 |
 |
Hi Raul, I wish we could and I asked for it, but it is not possible (yet?). The EGI portal displays monthly numbers or in any case cumulative over a month so it is not possible to make a daily comparison. Only month complete months comparison is reliable. cheers alessandra On 03/10/2016 18:32, RAUL H C LOPES wrote: > Hi again, > > Is there any way to check the records for one site of daily basis? > There is to be the that EGEE site. > > Thanks, raul > On 03/10/16 17:29, RAUL H C LOPES wrote: >> Hi Alessandra, >> >> I haven't changed anything at Brunel that would explain that >> difference. Two weeks ago (Sep 15), I corrected the HS06 in one the >> our clusters. However, the correction was minimal. It changed the >> numbers in in around 5%. It would explain that difference. >> >> BTW, I had noticed that Apel numbers for Brunel for August and >> September were unexpected. If you compare with July for example. And >> we did have a problem last year, when Apel simply declared >> lost one month of our records. >> >> Thanks, raul >> >> >> >> On 03/10/16 17:19, Alessandra Forti wrote: >>> PS what it is not clear to me mostly is how is it possible that a >>> site can move from green to red like RHUL and Brunel for example. >>> Did you change anything that may have caused this changes? >>> >>> On 03/10/2016 18:03, Alessandra Forti wrote: >>>> Dear all, >>>> >>>> the accounting TF has now a dashboard to compare results from the >>>> experiments accounting records with the APEL results. >>>> >>>> This contains all the experiments latest values which are updated >>>> each month with the numbers of the previous finished month. >>>> >>>> http://wlcg-mon-dev.cern.ch/dashboard/request.py/siteview#currentView=ACCOUNTING_VALIDATION&search_0=UKI-&sorting_Site+Name=asc >>>> >>>> >>>> There is also an historical view these are the UK sites from >>>> January until August >>>> >>>> http://tinyurl.com/gvuxue2 >>>> >>>> Most UK sites are ok, there are however 3 problematic sites which >>>> have RED more often than not UCL, RHUL and ECDF. Brunel was red in >>>> August and September too. >>>> >>>> If you click on each month there is a json file with the numbers >>>> for the month contained in the Status field the elements of the >>>> array of numbers are: HS06 on the atlas dashboard, HS06 in APEL, >>>> ratio, wallclock in ATLAS, wallclock in APEL, wallclock ratio. For >>>> example ECDF in August http://tinyurl.com/hq8tfvt >>>> >>>> Status": "123600,55882,45,11771,4299,37" >>>> >>>> APEL HS06 value 55882 is 45% the ATLAS dashb value 123600 >>>> The APEL wallclock 4299 is 37% of the ATLAS dashb value 11771 >>>> >>>> Sometimes the discrepancies are the other way, i.e. APEL much >>>> bigger than ATLAS. ATLAS gets the numbers only for the payload, >>>> while APEL gets the numbers from the batch system including all the >>>> pilot time, so in theory APEL should always be slightly bigger. >>>> Whatever the numbers they should be within 15-20% difference once >>>> way or the other. >>>> >>>> I've put these pages in the links page >>>> >>>> https://www.gridpp.ac.uk/wiki/Links_Monitoring_pages#Accounting >>>> >>>> Let me know if you have any comment. Can UCL, RHUL, ECDF and Brunel >>>> help me understand their discrepancies please? >>> >>>> >>>> cheers >>>> alessandra >>>> >>>> >>>> >>> >> > -- Respect is a rational process. \\// Fatti non foste a viver come bruti (Dante)