Re:

[Alex Elder <elder@inktank.com>]

On 10/25/2012 07:15 AM, Gregory Farnum wrote:
> Sorry, I was unclear — I meant I think[1] it was fixed in our linux
> branch, for future kernel releases. The messages you're seeing are
> just logging a perfectly normal event that's part of the Ceph
> protocol. -Greg [1]: I'd have to check to make sure. Sage, Alex, am I
> remembering that correctly?

I see those too.  I think the socket (the other end?) closes after
a period of inactivity, but it does re-open and reconnect again
whenever necessary so that should really be fine.

The messages have not gone away yet, I personally think they should.
They originate from here, in net/ceph/messenger.c:

  static void ceph_fault(struct ceph_connection *con)
          __releases(con->mutex)
  {
          pr_err("%s%lld %s %s\n", ENTITY_NAME(con->peer_name),
                 ceph_pr_addr(&con->peer_addr.in_addr), con->error_msg)

Perhaps this should become pr_info() or something.  Sage?

					-Alex

> On Wednesday, October 24, 2012 at 11:45 PM, jie sun wrote:
> 
>> What is the version of the master branch ? I use the stable version
>> 0.48.2 Thank you! -SunJie
>> 
>> 2012/10/24 Gregory Farnum <greg@inktank.com>:
>>> On Tuesday, October 23, 2012 at 10:48 PM, jie sun wrote:
>>>> My vm kernel version is "Linux ubuntu12 3.2.0-23-generic".
>>>> 
>>>> "ceph-s" shows " health HEALTH_OK monmap e1: 1 mons at
>>>> {a=10.100.211.146:6789/0}, election epoch 0, quorum 0 a osdmap
>>>> e152: 10 osds: 9 up, 9 in pgmap v48479: 2112 pgs: 2112
>>>> active+clean; 23161 MB data, 46323 MB used, 2451 GB / 2514 GB
>>>> avail mdsmap e31: 1/1/1 up {0=a=up:active} "
>>>> 
>>>> In my vm, I do operations like: I install 4 debs on my vm, such
>>>> as libnss3, libnspr4, librados2, librbd1. And then execute
>>>> "modprobe rbd" so that I can map a image to my vm. Then "rbd
>>>> create foo --size 10240 -m $monIP(my ceph mon IP)", "rbd map
>>>> foo -m $monIP" ------ Here a device /dev/rbd0 can be used as a
>>>> local device "mkfs -t ext4 /dev/rbd0" "mount /dev/rbd0 /mnt(or
>>>> some other directory)" After the operations above, I can use
>>>> this device. But it oftern prompt some log like "libceph: osd9
>>>> 10.100.211.68:6809 socket closed". I just want to mount a
>>>> device to my vm, so I didn't install a ceph client. Is this
>>>> proper to do so?
>>> 
>>> 
>>> 
>>> You might consider using the native QEMU/libvirt instead; it
>>> offers some more advanced options. But if you're happy with it,
>>> this certainly works!
>>> 
>>> The "socket closed" messages are just noise; it's nothing to be
>>> concerned about (you'll notice they're happening every 15 minutes
>>> for each OSD; probably you aren't doing any disk accesses). I
>>> think these warnings actually got removed from our master branch
>>> a few days ago. -Greg
>> 
> 
> 
> 
> -- To unsubscribe from this list: send the line "unsubscribe
> ceph-devel" in the body of a message to majordomo@vger.kernel.org 
> More majordomo info at  http://vger.kernel.org/majordomo-info.html
> 

--
To unsubscribe from this list: send the line "unsubscribe ceph-devel" in
the body of a message to majordomo@vger.kernel.org
More majordomo info at  http://vger.kernel.org/majordomo-info.html