 0.61.

What's the best way to check the Ceph version on a running cluster?

--- End Message ---
_______________________________________________
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]     [Ceph Dev]     [Linux USB Development]     [Video for Linux]     [Linux Audio Users]     [Yosemite News]     [Linux Kernel]     [Linux SCSI]     [xfs]


  Powered by Linux