September 16, 20169 yr I am using 2 SE3016 silicon graphics SAS expanders, running 6.2 and have been having lots of problems - I recently upgraded but these problems existed for the most part before i upgraded, just didn't happen very often. Now i am experiencing them more often. I will just list them all and perhaps be advised to have separate threads? 1. Drive expander is occasionally becoming unreadable and sometimes dropping all drives in it. (this has happened 3x in the last 2 days) - 1 time with 6.1.9 and 2 since i upgraded to 6.2) 2. Unraid creates a packet storm or broadcast storm when unraid crashes - this has been happening once in a while for a long time. I have to reboot at least once to get this to go away. In the meantime no wired computers on the network can get online or see eachother. It seems to go away if unraid boots and everything is working good. If there is still a problem (i think with disk IO) it can still continue to kill the network. - also i am only using 1 NIC on my unraid machine, no bonding etc. 3. My unraid install is marking drives as faulty after a few reboots, often hard reboots as unraid is crashed and i can't do anything else without network or functioning keyboard. It has only done this 2x and the drives are showing fine in smart short smart tests as well as no obvious problems in smart attributes. 4. If the unraid machine is crashed can i pull the mem stick and copy the logs etc before rebooting? i see no reference to this in the before you post faq, so assuming you can't. At this point i have been swapping out parts, trying to figure out what has caused the expander failure or crashes and in that process got another drive flagged as faulty due i think to the hard shut downs/resets. I am triggering crashes by disk intensive tasks, like the mover, a cleardisk and possibly a mover last night at 2:30 am.. I am not sure. Since i have 2 of these 16 drive expanders i pulled all the drives and put them in 1 expander, plus 4 and a cache plugged directly into the motherboard sata ports. I had a problem during move, so switched them to the other enclosure, but can't really do much because it marked my 5tb as faulty. I am out of spare drives - the first rebuild went fine after a 4tb drive was marked as faulty went fine but said it had 300 errors (which i was told is a bug) - this was done on 6.1.9 a few days ago. Again i think there may possibly be nothing wrong with this drive. I used my hot spare 6tb on this. Now i have a 5tb and 4tb that are marked faulty. Perhaps i can just pull them and do preclears with a separate box and trial unraid to test them? I expect 2 fresh toshiba x300 6tbs to arrive hopefully in approximately a week. They have been delayed by amazon shipping for a while but i'm told they will arrive by the 23rd or sooner. I have some logs and diagnostics and will attach them. Often i couldn't get any diagnostics because of lockups and dead network. - i am first attaching syslogs which are just copied from web log interface. Then the following posts will have some diagnostics, of which i think 1 i was able to capture in a crashing state. Suggestions please and any corrections on my assumptions are appreciated since i don't know that much. Edit - at this point i am just preclearing the drives that were marked failed on a trial install of unraid 6.2 - I shut down the main unraid server as i am really worried about losing another drive and not having parity to rebuild, much less a spare drive. syslogs.zip
Archived
This topic is now archived and is closed to further replies.