[Cado-nfs-discuss] Uncompleted polyselect phase 1

Christophe Clavier christophe.clavier at unilim.fr
Wed Jun 3 19:38:58 CEST 2015


Paul,

Thank you for your insights.
I succedeed in canceling the workunits.
When I restart the server it does not seem to assign new workunits 
(neither for phase 1 nor phase 2):

/.../
/Info:HTTP server: If you want to start additional clients, remember to 
add their hosts to server.whitelist//
//Info:Polynomial Selection (size optimized): Starting//
//Info:Polynomial Selection (size optimized): 700 polynomials in queue 
from previous run, worst lognorm 55.150000//
/
When I start a client, it waits for a workunit to proceed with:

/.../
/INFO:root:Running wget -V//
//INFO:root:Downloading 
https://opn.msi.unilim.fr:8120/cgi-bin/getwu?clientid=node41.c0212823 to 
download/WU.node41.c0212823 (cafile = download/server.a6da95a9.pem)//
//ERROR:root:Download failed, URL error: HTTP Error 404: No work available//
//ERROR:root:Waiting 10.0 seconds before retrying//
//ERROR:root:Download failed, URL error: HTTP Error 404: No work available//
//ERROR:root:Waiting 10.0 seconds before retrying (I have been waiting 
since 10.0 seconds)//
//ERROR:root:Download failed, URL error: HTTP Error 404: No work available//
//ERROR:root:Waiting 10.0 seconds before retrying (I have been waiting 
since 20.0 seconds)/

Any idea?

Christophe


Le 03/06/2015 13:38, paul zimmermann a écrit :
>         Christophe,
>
>> Date: Wed, 03 Jun 2015 13:24:55 +0200
>> From: Christophe Clavier <christophe.clavier at unilim.fr>
>>
>> Hello,
>>
>> During the first phase of the polynomial selection, 3 out of 100 range
>> have been given to clients that unfortunately happened to be killed.
>> Consequently, at the end of this phase the server is waiting for results
>> from these clients and I would prefer not to wait for the timeout to elapse.
>> Is there any means to restart the server so that it can give these
>> ranges to some new clients, or alternatively so that it starts the
>> second phase with the priority queue in its current state ?
>>
>> Regards
>>
>> Christophe
> the following should do it:
>
> $ wudb.py -dbfile cxxx.db -cancel <wuname>
>
> You can get the names of the 3 remaining work-units as follows (see README
> in scripts/cadofactor):
>
> $ wudb.py -dbfile cxxx.db -dump -assigned
>
> Please confirm it works,
> Paul

-------------- section suivante --------------
Une pi?ce jointe HTML a ?t? nettoy?e...
URL: <http://lists.gforge.inria.fr/pipermail/cado-nfs-discuss/attachments/20150603/f390d820/attachment.html>


More information about the Cado-nfs-discuss mailing list