Dell PowerEdge T610 server had a failed hdd (4 hdd raid 5 set) last week. called dell. ran dset diagnostics and sent to dell. rebuilt raid set by reseating failed hdd. dell said there were 2 hdd failures although only one had been reported via server manager and was visible with amber lights on. checked next day and rebuilt hdd had failed again. dell sent 2 replacement hdd's. rebuilt both hdd's and updated f/w, s/w, bios etc and all ok except the virtual disc was showing 'virtual disc bad blocks'. all physcial discs ok. system running ok. checked on system error log and there are occasional 'event id 51 hdd paging errors'. contacted dell again. ran online dell gui diagnostics but this just went around on a loop so aborted (with dells guidance) after @60 mins on hdd0 - same happened on hdd1. engineer said this should not happen and there must be a serious problem with the hdd's and recommended I run it on all hdd's overnight which I did not want to do. at this point dell said the only answer was to recreate the virtual disc and raid set. on the open server manager there is an option to 'clear virtual disc bad blocks'. when I mentioned this to the dell engineer he said I could try to run this but he did not sound too clued up on it. he said it could be run on a live system with no issues but did not sound 100% sure
Has anybody run a 'clear virtual disc bad blocks' task on a live server? Is there any system impact (i.e. slow running presumably whilst hdds are in action)
any other thoughts. I don't really want to have to do a complete server/raid rebuild as the system is running ok and there are only sporadic hdd warnings now on the system error log when the overnight backup job or av scan runs.
the only other oddity is the open server manager is reporting the raid f/w is below the minimum required level but it is not and dell acknowledge this.
any thoughts please?