mejutty

If I already have a ibm 1015 that was already flashed with the IT firmware do i need to go through the whole flash process ie ge the address earace flash etc etc or can I just flash the new it firmware??

I seem to have my array in a state where I can hit the so called shingle wall. I have 3 of the Seagate 8tb drives 2 data with about 700gb free with the other parity. I also have 2 8tb reds one a second parity the other a replacement for a 3tb green. when mover kicks in and starts writing the 40gb files I have siting there or my 120gb disk image backups as soon as about 20gb of data has been written the array grinds I even see read performance issues on other disks. stats show single digit to no reads or writes for periods of time.and I get video playback issues.

rsync -av --progress --remove-source-files /mnt/diskX/ /mnt/diskY Can this command be used to add files to a disk that already has data on it Ie diskx has a folder call movies with files and disky has a folder called movies with files will it merge or overwrite??

I am looking to do the transfer but want to check something. I have a 4tb drive that I just put into the array, it was my parity disk that was replaced with an 8tb drive. As I have other 8tb drives in the array with about 2.5tb on them and more free space than an empty 4tb drive the 4tb drive has no data on it at all. Is it just a simple stop array change file system bring back online then format to have the same empty disk change the filesystem it uses while keeping parity intact?? Then I can do the data moves between the disks??

I have just done direct path io passthrough of the whole usb controller rather than adding a usb device and the issue has gone. Web gui very fast no reset errors. Only downside is plop will not boot had to use PlopKexec livecd now just trying to work out how to have it autoboot rather than needing to select to boot unraid. Another thing to note is now the dashboard shows cpu usage as being 1-2% not the 99-100% it was before even tho top didn't show any cpu usage.

Add me to the list of people with the issue. I did a fresh install of 6.0.1 last night on ESX6 with 2 hds and getting the really slow web gui and the reset errors. I just haven't been able to start up yet as I am preclearing 2 8TB drives and I don't see a way to start without any assigned drives.

Thank you very much just the answer I was looking for.

I am about to do a test build of an esxi system with 3 pass through M1015 on a X9SCM-F board with a E-1230V2 which is supposed to not work. I suspect that the new firmware may and I stress may have fixed the issue. I have got myself some 20 odd 2tb sata drives to play with for the test but I don't have the time to preclear them. Basically what I want to do is add 20 to 22 of the disks to my build add them to an array without a cache drive and there will be no data as it is a test. Once I add the drives prove that it works copy some data up and down then add another drive as cache and let it start to do a parity creation to stress reading and writing over all the disks. Now because this is only a test and the information of the drives is irrelevant is there a way to write the preclear signature with enough of the system erased without having to do a full preclear on all 20 odd drives?? It appears as tho I can skip pre and post reads and zero the mbr but this option says nothing else so I don't know if that means it does nothing else or will write a signature with only the mbr being zeroed. Any help appreciated. Found plenty of posts saying your data will be unprotected by adding a non zeroed drive but I have no data and don't care it is a fresh no data test build.

I do it the old fashioned way download the vmdk to my pc mount change the files then upload. If you are going to mount the boot volume why not just share it out as say vmflash then update as you would the usual flash share?? That being said I do all my work from a windows machine as as little as possible at a Unix prompt

I use the screws provided by Norco for this reason. Since they don't give you enough for the cages and the drive trays, you have to buy more: link. $10 for a bag of screws is pretty ridiculous, but they are the only ones I've found that are proven to work. I notice in you blog that you used the antec 1200 v3 thumb screws for the for the norco cages how did you manage this

I found that I had issue mounting the Norco SS-500 in my Antec 1200 case, apart from the removal of the tabs with a dremel I found that using normal screws to hold the Norco SS-500 inplace resulted in the screws going to far into the cage itself and stopping the caddies/drives going in and out. I replaced my fans with Enermax Magma Batwing Blade I'm still not 100% happy with the temp the drives get up to but I can't hear any of the fans running in the case so I may look at getting some fans that may not be on the quiet side to see if I can get some extra cooling. Might even try the original fans since I never did try them.

One of the two drives in that top area is my cache drive, so it's spun down during the parity check. Also, in the 5.25" opening I inserted a mesh grill thingy from another case: My temps in that area during a parity check are about the same (or lower) than in the regular 4 bays. well I'll be dammed. Did exaclty what you did put a mesh blanking plate in there and bang temps on all drives identical. Got 36degrees while doing a drive rebuild. Much better temps than with the lid off.

I am currenlty running with the lid off due to high temps when doing parity checks. With the lid on tems for the top 2 drives get to around 47c with the lid off around 41c and again only while doing parity checks/drive rebuilds. I am currenlty working to place 2 small 4cm fans just behind the front cd slot so I can run with the lid on. I have 6 3tb drives in my box with the top 2 drives being Hitachi 5400 rpm drives.

I have 2 3tb Hitachi coolspin drives in there and they run at 42-44 while doing a parity check.

Yes I enabled write cache in the bios. I had 4 3tb Hitachi drives all preclearing at over 100mb/sec at the same time when they started out just seems to be when writing through unraid. Not sure have another box will do some testing. I did post my syslog in the gerneral foumr to have looked at as I noted the 250Gb drive that came with the unit which I still have in there is only running at UDMA100 as opposed to UDMA133 but I don't know if that will affect all the drives esp with the cache drive off.

I already have the russian hack in place (I think). Before the russian hack I noted during preclears that the upper 2 disks appeared to share BW ie they were doing the full read at about 50mb/sec each where as the others were doing it at 100mb/sec each, aftert he hack all drives were listing preclear read speeds of 100Mb/sec when stating out. I also noted that the preclear writes were at about 80mb/sec it's only when writing to the drives through unraid that I can only get up to 30mb sec transfers. I have a second N36l I am going to flash and patch up to preclear some other disks and do some testing to see if I can work out if it is an issue or the way it is. I have the nexus double twin but I also found another brand that I can't find now that did the same thing.

I have almost finished my build and I noticed that the array without a parity drive still only got 30mb writes, without a cache drive. I am waiting for parity to finish it's initial build but do you think it should have been higher without parity?? I also note that my parity is only building at 28-29mb/sec which seems to show something may be limiting write speeds. BTW here is a neater way to stuff 2 drives into the top of an N36l. Will post pics of a fan mod to keep them cool. They currenlty run at around 40-42 with case on doing preclear/parity build.