On 04/11/13 14:12, Brian Davies wrote:
> Try
> http://dashb-wlcg-transfers.cern.ch/ui/#dst.site=%28QMUL%29&fts.servers=%28lcgfts3.gridpp.rl.ac.uk%29&grouping.dst=%28country,site%29&grouping.src=%28country,site%29&m.content=%28errors,successes%29&p.grouping=src&tab=transfer_plots&technology=%28fts%29&vo=%28atlas%29
>
> detail is to select lcgfts3.gridpp.rl.ac.uk in the "technologies tab"
>
That's interesting - can it be listed on
https://www.gridpp.ac.uk/wiki/Links_Monitoring_pages
However it only lists transfers for which QMUL is the destination, and
not those for which QMUL is the source (it's also possible to list
transfers for which QMUL is the source, but that has to be on another
web page - and a QMUL centric plot would be better so I don't have to
look at quite so many pages).
Chris
> -----Original Message-----
> From: Testbed Support for GridPP member institutes [mailto:[log in to unmask]] On Behalf Of Christopher J. Walker
> Sent: 04 November 2013 14:07
> To: [log in to unmask]
> Subject: FTS3 monitoring
>
> I previously used to watch
> http://ganglia.gridpp.rl.ac.uk/cgi-bin/ganglia-fts/fts-page.pl?r=12+hours&g=UKILT2QMUL&v=All&f=&h=Services_Grid%2Flcgfts03.gridpp.rl.ac.uk&s=normal&.submit=Submit+Query
>
> to see that file transfers were going OK.
>
> With the switch to FTS3, and as a site admin, I'm looking for a similar "site centric" page - where I can see transfers in and out - what bandwidth I'm using, how many transfers are queued and whether things are failing.
>
>
> I can't see one - the two links on
> https://www.gridpp.ac.uk/wiki/Links_Monitoring_pages
>
> are to but a there one?
>
> The top one gives a text page - which you can filter to QMUL as source or QMUL as destination, but not both:
>
> http://www-ftsmon.gridpp.rl.ac.uk/fts3/ftsmon/#/?source_se=srm:%2F%2Fse03.esc.qmul.ac.uk&vo=&dest_se=
>
> The ganglia plots:
>
> http://ganglia.gridpp.rl.ac.uk/cgi-bin/ganglia-fts/fts3-sites.pl
>
> Have a similar problem - and also only seem to give an indication of transfers in progress, not the bandwidth used.
>
> Chris
|