Hi Graeme,
I've just upgraded to DPM-rfio-server-1.3.8-2sec_sl3 and
DPM-gridftp-server-1.3.8-2sec_sl3 and so far the response seems ok:
epdt82$ time dpns-ls -l /dpm/ph.bham.ac.uk/home/dteam/generated/2005-10-18
-rw-rw-r-- 1 15007 2688 37 Oct 18 00:51 file8976ef4f-bf6d-41eb-95e4-ebc1a87a3106
0.48s real 0.07s user 0.07s system
epdt82$
The load on my SE is also ok:
[root@epgse1 sysconfig]# uptime
16:43:40 up 10 days, 21:58, 2 users, load average: 0.03, 0.04, 0.04
Yves
On Tue, 18 Oct 2005, Greig A Cowan wrote:
> Hi Graeme,
>
> I upgraded to 1.3.8 yesterday and have been seeing the same error messages
> in our DPM (log at end), although I haven't seem the large time delays
> that you are experiencing.
>
> [gcowan@ui gcowan]$ time dpns-ls -l
> /dpm/epcc.ed.ac.uk/home/dteam/greig_test_dir/20051017_133023.txt
> -rw-rw-r-- 1 18119 2688 17371 Oct 17 13:30
> /dpm/epcc.ed.ac.uk/home/dteam/greig_test_dir/20051017_133023.txt
>
> real 0m0.310s
> user 0m0.040s
> sys 0m0.000s
>
> I am still having issues with my migration of our Classic SE files into
> our DPM (converting the classic SE into a pool node). As an update to my
> email from yesterday, it now appears that although the migration created a
> whole set of /dpm/epcc.ed.ac.uk/home/dteam/generated/YYYY-MM-DD
> directories and populated them with files, they don't match to the files
> in the corresponding classic SE directories (the names are different and
> not all of the YYYY-MM-DD dirs are present in the DPNS). I'm confused :|
>
> Cheers,
> Greig
>
> 10/18 16:13:05 25673,0 Cns_srv_lstat: NS092 - lstat request by
> /C=UK/O=eScience/OU=Edinburgh/L=NeSC/CN=greig cowan (18119,2688) from
> ui.epcc.ed.ac.uk
> 10/18 16:13:05 25673,0 Cns_srv_lstat: NS098 - lstat 0
> /dpm/epcc.ed.ac.uk/home/dteam/generated
> 10/18 16:13:05 25673,0 Cns_srv_lstat: returns 0
> 10/18 16:13:05 25673,0 Cns_srv_opendir: NS092 - opendir request by
> /C=UK/O=eScience/OU=Edinburgh/L=NeSC/CN=greig cowan (18119,2688) from
> ui.epcc.ed.ac.uk
> 10/18 16:13:05 25673,0 Cns_srv_opendir: NS098 - opendir
> /dpm/epcc.ed.ac.uk/home/dteam/generated
> 10/18 16:13:05 25673,0 Cns_srv_opendir: returns 0
> 10/18 16:13:05 25673,0 Cns_srv_readdir: NS092 - readdir request by
> /C=UK/O=eScience/OU=Edinburgh/L=NeSC/CN=greig cowan (18119,2688) from
> ui.epcc.ed.ac.uk
> 10/18 16:13:05 25673,0 Cns_srv_readdir: returns 0
> 10/18 16:13:05 25673,0 Cns_srv_readdir: NS092 - readdir request by
> /C=UK/O=eScience/OU=Edinburgh/L=NeSC/CN=greig cowan (18119,2688) from
> ui.epcc.ed.ac.uk
> 10/18 16:13:05 25673,0 Cns_srv_readdir: returns 0
>
>
>
> On Tue, 18 Oct 2005, Graeme Stewart wrote:
>
> > Hi Folks
> >
> > I upgraded to the new releases of dpm-rfio and dpm-gridftp-server (1.3.8-2)
> > today. This has fixed the rfio logging problem previously reported, however
> > dpns queries have become terribly slow:
> >
> > [graeme@ui1-gla graeme]$ time dpns-ls -l /dpm/scotgrid.ac.uk/home/dteam/generated/2005-10-18
> > -rw-rw-r-- 1 dteam004 dteam 2908624 Oct 18 15:45 fileffdba218-8fb8-4baa-bc48-54a4fe39a136
> >
> > real 0m30.444s
> > user 0m0.080s
> > sys 0m0.000s
> >
> > Borne out by the dpns logs:
> >
> > 10/18 15:56:18 3581,0 Cns_srv_lstat: NS092 - lstat request by /C=UK/O=eScience/OU=Glasgow/L=Compserv/CN=graeme stewart (1454,2010) from ui1-gla.scotgrid.ac.uk
> > 10/18 15:56:18 3581,0 Cns_srv_lstat: NS098 - lstat 0 /dpm/scotgrid.ac.uk/home/dteam/generated/2005-10-18
> > 10/18 15:56:18 3581,0 Cns_srv_lstat: returns 0
> > 10/18 15:56:28 3581,0 Cns_srv_opendir: NS092 - opendir request by /C=UK/O=eScience/OU=Glasgow/L=Compserv/CN=graeme stewart (1454,2010) from ui1-gla.scotgrid.ac.uk
> > 10/18 15:56:28 3581,0 Cns_srv_opendir: NS098 - opendir /dpm/scotgrid.ac.uk/home/dteam/generated/2005-10-18
> > 10/18 15:56:28 3581,0 Cns_srv_opendir: returns 0
> > 10/18 15:56:33 3581,0 Cns_srv_readdir: NS092 - readdir request by /C=UK/O=eScience/OU=Glasgow/L=Compserv/CN=graeme stewart (1454,2010) from ui1-gla.scotgrid.ac.uk
> > 10/18 15:56:33 3581,0 Cns_srv_readdir: returns 0
> > 10/18 15:56:38 3581,0 Cns_srv_readdir: NS092 - readdir request by /C=UK/O=eScience/OU=Glasgow/L=Compserv/CN=graeme stewart (1454,2010) from ui1-gla.scotgrid.ac.uk
> > 10/18 15:56:38 3581,0 Cns_srv_readdir: returns 0
> >
> > Every part of the DPNS query is taking 5 or 10 seconds. These used to be "instant", e.g.,
> >
> > 10/10 11:30:14 28997,0 Cns_srv_lstat: NS092 - lstat request by /C=UK/O=eScience/OU=Glasgow/L=Compserv/CN=graeme stewart (1453,2010) from ui1-gla.scotgrid.ac.uk
> > 10/10 11:30:14 28997,0 Cns_srv_lstat: NS098 - lstat 0 /dpm/scotgrid.ac.uk/home/dteam/
> > 10/10 11:30:14 28997,0 Cns_srv_lstat: returns 0
> > 10/10 11:30:14 28997,0 Cns_srv_opendir: NS092 - opendir request by /C=UK/O=eScience/OU=Glasgow/L=Compserv/CN=graeme stewart (1453,2010) fromui1-gla.scotgrid.ac.uk
> > 10/10 11:30:14 28997,0 Cns_srv_opendir: NS098 - opendir /dpm/scotgrid.ac.uk/home/dteam/
> > 10/10 11:30:14 28997,0 Cns_srv_opendir: returns 0
> > 10/10 11:30:14 28997,0 Cns_srv_readdir: NS092 - readdir request by /C=UK/O=eScience/OU=Glasgow/L=Compserv/CN=graeme stewart (1453,2010) fromui1-gla.scotgrid.ac.uk
> > 10/10 11:30:14 28997,0 Cns_srv_readdir: returns 0
> > 10/10 11:30:14 28997,0 Cns_srv_readdir: NS092 - readdir request by /C=UK/O=eScience/OU=Glasgow/L=Compserv/CN=graeme stewart (1453,2010) fromui1-gla.scotgrid.ac.uk
> > 10/10 11:30:14 28997,0 Cns_srv_readdir: returns 0
> >
> > I have restarted all the daemons, in case any were in a fankle, but has
> > anyone else see the same?
> >
> > Graeme
> >
> >
>
>
|