1 / 8

Mario Reale IGI / GARR GARR - mario.reale@garr.it EGEE-III to EGI Transition Meeting,

Coordination of Network Support (O-E-12 ) EGI International Task : What do all NGIs need to do ? Initial proposal and questions around task implementation. Mario Reale IGI / GARR GARR - mario.reale@garr.it EGEE-III to EGI Transition Meeting, March 2, 2010 - Amsterdam.

ebonner
Download Presentation

Mario Reale IGI / GARR GARR - mario.reale@garr.it EGEE-III to EGI Transition Meeting,

An Image/Link below is provided (as is) to download presentation Download Policy: Content on the Website is provided to you AS IS for your information and personal use and may not be sold / licensed / shared on other websites without getting consent from its author. Content is provided to you AS IS for your information and personal use only. Download presentation by click this link. While downloading, if for some reason you are not able to download a presentation, the publisher may have deleted the file from their server. During download, if you can't get a presentation, the file might be deleted by the publisher.

E N D

Presentation Transcript


  1. Coordination of Network Support (O-E-12 )EGI International Task:What do all NGIs need to do ? Initial proposal and questions around task implementation Mario Reale IGI / GARR GARR - mario.reale@garr.it EGEE-III to EGI Transition Meeting, March 2, 2010 - Amsterdam

  2. Goals of O-E-12 • The main goal of O-E-12 is to coordinate the support for the overall interfacing between the EGI infrastructure and the underlying network layer w.r.t.: • Link and Network faults • Scheduled downtimes and maintenance • Network Monitoring for the Grid infrastructure • Network Performance Issues • The network support coordination activity will be managed by IGI / GARR • The corresponding manpower is 0.5FTE • EGI network support will require a commitment by the NGIs to contribute to joint, organized activities aimed at supporting EGI on network related issues • Contact points and contributing members should be identified within each NGI SA2 – Transition toward EGI – X. Jeannin - 2 Feb. 2010 2

  3. In particular, we consider specifically relevant for O-E-12: 1)  The notification to Grid site administrators and Usersabout planned/scheduled network outages due to maintenance or interventions on the links/routers 2)  For a specificsubset of all possible e2e paths (jointly identified by many actors: VRCs/VOs, NGIs, EGI.eu..), the set up of a multi-domain monitoring systemcarefully reporting about the network performances and able to provide useful information both to ROCs(NGIs operations) and  NOCs (NRENs) 3)  The provisioning of an EGI  global network troubleshooting tool for usage both by ROCs(NGIs) and NOCs (NRENs) 4)  A central monitoring server to report about the downtimesof all the publicly advertised and available Grid services (registered in the GOC-DB) over the EGI infrastructure. Networking Issues around O-E-12

  4. O-E-12 Initial Strategy The 3 main network-related services we intend to set up initially for EGI are PerfSONAR_Lite-TSS General on-demand troubleshooting tool being release by SA2 ENOC DownCollector Central server testing Grid TCP ports on all Grid nodes registered in GOCDB every 2 minutes Keep using GGUS to collect and store planned network interventions related tickets Depending on the amount of collaboration from the NGIs we will consider the introduction of new tools and monitoring systems PerfSONAR multi-domain monitoring and visualization tools NAGIOS Zabbix Other tools NGIs will propose or are interested in

  5. PerfSONAR-Lite TSS Network troubleshooting tool Launches tests on demand from the Grid site under central server control: ping, traceroute, DNS lookup, nmap and bandwidth measurements Network Coordination Team, NGI orsite administrator 1 2 6 3 Network Coordination 5 4 administrator Grid site B Grid site A Local site light PerfSONAR’s sensor Central monitoring server PerfSONAR-Lite TSS — X. Jeannin — TNLC February 2010 5

  6. DownCollector • Currently run by ENOC at • https://ccenoc.in2p3.fr/DownCollector/ 6

  7. 1)How should overall new possible  specific requirements from users communities or specific application networking requirements be transferred to NRENs and  GEANT ? 2) If your NREN is provided with a PERT Team ( Team of experts on solving network performance issues): How familiar is your NGI operation team with the PERT management team provided by your NREN ? 3)Which are the network monitoring tools your NREN is mostly familiar with ? Which possible new ones would it be interested in ? 4)In particular, how familiar are you with the PerfSONAR multi-domain monitoring and visualization tools ? 5)Would you possibly be interested in an approach to network monitoring based on selected used and VOs submitting Grid jobs aimed at providing network monitoring information on selected paths / links ?  (thus requiring no  monitoring software to be installed at the site, rather only accepting a set of specific Grid jobs) Questions for all NGIs

  8. A few final commentson what all NGIs have to do Network tickets/issues are normally occurring very rarely A few ones per year Nevertheless, clearly all of us (NGIs) should Discuss and agree on a minimal set of basic useful things to be pursued Tools and workflow for network troubleshooting How to implement network monitoring for specific, relevant e2e path Which are the centralized tools / services to be set up Commit to contribute to an overall EGI network support Focusing on improving the overall reliability and predictability of the EGI infrastructure Identify our NGI contacts towards network related issue and the central Network Support coordination at GARR Rely on the existing NRENs and GEANT tools and teams to provide network support Within GEANT a permanent support/coordination liaison for EGI should be set up

More Related