Prototype Data Access Center (PDAC) meetings take place every other Thursday at 9:00 Project Time on the BlueJeans infrastructure-meeting channel: https://bluejeans.com/383721668
Date 02 Nov 2017
Attendees
Forms - Checkbox button group0_label Paul Domagala 3_value Fabio Hernandez 2_selected true 5_label Kian-Tat Lim required false 1_selected false 4_selected false 7_selected false 2_label Igor Gaponenko 5_value Kian-Tat Lim 0_value Paul Domagala 4_label Simon Krughoff 7_value Xiuqin Wu labelLength 5% 7_label Xiuqin Wu 2_value Igor Gaponenko 6_label Fritz Mueller 4_value Simon Krughoff 1_label Gregory Dubois-Felsmann 3_selected false counter 8 label Usual attendees 6_selected false multiselect false 3_label Fabio Hernandez 6_value Fritz Mueller name Attendees 1_value Gregory Dubois-Felsmann 0_selected true 5_selected false
Goals Discussion items Key: AGENDA TOPICS, DISCUSSION , FOLLOWUP
Time Item Who Notes qserve master status Currently blocked on GPFS instability. Igor needs 48hrs of uninterrupted system uptime to load catalogs. The group has a milestone to load WISE data by end of Nov. May move to a strategy of keeping 2-3 copies of data across multiple nodes. Igor Gaponenko suggests investigating using native (RO) GPFS connections on PDAC.So far, system has met expectations (modulo GPFS issues). Local RAID is used for backups. Igor Gaponenko advised to go ahead and load the system to see if the problem recurs. Stress test.GPFS instability status The latest:
One of 2 GPFS servers is down. The GPFS client’s failed over automatically, and the admin's manually failed over the NFS this morning. IBM is at NCSA today figuring out problem. Request to increase limits and swap space for qserv pdac Unknown User (pdomagala) Ref. Jiraserver JIRA serverId 9da94fb6-5771-303d-a785-1b6c5ab0f2d2 key IHS-488
Since this is a system-wide change that effects all users. We'll be redirecting this to the DM chage control process.
Action items Attachments