* client released lease on dn which does not exist
@ 2010-07-17 17:07 Sébastien Paolacci
2010-07-28 11:27 ` Sébastien Paolacci
0 siblings, 1 reply; 3+ messages in thread
From: Sébastien Paolacci @ 2010-07-17 17:07 UTC (permalink / raw)
To: ceph-devel
Hello all,
I'm getting some warnings that I've not been able to find informations
for when running dbench on a newly setup box (nettop, atom dual core
2.5" very low-end hdd). Seems however to be easily reproducible
(unstable server and kernel client from yesterday) :
srv:# mkcephfs -c ceph.conf --allhosts --mkbtrfs
srv:# for srv in cosd cmon cmds; do $srv -c ceph.conf -i 0; done
client:# mount -t ceph srv:/ /mnt/
client:# dbench -D /mnt/ 1
Few seconds later:
srv:# ceph -w
10.07.17_18:40:12.797372 pg v7: 264 pgs: 264 active+clean+degraded;
69 KB data, 508 KB used, 450 GB / 450 GB avail; 10/20 degraded
(50.000%)
10.07.17_18:40:12.800218 mds e4: 1/1/1 up {0=up:active}
10.07.17_18:40:12.800303 osd e3: 1 osds: 1 up, 1 in
10.07.17_18:40:12.800448 log 10.07.17_18:39:56.072957 mon0
192.168.0.3:6789/0 4 : [INF] mds0 192.168.0.3:6802/6754 up:active
10.07.17_18:40:12.800598 mon e1: 1 mons at 192.168.0.3:6789/0
10.07.17_18:40:14.234588 pg v8: 264 pgs: 264 active+clean+degraded;
1166 KB data, 1280 KB used, 450 GB / 450 GB avail; 28/56 degraded
(50.000%)
10.07.17_18:40:18.159014 log 10.07.17_18:40:12.067141 mds0
192.168.0.3:6802/6754 1 : [WRN] client4099 released lease on dn
1000000000c/RESULTS3.PRN which dne
10.07.17_18:40:19.199284 pg v9: 264 pgs: 264 active+clean+degraded;
2103 KB data, 2536 KB used, 450 GB / 450 GB avail; 33/66 degraded
(50.000%)
Does anyone know how "severe" the warning is and what could I
undertake to get rid of it?
Thanks,
Sebastien
ps: I can (separately) provide the 90MB mds log file should it be needed.
[global]
[mon]
mon data = /data/mon$id
[mon0]
host = srv
mon addr = 192.168.0.3:6789
[mds]
debug ms = 1
debug mds = 20
[mds0]
host = srv
[osd]
osd data = /data/osd$id
osd journal = /dev/sda3
[osd0]
host = srv
btrfs devs = /dev/sda4
^ permalink raw reply [flat|nested] 3+ messages in thread
* Re: client released lease on dn which does not exist
2010-07-17 17:07 client released lease on dn which does not exist Sébastien Paolacci
@ 2010-07-28 11:27 ` Sébastien Paolacci
2010-07-28 16:37 ` Sage Weil
0 siblings, 1 reply; 3+ messages in thread
From: Sébastien Paolacci @ 2010-07-28 11:27 UTC (permalink / raw)
To: ceph-devel
FYI, the same test pass without any warning now (unstable from yesterday).
Sebastien
2010/7/17 Sébastien Paolacci <sebastien.paolacci@gmail.com>:
> Hello all,
>
> I'm getting some warnings that I've not been able to find informations
> for when running dbench on a newly setup box (nettop, atom dual core
> 2.5" very low-end hdd). Seems however to be easily reproducible
> (unstable server and kernel client from yesterday) :
>
> srv:# mkcephfs -c ceph.conf --allhosts --mkbtrfs
> srv:# for srv in cosd cmon cmds; do $srv -c ceph.conf -i 0; done
>
> client:# mount -t ceph srv:/ /mnt/
> client:# dbench -D /mnt/ 1
>
> Few seconds later:
> srv:# ceph -w
> 10.07.17_18:40:12.797372 pg v7: 264 pgs: 264 active+clean+degraded;
> 69 KB data, 508 KB used, 450 GB / 450 GB avail; 10/20 degraded
> (50.000%)
> 10.07.17_18:40:12.800218 mds e4: 1/1/1 up {0=up:active}
> 10.07.17_18:40:12.800303 osd e3: 1 osds: 1 up, 1 in
> 10.07.17_18:40:12.800448 log 10.07.17_18:39:56.072957 mon0
> 192.168.0.3:6789/0 4 : [INF] mds0 192.168.0.3:6802/6754 up:active
> 10.07.17_18:40:12.800598 mon e1: 1 mons at 192.168.0.3:6789/0
> 10.07.17_18:40:14.234588 pg v8: 264 pgs: 264 active+clean+degraded;
> 1166 KB data, 1280 KB used, 450 GB / 450 GB avail; 28/56 degraded
> (50.000%)
> 10.07.17_18:40:18.159014 log 10.07.17_18:40:12.067141 mds0
> 192.168.0.3:6802/6754 1 : [WRN] client4099 released lease on dn
> 1000000000c/RESULTS3.PRN which dne
> 10.07.17_18:40:19.199284 pg v9: 264 pgs: 264 active+clean+degraded;
> 2103 KB data, 2536 KB used, 450 GB / 450 GB avail; 33/66 degraded
> (50.000%)
>
> Does anyone know how "severe" the warning is and what could I
> undertake to get rid of it?
>
> Thanks,
> Sebastien
>
> ps: I can (separately) provide the 90MB mds log file should it be needed.
>
> [global]
> [mon]
> mon data = /data/mon$id
> [mon0]
> host = srv
> mon addr = 192.168.0.3:6789
> [mds]
> debug ms = 1
> debug mds = 20
> [mds0]
> host = srv
> [osd]
> osd data = /data/osd$id
> osd journal = /dev/sda3
> [osd0]
> host = srv
> btrfs devs = /dev/sda4
>
--
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
^ permalink raw reply [flat|nested] 3+ messages in thread
* Re: client released lease on dn which does not exist
2010-07-28 11:27 ` Sébastien Paolacci
@ 2010-07-28 16:37 ` Sage Weil
0 siblings, 0 replies; 3+ messages in thread
From: Sage Weil @ 2010-07-28 16:37 UTC (permalink / raw)
To: Sébastien Paolacci; +Cc: ceph-devel
[-- Attachment #1: Type: TEXT/PLAIN, Size: 2848 bytes --]
On Wed, 28 Jul 2010, Sébastien Paolacci wrote:
> FYI, the same test pass without any warning now (unstable from yesterday).
Yeah, sorry I missed this email. I fixed a few different lease related
problems on Monday. There may still be (rare) cases where that warning
comes up which need some additional scrutity, so if you do see it again,
please let me know.
Thanks!
sage
>
> Sebastien
>
> 2010/7/17 Sébastien Paolacci <sebastien.paolacci@gmail.com>:
> > Hello all,
> >
> > I'm getting some warnings that I've not been able to find informations
> > for when running dbench on a newly setup box (nettop, atom dual core
> > 2.5" very low-end hdd). Seems however to be easily reproducible
> > (unstable server and kernel client from yesterday) :
> >
> > srv:# mkcephfs -c ceph.conf --allhosts --mkbtrfs
> > srv:# for srv in cosd cmon cmds; do $srv -c ceph.conf -i 0; done
> >
> > client:# mount -t ceph srv:/ /mnt/
> > client:# dbench -D /mnt/ 1
> >
> > Few seconds later:
> > srv:# ceph -w
> > 10.07.17_18:40:12.797372 pg v7: 264 pgs: 264 active+clean+degraded;
> > 69 KB data, 508 KB used, 450 GB / 450 GB avail; 10/20 degraded
> > (50.000%)
> > 10.07.17_18:40:12.800218 mds e4: 1/1/1 up {0=up:active}
> > 10.07.17_18:40:12.800303 osd e3: 1 osds: 1 up, 1 in
> > 10.07.17_18:40:12.800448 log 10.07.17_18:39:56.072957 mon0
> > 192.168.0.3:6789/0 4 : [INF] mds0 192.168.0.3:6802/6754 up:active
> > 10.07.17_18:40:12.800598 mon e1: 1 mons at 192.168.0.3:6789/0
> > 10.07.17_18:40:14.234588 pg v8: 264 pgs: 264 active+clean+degraded;
> > 1166 KB data, 1280 KB used, 450 GB / 450 GB avail; 28/56 degraded
> > (50.000%)
> > 10.07.17_18:40:18.159014 log 10.07.17_18:40:12.067141 mds0
> > 192.168.0.3:6802/6754 1 : [WRN] client4099 released lease on dn
> > 1000000000c/RESULTS3.PRN which dne
> > 10.07.17_18:40:19.199284 pg v9: 264 pgs: 264 active+clean+degraded;
> > 2103 KB data, 2536 KB used, 450 GB / 450 GB avail; 33/66 degraded
> > (50.000%)
> >
> > Does anyone know how "severe" the warning is and what could I
> > undertake to get rid of it?
> >
> > Thanks,
> > Sebastien
> >
> > ps: I can (separately) provide the 90MB mds log file should it be needed.
> >
> > [global]
> > [mon]
> > mon data = /data/mon$id
> > [mon0]
> > host = srv
> > mon addr = 192.168.0.3:6789
> > [mds]
> > debug ms = 1
> > debug mds = 20
> > [mds0]
> > host = srv
> > [osd]
> > osd data = /data/osd$id
> > osd journal = /dev/sda3
> > [osd0]
> > host = srv
> > btrfs devs = /dev/sda4
> >
> --
> 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
>
>
^ permalink raw reply [flat|nested] 3+ messages in thread
end of thread, other threads:[~2010-07-28 16:35 UTC | newest]
Thread overview: 3+ messages (download: mbox.gz / follow: Atom feed)
-- links below jump to the message on this page --
2010-07-17 17:07 client released lease on dn which does not exist Sébastien Paolacci
2010-07-28 11:27 ` Sébastien Paolacci
2010-07-28 16:37 ` Sage Weil
This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox;
as well as URLs for NNTP newsgroup(s).