General information
Middleware
UMD
About the new Linux distribution
- CERN and FNAL decided to jointly support AlmaLinux : https://linux.web.cern.ch/#fermilabcern-recommendation-for-linux-distribution
- check also the presentation at the January GDB
- on Jan16 the Linux Future Committee met to discuss the adoption of Almalinux: https://indico.cern.ch/event/1229565/
- presentation about Almalinux on Jan31st https://indico.cern.ch/event/1242311/
- middleware will be released for Almalinux 9 (no more for CS9)
- CentOS7 is the recommended OS until UMD5 is released
- CentOS7 supported until Jun 2024
- recommended path is C7→AL9
- survey about Linux future circulated by WLCG to site-admins
- results presented during February GDB
About UMD
- collecting the products available on EL9 to prepare the first UMD5 release
Operations
- New NGI_IE Operation Centre: https://ggus.eu/index.php?mode=ticket_info&ticket_id=159433
- contributing with one site in the EGI FedCloud
ARGO/SAM
- request to enable the notification of failures for the APEL metrics:
- the APEL metrics are included in the ARGO_MON_OPERATORS
- notifications will be enabled for all of the metrics in this profile instead of ARGO_MON_CRITICAL
- a new report associated to this profile needs to be created in ARGO
FedCloud
- EGI Check-in migration from MitreID to Keycloak:
- two cases are pursued individually, with minimal or no impact
- full tickets list
Feedback from DMSU
New Known Error Database (KEDB)
The KEDB has been moved to Jira+Confluence: https://confluence.egi.eu/display/EGIKEDB/EGI+Federation+KEDB+Home
- problems are tracked with Jira tickets to better follow-up their evoulution
- problems can be registered by DMSU staff and EGI Operations team
Monthly Availability/Reliability
Under-performed sites in the past A/R reports with issues not yet fixed:
- AsiaPacific: https://ggus.eu/index.php?mode=ticket_info&ticket_id=159223
- Taiwan-LCG2: SRM decommissioned, but not yet disabled on GOCDB
- NGI_DE: https://ggus.eu/index.php?mode=ticket_info&ticket_id=159116
- MPPMU: webdav configuration issues; a new endpoint has been setup.
- UNI-FREIBURG: webdav and SRM configuration issues have been fixed, A/R figures are improving.
- NGI_GRNET: https://ggus.eu/index.php?mode=ticket_info&ticket_id=158231
- GR-07-UOI-HEPLAB: SRM failures have been fixed
- NGI_UK: https://ggus.eu/index.php?mode=ticket_info&ticket_id=156742
- RAL-LCG2: authz issues with the webdav endpoint
- they provide an object store endpoint, and the commands executed by the webdav probe don't work (GGUS 157748). The webdav probe is not maintained any more, anyway we are discussing a possible modification.
- new version of the probe to be released and tested
- the "ls" check can be disabled by setting a proper information in GOCDB
- they provide an object store endpoint, and the commands executed by the webdav probe don't work (GGUS 157748). The webdav probe is not maintained any more, anyway we are discussing a possible modification.
- RAL-LCG2: authz issues with the webdav endpoint
Under-performed sites after 3 consecutive months, under-performed NGIs, QoS violations: (Jan 2023):
- AsiaPacific: https://ggus.eu/index.php?mode=ticket_info&ticket_id=160323
- Australia-T2
- TW-NCUHEP
- NGI_BG: https://ggus.eu/index.php?mode=ticket_info&ticket_id=160327
- BG05-SUGrid
- NGI_DE: https://ggus.eu/index.php?mode=ticket_info&ticket_id=160331
- FZJ: expired certificates on some dCache servers; fixed.
- NGI_HR: https://ggus.eu/index.php?mode=ticket_info&ticket_id=160330
- egee.irb.hr
- NGI_IBERGRID: https://ggus.eu/index.php?mode=ticket_info&ticket_id=160329
- CIEMAT-LCG2: in November there were job failures due to authz issues with ops VO; between Dec and Jan problem with the air conditioning in the computing facility, preventive shutdown of the cluster.
- NGI_IT: https://ggus.eu/index.php?mode=ticket_info&ticket_id=160332
- RECAS-NAPOLI
sites suspended:
Documentation
- MediaWiki in read-only mode
- content to be moved to different locations (confluence and https://docs.egi.eu/)
- confluence space hosting policies and procedures: EGI Policies and Procedures
- EGI Federation Operations
- Change Management, Release and Deployment Management, Incident and Service Request Management, Problem Management, Information Security Management
- Manuals, How-Tos, Troubleshooting, FAQs:
- huge number of material need to be reviewed and in case updated when moved to the new place
- location will be https://docs.egi.eu/providers/operations-manuals/
- Guidelines for providers to join EGI: https://docs.egi.eu/providers/joining/
- Tutorial on submitting HTC jobs: https://docs.egi.eu/users/tutorials/htc-job-submission/
IPv6 readiness plans
- please provide updates to the IPv6 assessment (ongoing) https://wiki.egi.eu/w/index.php?title=IPV6_Assessment
- if any relevant, information will be summarised at OMB
Transition from X509 to federated identities (AARC profile token)
- In Feb 2022 OSG fully moved to token-based AAI, abandoning X509 certificates
- HTCondorCE: replacement of Grid Community Toolkit
- The long-term support series (9.0.x) from the CHTC repositories will support X509/VOMS authentication until May 2023
- Starting in 9.3.0 (released in October 2021), the HTCondor feature releases does NOT contain this support
- EGI sites are recommended to stay with the long-term support series for the time being
Migration of the VOs from VOMS to Check-in
- transition period where both X509 and tokens can be used
- delays in updating the GRID elements to the latest version compliant with tokens
- not all of the middleware products can be compliant with tokens at the same time
- the same VO has to interact with element supporting different authentications
Testing HTCondorCE and AARC Profile token
- INFN-T1 did some tests with the AARC Profile token using its HTCondorCE endpoints
- Configuring authentication on HTCondorCE
- HTCondor CE token configuration tips (by INFN-T1)
- dteam VO registered in Check-in/Comanage:
- Entitlements:
- urn:mace:egi.eu:group:dteam:role=member#aai.egi.eu
- urn:mace:egi.eu:group:dteam:role=vm_operator#aai.egi.eu
- Entitlements:
- The HTCondorCE expects to find in the token the scope claim to authorise the jobs submission
- in that moment Check-in didn't release this claim: it does since the migration to Keykloak technology replacing MitreID
WLCG Campaign
- WLCG started the CE Token support campaign
- https://twiki.cern.ch/twiki/bin/view/LCG/CEtokenSupportCampaign
- sites should upgrade the CEs to the version supporting also tokens, and configure tokens
Hackathon events
- 15th - 16th September ARC/HTCondor CE Hackathon, organised by WLCG, with HTCondorCE and ARC-CE to mostly investigating data staging issues (see GDB introduction)
- agreed to enable the support of the several token profiles through plugins
- same plugin for the several CEs
- plugins provided by the "creators" of the token profiles
- CE teams to provide specifics to the AAI teams and to release a new CE version supporting the plugins
- agreed to enable the support of the several token profiles through plugins
Plans for the coming months:
- ARC-CE and HTCondorCE implemented a new API interface
- The Check-in team is working on a plugin for the Check-in/AARC token profile allowing the authz on the CE side
- according to the AARC guidelines, the claim to authorise the job submission is provided through a different attributes than the one used by the WLCG token
- the plugin will translate the attribute to be understandable to the CEs
- The plugin will be tested in production before its release in UMD
- Then we can start the decommission procedure for the HTCondorCE long-term support series (9.0.x)
- ideally before May 2023
- At the same time the VOs using voms will be cloned to Check-in in order to be ready to use the tokens when the first HTCondor 10.0.x endpoints are in productions (some of the VOs might be involved in testing tokens with the 9.0.x version)
DPM Decommission and migration
- DPM supported until June 2023
- Sites are encouraged to start the migration to a different storage element since the process will take time
- choosing the new storage solution depends on the expertise/experience of the sites and on the needs of the supported VOs
- See the slides presented by Petr Vokac at the EGI Conference 2022 about the migration tools to dCache
- DPM provides a migration script to dCache (migration guide)
- Transparent migration
- Migrate just catalog (database) and keep files untouched
- both SE store files on posix filesystem
- Transparent migration
- Migration in three steps
- verify the DPM data consistency
- no downtime needed
- the operation can last several days or some weeks
- DPM dump and dCache import
- downtime lasting about 1 day
- verify the DPM data consistency
- In September opened tickets to the sites to plan the migration and decommission:
- tickets list (9 out of 55 were solved)
- Please let us know your plans for DPM EOL and in case you decide to use dCache migration tools the tickets will be used to support you on this storage migration method.
- dCache migration should be done by June 2023
Planned completion date | |
By 2022 | 1 |
Beginning of 2023 | 5 |
By Feb 2023 | 5 |
By Q1 2023 | 8 |
By Q2 2023 | 11 |
undefined | 12 |
Chosen technology | |
Migration to dCache | 28 |
Migration to EOS | 6 |
Migration to xroot/ceph | 3 |
Migration to XrootD | 1 |
Migration to Xcache | 1 |
Migration to Dynafed | 2 |
Not yet decided/no clear plan | 11 |
Decommissioning SE or site | 3 |
New benchmark replacing HEP-SPEC06
The benchmark HEPSCORE is going to replace the old Hep-Spec06
Main points agreed:
- Existing resources at the sites will not be re-benchmarked with HEPscore (unless the site has modern resources and would like to re-benchmark them in order to get higher consumption in the accounting reports)
- New resources purchased by the site will be benchmarked with HEPscore
- HEPscore will be normalized with HS06 with factor 1
- The switch to HEPscore in the accounting reports will happen on the 1st of April 2023 (when WLCG switch yearly pledges)
- This implies that two benchmarks will co-exist on the infrastructure for quite some time
- We would like to follow the progress regarding amount of the resources benchmarked with HEPscore
- No need for reporting of measurements for two benchmarks in parallel for the same set of resources
- This implies that accounting record should contain one metric for a single benchmark and benchmark name has to be properly defined in the accounting record.
Plans for the coming months:
- A new APEL version compliant with the new benchmark will be released
- Planned some tests in particular with sites sending normalised reports
- Change of units: in the Accounting Portal all of the references to HepSpec06 will be replaced with the new benchmark
- this will be implemented during last week of March
- Need to distinguish between the resources benchmarked with one benchmark or the other to monitor the move to the new benchmark over the time
- The accounting records will be tagged with the name of the benchmark used for normalisation
- A new APEL version is going to introduce this feature
- The Accounting Portal will allow to see the split of the resources by benchmark
- This change can be introduced a couple of months after April
HEPSCORE application:
- link to the gitlab page: https://gitlab.cern.ch/hep-benchmarks/hep-score
AOB
Next meeting
March