ASGC T1 Report ============== Oct 23 * wcp007 and wcp006 crash around Oct 22 20:42:27 * reboot at Mon Oct 23 01:30:23 UTC 2006 Oct 21 * ct error found at ASGC CE, at 2006-10-20 18:51:23 * prob arise from generic timeout error, prob resolved next iter sft executed * Two new disk servers added CMS report the transfer quality drop a little bit when two disk server added: * quote: Since they were added, the transfer quality of PHEDEX starts dropping a little bit The most errors are related to "timed out", transfers are longer than 8000 seconds Oct 20 * wcp006 reboot at 2:30 utc, due to the highload * ct error found at ASGC , as well as other sites using ASGC's top bdii * prob only found at quanta.grid.sinica.edu.tw * sft ct error detect at: 2006-10-20 09:39:56 and 2006-10-20 10:03:08 * the errors are referring to generic timeout prob in Replica Management tests using lcg-utils Oct 17 * JL/CT error occured at the same time as the LHCOPN maintence. This seems to have caused SAM failures. * [LHCOPN] Maintenance on 18/10/2006: routers' configuration changes Oct 16 * Oracle streams server increased to 4M window size result increase from 4MB to 40MB performance.