Speakers:
Alessandra Forti
(University of Manchester (GB)),
Diego Davila Foyo
(Univ. of California San Diego (US)),
Petr Vokac
(Czech Technical University (CZ))
ATLAS
- Status: 51 / 77 (T0&T1&T2)
- all T0+T1 except RAL
- 40% data volume transferred with HTTP-TPC (15-20% with TAPE that can't be move yet)
- EOS ATLAS back with HTTP-TPC
- works fine except for one site GGUS:151000
- weird CentOS7 libcurl behavior for KEK CA
- NSS error -8054 (SEC_ERROR_REUSED_ISSUER_AND_SERIAL)
- GGUS:150896 StoRM site with Lustre backend and STOR-932
- does anybody knows if `du --bytes --summarize /path/to/storage` is usable?
- how long does it take to walk through 10M directories with 10M files using `du`?
- we expect updates at least every hour - it should be cheap
- did not we discussed configurable script for StoRM?
- Storage issued tokens validity vs. X509 proxy/voms expiration
- I guess it is OK to obtain (restricted) token which can be valid for longer time
- dCache - ignore proxy & voms expiration, max lifetime 168H (rejects request with longer validity)
- DPM, EOS, XRootD - ignore proxy & voms expiration, max lifetime 24H (requests with longer validity capped)
- StoRM use voms extension lifetime and ignore proxy expiration