SC WEEKLY REPORT

1) INTERVENTIONS ON CASTOR2

- Migration of lhcb from castor1 to castor2 (to be completed on july 07): two different disk pool have been configured, one with tape beckend and the other one with disk-only backend.
srm://castorsrm.cr.cnaf.infn.it will be used for the tape pool srm://sc.cr.cnaf.infn.it for the disk pool
- two additional disk servers made available to ATLAS, for a total of 3 disk servers available to the VO.

2) LFC AT CNAF
a new server has been installed in order to have a pool of two hosts to be used in load balancing mode. Configuration is still ongoing.

3) SCHEDULED DOWNTIME
Provable downtime next Mon affecting FTS and CASTOR2. If confirmed, it will be announced in due time.
---------------------------------------------------

4) SC4 DAILY LOG
07/03: SAM SRM FUNCTIONALITY PROBLEMS
Many SRM failures detected by SAM on Mon July 03, affecting both T1 and T2 sites at INFN. A number of tickets issued, of which just the one concerning Roma1 and Torino have been closed.

Tickets (on Italian system support):
According with: https://lcg-sam.cern.ch:8443/sam/sam.cgi?sensors=SRMŽions=Italy&vo=DTeam&order=RegionName&funct=ShowSensorTests

INFN-BARI pccms5.cmsfarm1.ba.infn.it  Permission denied
https://grid-it.cnaf.infn.it/checklist/modules/xhelp/ticket.php?id=1634

INFN-MILANO grid006.mi.infn.it Permission denied
https://grid-it.cnaf.infn.it/checklist/modules/xhelp/ticket.php?id=1635

INFN-ROMA1 grid-cert-03.roma1.infn.it Transport endpoint is not connected
https://grid-it.cnaf.infn.it/checklist/modules/xhelp/ticket.php?id=1636

INFN-ROMA1-CMS cmsrm-se01.roma1.infn.it Transport endpoint is not connected
https://grid-it.cnaf.infn.it/checklist/modules/xhelp/ticket.php?id=1637

INFN-T1 castorsrm.cr.cnaf.infn.it Timeout when executing test SRM-put after 240 seconds
No ticket (under investigation)

INFN-T1 sc.cr.cnaf.infn.it Timeout when executing test SRM-put after 240 seconds
No ticket (under investigation)

INFN-TORINO grid008.to.infn.it No information found for SE
No ticket (INFN-TORINO down per aggiornamento, vedi GOCDB)

07/04: NETWORK FAULT
GEANT network fault problem (fibre cut) on the Milan-Geneva path affecting the LHC OPEN connection between CERN and CNAF, problem fixed via traffic rerouting at 7 pm. LHC OPN path back to operation on 07/06.

07/05
* ATLAS: ATLAS importing from CERN at around 100 MB/S, thanks to the addition of two new disk servers at Castor2. AOD data transfer from CNAF to Milan and Naples started. High VO BOX load at CNAF (around 30).
* Functionality of DPM in Roma1 fixed.

07/06
Implementation of disk area for dteam (t1 --> t0 transfers) not subject to garbage collection ongoing.

07/07
Fixes to FTS configuration: (GGUS id# 9426):
FTA_LHCB_ACTIONS_RETRYPARAMS="MaxFailures = 3; HoldEnabled = false ; OverwriteFailedFiles = true ; OverwriteExistingFiles = false ; DefaultRetryDelay = 300 ; RetryDelayForTimeoutOnGet = 1800 ; RetryDelayForDestFileExists = 300 ;" FTA_LHCB_AGENT_RETRY_INTERVAL="30"

[end]