Hello, while reproducing the problem, objecter_requests looks like this:
{
"ops": [
{
"tid": 42029,
"pg": "5.bd9616ad",
"osd": 46,
"object_id": "rbd_data.e10ca56b8b4567.000000000000311c",
"object_locator": "@5",
"target_object_id": "rbd_data.e10ca56b8b4567.000000000000311c",
"target_object_locator": "@5",
"paused": 0,
"used_replica": 0,
"precalc_pgid": 0,
"last_sent": "2.28854e+06s",
"attempts": 1,
"snapid": "head",
"snap_context": "a07c2=[]",
"mtime": "2017-05-16 21:53:22.0.069541s",
"osd_ops": [
"delete"
]
}
],
"linger_ops": [
{
"linger_id": 1,
"pg": "5.5f3bd635",
"osd": 17,
"object_id": "rbd_header.e10ca56b8b4567",
"object_locator": "@5",
"target_object_id": "rbd_header.e10ca56b8b4567",
"target_object_locator": "@5",
"paused": 0,
"used_replica": 0,
"precalc_pgid": 0,
"snapid": "head",
"registered": "1"
}
],
"pool_ops": [],
"pool_stat_ops": [],
"statfs_ops": [],
"command_ops": []
}
Yes they've an established TCP connection. Qemu <=> osd.46. Attached is
a pcap file of the traffic between them when it got stuck.
Greets,
Stefan
Am 16.05.2017 um 21:45 schrieb Jason Dillaman:
> On Tue, May 16, 2017 at 3:37 PM, Stefan Priebe - Profihost AG
> <[email protected]> wrote:
>> We've enabled the op tracker for performance reasons while using SSD
>> only storage ;-(
>
> Disabled you mean?
>
>> Can enable the op tracker using ceph osd tell? Than reproduce the
>> problem. Check what has stucked again? Or should i generate an rbd log
>> from the client?
>
> From a super-quick glance at the code, it looks like that isn't a
> dynamic setting. Of course, it's possible that if you restart OSD 46
> to enable the op tracker, the stuck op will clear itself and the VM
> will resume. You could attempt to generate a gcore of OSD 46 to see if
> information on that op could be extracted via the debugger, but no
> guarantees.
>
> You might want to verify that the stuck client and OSD 46 have an
> actual established TCP connection as well before doing any further
> actions.
>
osd.46_qemu_2.pcap.gz
Description: GNU Zip compressed data
_______________________________________________ ceph-users mailing list [email protected] http://lists.ceph.com/listinfo.cgi/ceph-users-ceph.com
