On Wed, Jul 9, 2008 at 1:12 PM, Tim <tim at tcsac.net> wrote:> Perfect. Which means good ol'' supermicro would come through :) WOHOO! > > AOC-USAS-L8i > > http://www.supermicro.com/products/accessories/addon/AOC-USAS-L8i.cfmIs this card new? I''m not finding it at the usual places like Newegg, etc. It looks like the LSI SAS3081E-R, but probably at 1/2 the cost. -B -- Brandon High bhigh at freaks.com "The good is the enemy of the best." - Nietzsche
On Thu, Jul 10, 2008 at 10:34 AM, Brandon High <bhigh at freaks.com> wrote:> On Wed, Jul 9, 2008 at 1:12 PM, Tim <tim at tcsac.net> wrote: >> Perfect. Which means good ol'' supermicro would come through :) WOHOO! >> >> AOC-USAS-L8i >> >> http://www.supermicro.com/products/accessories/addon/AOC-USAS-L8i.cfm > > Is this card new? I''m not finding it at the usual places like Newegg, etc. > > It looks like the LSI SAS3081E-R, but probably at 1/2 the cost.It appears to be the non--RAID version of the card. (That''s what the R suffix indicates). If it is that is the case, then I''ve got one running quite happily in my workstation already, using the mpt driver. James C. McPherson -- Solaris kernel software engineer, system admin and troubleshooter http://www.jmcp.homeunix.com/blog http://blogs.sun.com/jmcp Find me on LinkedIn @ http://www.linkedin.com/in/jamescmcpherson
Dunno how old it is, but James is right, no Raid which is why it''s cheaper. Also why I like it ;) On Wed, Jul 9, 2008 at 7:34 PM, Brandon High <bhigh at freaks.com> wrote:> On Wed, Jul 9, 2008 at 1:12 PM, Tim <tim at tcsac.net> wrote: > > Perfect. Which means good ol'' supermicro would come through :) WOHOO! > > > > AOC-USAS-L8i > > > > http://www.supermicro.com/products/accessories/addon/AOC-USAS-L8i.cfm > > Is this card new? I''m not finding it at the usual places like Newegg, etc. > > It looks like the LSI SAS3081E-R, but probably at 1/2 the cost. > > -B > > -- > Brandon High bhigh at freaks.com > "The good is the enemy of the best." - Nietzsche >-------------- next part -------------- An HTML attachment was scrubbed... URL: <http://mail.opensolaris.org/pipermail/zfs-discuss/attachments/20080709/e4f8026c/attachment.html>
On Wed, Jul 09, 2008 at 07:53:30PM -0500, Tim wrote:> Dunno how old it is, but James is right, no Raid which is why it''s cheaper. > Also why I like it ;)I have the HP badged LSA SAS3080X in my Ultra80, it''s a fantastic card. If I ever get a box with PCI-E (I''m looking to upgrade the U80 soon, so it might just happen) that card looks like it would be *perfect*.> On Wed, Jul 9, 2008 at 7:34 PM, Brandon High <bhigh at freaks.com> wrote: > > > On Wed, Jul 9, 2008 at 1:12 PM, Tim <tim at tcsac.net> wrote: > > > Perfect. Which means good ol'' supermicro would come through :) WOHOO! > > > > > > AOC-USAS-L8i > > > > > > http://www.supermicro.com/products/accessories/addon/AOC-USAS-L8i.cfm > > > > Is this card new? I''m not finding it at the usual places like Newegg, etc. > > > > It looks like the LSI SAS3081E-R, but probably at 1/2 the cost. > > > > -B > > > > -- > > Brandon High bhigh at freaks.com > > "The good is the enemy of the best." - Nietzsche > >> _______________________________________________ > zfs-discuss mailing list > zfs-discuss at opensolaris.org > http://mail.opensolaris.org/mailman/listinfo/zfs-discuss
Here''s to hoping it works, I just pulled the trigger on this at provantage for my Home NAS in hopes to replace the PCI card I have in their now from Supermicro. I also picked up a couple of SAS to SATA cables for it too. If anyone has verified it works, please let me know. Thanks, ~Bryan This message posted from opensolaris.org
Bryan, Where did you find the sas to sata cables? I''ve been looking but haven''t found anything at the usual watering holes. I assume you grabbed mini-sas to 4 sata? Thanks! --tim On 7/12/08, Bryan Wagoner <bryan at wagz.tv> wrote:> Here''s to hoping it works, I just pulled the trigger on this at provantage > for my Home NAS in hopes to replace the PCI card I have in their now from > Supermicro. I also picked up a couple of SAS to SATA cables for it too. If > anyone has verified it works, please let me know. > > Thanks, > ~Bryan > > > This message posted from opensolaris.org > _______________________________________________ > zfs-discuss mailing list > zfs-discuss at opensolaris.org > http://mail.opensolaris.org/mailman/listinfo/zfs-discuss >
Tim wrote:> Bryan, > > Where did you find the sas to sata cables? I''ve been looking but > haven''t found anything at the usual watering holes. I assume you > grabbed mini-sas to 4 sata?Hi Tim, I did a quick search with google: http://www.google.com/search?q=sff8087+sff+cable and came up with quite a few hits immediately. Some that might be worth a look include http://www.specialty-cables.com/sata-sas-cables.html http://www.eaglebit.com/Multi_lane_Cable_Forward_Breakout_Cable_Infiniband_SATA_s/192.htm http://www.acmemicro.com/estore/merchant.ihtml?pid=3886&step=4 cheers, James C. McPherson -- Senior Kernel Software Engineer, Solaris Sun Microsystems http://blogs.sun.com/jmcp http://www.jmcp.homeunix.com/blog
I was a little confused on what to get, so I ended up buying this off the Provantage website where I''m getting the card. The card was like $123 and each of these cables was like $22. CBL-0118L-02 IPASS to 4 SATA Octopus Cable I picked up this cable based on a part number for a different SAS cable mentioned in the manual that I downloaded online for that card. This was the next part number in line that had an IPASS connector and 4 sata.There seems to be a lot of different types of mini-sas cables out there. Not entirely sure if it''s the right one. I''m hoping to get a nice performance increase as I am currently running my desktop diskless and booting vista from iscsi targets on ZFS. This message posted from opensolaris.org
Did you have success? What version of Solaris? OpenSolaris? etc? I''d want to use this card with the latest Solaris 10 (update 5?) The connector on the adapter itself is "IPASS" and the Supermicro part number for cables from the adapter to standard SATA drives is CBL-0118L-02 "IPASS to 4 SATA Cable, 23-cm Pb-free" - I confirmed that with the Supermicro guys. I suppose by now you''ve been able to confirm if your cables worked as well, I''m just hoping maybe you have news as to everything else... This message posted from opensolaris.org
On Sun, Jul 13, 2008 at 3:37 AM, Bryan Wagoner <bryan at wagz.tv> wrote:> I was a little confused on what to get, so I ended up buying this off the Provantage website where I''m getting the card. The card was like $123 and each of these cables was like $22. > > CBL-0118L-02 IPASS to 4 SATA Octopus CableAdaptec also sells a lot of different SAS/SATA cables. The cost is a little higher but there''s a lot of selection. http://www.adaptec.com/en-US/products/cables/cables/sas/ Any of these cables will work: I-MSASX4-SAS4X1-FO-0.5M R internal mini Serial Attached SCSI x4 (SFF-8087) to (4) x1 (SFF-8482) Serial Attached SCSI (controller based) fan-out cable with removable power dongles. I-MSASX4-4SATAX1 0.5M R internal mini Serial Attached SCSI x4 (SFF-8087) to (4) x1 Serial ATA (controller based) fan-out cable ACK-I-mSASx4-4SATAx1-SB-0.5m R internal Mini Serial Attached SCSI x4 (SFF-8087) to (4) x1 Serial ATA (controller based) fan-out cable with SFF-8448 sideband signals. -B -- Brandon High bhigh at freaks.com "The good is the enemy of the best." - Nietzsche
Did anybody ever get this card working? SuperMicro only have Windows and Linux drivers listed on their site. Do Sun''s generic drivers work with this card? This message posted from opensolaris.org
On Mon, Aug 4, 2008 at 8:02 AM, Ross <myxiplx at hotmail.com> wrote:> Did anybody ever get this card working? SuperMicro only have Windows and > Linux drivers listed on their site. Do Sun''s generic drivers work with this > card? > >Still waiting to buy a set. I''ve already got the supermicro marvell based cards, as well as a motherboard with 2xPCI-X 133mhz slots and no PCI-E. It''ll be a motherboard plus two of the cards when I do buy them. Haven''t really had the motivation or the cash to do so yet. I''ve been keeping my eye out for a board that supports the opteron 165 and the wider lane dual pci-E slots that isn''t stricly a *gaming* board. I''m starting to think the combination doesn''t exist. -------------- next part -------------- An HTML attachment was scrubbed... URL: <http://mail.opensolaris.org/pipermail/zfs-discuss/attachments/20080804/88385ca0/attachment.html>
On Mon, Aug 4, 2008 at 6:49 AM, Tim <tim at tcsac.net> wrote:> really had the motivation or the cash to do so yet. I''ve been keeping my > eye out for a board that supports the opteron 165 and the wider lane dual > pci-E slots that isn''t stricly a *gaming* board. I''m starting to think the > combination doesn''t exist.The AMD 790GX boards are starting to show up: http://www.newegg.com/Product/Product.aspx?Item=N82E16813128352 Dual 8x PCIe slots, integrated video and 6 AHCI SATA ports. -B -- Brandon High bhigh at freaks.com "The good is the enemy of the best." - Nietzsche
Thanks for the link. I''ll consider those, but it still means a new CPU, and it appears it does not support any of the opteron line-up. On Mon, Aug 4, 2008 at 3:58 PM, Brandon High <bhigh at freaks.com> wrote:> On Mon, Aug 4, 2008 at 6:49 AM, Tim <tim at tcsac.net> wrote: > > really had the motivation or the cash to do so yet. I''ve been keeping my > > eye out for a board that supports the opteron 165 and the wider lane dual > > pci-E slots that isn''t stricly a *gaming* board. I''m starting to think > the > > combination doesn''t exist. > > The AMD 790GX boards are starting to show up: > http://www.newegg.com/Product/Product.aspx?Item=N82E16813128352 > > Dual 8x PCIe slots, integrated video and 6 AHCI SATA ports. > > -B > > -- > Brandon High bhigh at freaks.com > "The good is the enemy of the best." - Nietzsche >-------------- next part -------------- An HTML attachment was scrubbed... URL: <http://mail.opensolaris.org/pipermail/zfs-discuss/attachments/20080804/a65c622c/attachment.html>
And I can certainly vouch for that series of chipsets... I have a 750a-sli chipset (the one below the 790) and the SATA ports (in AHCI mode) Just Work(tm) under nevada / opensolaris. I''m yet to give it a while on S10, mostly as I pretty much run nevada everywhere... As S10 does indeed have an AHCI driver, I''d expect it would work just fine there too. Oh - and the ports go like stink!* For what it''s worth, even with Nevada, you will need the newest NVidia Xorg drivers from nvidia''s website to get the video working properly, and will need to add in it''s PCI ID''s in /etc/driver_aliases (And, as yet, I''m unable to run compiz in a stable way - Tends to hard lock up the machine after about 5 minutes use...), a very new hdaudio driver (I needed a bodgied up one from the Beijing team to make it work) and last I checked, the nvidia ethernet did not work properly without assigning it a valid ethernet address... (The driver misreads the ethernet address and either delivers it backwards, or byte-swaps... I don''t remember exactly...) Oh - And just in case you forget, most boards I have seen use IDE mode for the controllers by default, which reeks. Expect less than 15 MB/s if reading and writing at the same time if you forget to change the controller mode to AHCI! For what it''s worth, the board I''m using is a giga-byte.. Manufacturer: Gigabyte Technology Co., Ltd. Product: M750SLI-DS4 Which also has the 6 X AHCI ports. It might seem like it''ll be a lot of hassle getting it working, but in the ZFS space, it works great pretty much out of the box (plus ethernet address change if the nvidia driver is still busted... ;) Cheers! Nathan. *Going like stink means going like a hairy goat - like lightning - like s*it off a shovel - like a zyrtec - fast. :) Brandon High wrote:> On Mon, Aug 4, 2008 at 6:49 AM, Tim <tim at tcsac.net> wrote: >> really had the motivation or the cash to do so yet. I''ve been keeping my >> eye out for a board that supports the opteron 165 and the wider lane dual >> pci-E slots that isn''t stricly a *gaming* board. I''m starting to think the >> combination doesn''t exist. > > The AMD 790GX boards are starting to show up: > http://www.newegg.com/Product/Product.aspx?Item=N82E16813128352 > > Dual 8x PCIe slots, integrated video and 6 AHCI SATA ports. > > -B >
On Mon, Aug 4, 2008 at 2:52 PM, Tim <tim at tcsac.net> wrote:> Thanks for the link. I''ll consider those, but it still means a new CPU, and > it appears it does not support any of the opteron line-up.It should support any AM2/AM2+ dual-core Opteron like the 1220, etc. as well as the quad-core stuff. Socket 939 has been phased out for 2-3 years now, it''s unlikely new motherboards will be available. -B -- Brandon High bhigh at freaks.com "The good is the enemy of the best." - Nietzsche
>>>>> "bh" == Brandon High <bhigh at freaks.com> writes: >>>>> "nk" == Nathan Kroenert <Nathan.Kroenert at Sun.COM> writes:nk> And I can certainly vouch for that series of chipsets... I nk> have a 750a-sli chipset (the one below the 790) um...what? 750a is an nVidia chip http://en.wikipedia.org/wiki/Comparison_of_Nvidia_chipsets#nForce_700 NVidia chips are *extremely* confusing because one chipset can have like four names (ex. NForce 630a, nForce4, MCP61P, GeForce 7050 could all refer to a single chip), and I can''t find all four names for each chip. They immitate Sun marketing for naming things. What''s the C compiler called these days? 790GX is an ATI chip http://en.wikipedia.org/wiki/AMD_700_chipset_series#790GX nk> and the SATA ports (in AHCI mode) Just Work(tm) under nevada / nk> opensolaris. The ATI SB600 has problems under Solaris: http://bugs.opensolaris.org/bugdatabase/view_bug.do?bug_id=6665032 The ATI SB600/700 is AHCI. The nVidia SATA, I _thought_, was not AHCI but rather nv_sata: http://src.opensolaris.org/source/xref/onnv/onnv-gate/usr/src/uts/common/io/sata/adapters/ so, maybe not to confuse AHCI ``mode'''' in Blue Screen of Setup with the ahci driver? Also, SLI is nVidia. The ATI equivalent, the northbridge Brandon is talking about, is CrossFire. In short, there are only two lines of AMD chipsets to choose from, and you''ve confused them. But the important thing is, you''re reporting opensolaris works well, no slowdowns or sporadic lockups, SATA and Ethernet, on the nVidia 750a? In that case, I will rush right out and buy one. Nothing that new has been reported to work yet. It looks like it''s the first nVidia chip to have the 2GHz HT bus. bh> It should support any AM2/AM2+ dual-core Opteron like the bh> 1220, etc. as well as the quad-core stuff. Are you inferring that based on the name/shape of the socket? I don''t think that''s a fair assumption. The boards I looked at, if you go to the taiwanese manufacturer''s web site, explicitly list the CPU''s they support, and for all the boards I looked at, it''s either phenom or opteron, not both---a strict divide between desktop and server. Also the server boards all need registered memory, and the desktops all need unregistered. There are also two problems with quad-core chips. One is the requirement for a split power plane. This seems to be stronger with Opterons. I don''t understand if it''s a work/notwork issue, a speed issue, or a power-saving issue. The other is requirement for workaround of the BA/B2 stepping TLB bug. and I can''t untangle whether it''s part of the above two requirements or something different, but many motherboards needed a BIOS update to boot with a Barcelona chip. Customers were told to install an older AMD chip, upgrade the BIOS, then install the new chip. I would not assume a chip will work based on socket name, because from reading the comments I''m certain that assumption is unsafe. I''d suggest going back to the motherboard manufacturer website and checking for explicit listing of the 4-digit CPU code. You will find a lot of posts w.r.t. the server boards at least saying ``i got a quad-core (Barcelona) opteron, and it just didn''t work.'''' The Tyan <nnnn>-E series is just like their old <nnnn> boards, but with the errata to support the new chips. -------------- next part -------------- A non-text attachment was scrubbed... Name: not available Type: application/pgp-signature Size: 304 bytes Desc: not available URL: <http://mail.opensolaris.org/pipermail/zfs-discuss/attachments/20080804/7aefb584/attachment.bin>
Miles Nordin wrote:> The other is requirement for workaround of the BA/B2 stepping TLB bug. > >This was fixed some months ago, and it should be hard to find the old B2 chips anymore (not many were made or sold). -- richard
>>>>> "re" == Richard Elling <Richard.Elling at Sun.COM> writes:re> This was fixed some months ago, and it should be hard to find re> the old B2 chips anymore (not many were made or sold). -- well, they all ended up on newegg. :) -------------- next part -------------- A non-text attachment was scrubbed... Name: not available Type: application/pgp-signature Size: 304 bytes Desc: not available URL: <http://mail.opensolaris.org/pipermail/zfs-discuss/attachments/20080805/56535887/attachment.bin>
Well, just an update I suppose since there are people waiting fora review on this card. I bought the card shortly after my last post and it has been sitting in the box for the last 3 weeks or so on my desk. The reason being is because the LSI IPASS to 4 sata cables I ordered along with the card are backordered and they are expected to ship them to me on the 11th to finish the order. So basically, I haven''t bothered doing anything with the card since I don''t have any SAS drives to test with. As soon as the cables come in I''ll let you guys know if anything comes up good or bad. Did anyone else give the card a shot yet? I''m running 2008.05 as my home NAS box and have been quite happy. I''m going to use some t5700 thin clients with it too since the Processor usage on the storage server is so low. This message posted from opensolaris.org
On Mon, Aug 4, 2008 at 6:03 PM, Miles Nordin <carton at ivy.net> wrote:> bh> It should support any AM2/AM2+ dual-core Opteron like the > bh> 1220, etc. as well as the quad-core stuff. > > Are you inferring that based on the name/shape of the socket? I don''t > think that''s a fair assumption.I''m basing it on experience, actually. The 165 was a popular for Socket 939 systems since it cost less than the Athlon at the same clock. AMD realized their mistake and now charges more for the Opteron at equal clocks across the board.> The boards I looked at, if you go to the taiwanese manufacturer''s web > site, explicitly list the CPU''s they support, and for all the boards I > looked at, it''s either phenom or opteron, not both---a strict divide > between desktop and server. Also the server boards all need > registered memory, and the desktops all need unregistered.That''s based more on the target market for the board. It''s mainly just marketing, though some manufacturers may not add the server CPUIDs to the desktop BIOS. Remember that the memory controller is in the CPU, so it really doesn''t matter what the board says. (In fact, the very first "desktop" Athlon 64 chips were socket 940 and required registered memory.) The current 1-way Opterons are just binned Athlons. If you look at the actual CPU specs (http://www.amd.com/us-en/assets/content_type/white_papers_and_tech_docs/23932.pdf) for the AM2 Opterons it reads: - 144-bit DDR2 SDRAM controller operating at up to 333 MHz - Supports up to four unbuffered DIMMs - ECC checking with double-bit detect and single-bit correct The Socket F chips (2xxx and 8xxx series) require registered memory.> The other is requirement for workaround of the BA/B2 stepping TLB bug.Any BIOS that can recognize an Phenom / 3rd-gen Opteron will implement this for the B2 stepping.> or something different, but many motherboards needed a BIOS update to > boot with a Barcelona chip. Customers were told to install an older > AMD chip, upgrade the BIOS, then install the new chip. I would notThe BIOS needs to know about the chip. The same thing happened on the Intel side when the 65nm Core 2 came out (E6xxx and Q6xxx), and again with the 45nm Core 2 (E8xxx and Q8xxx). -B -- Brandon High bhigh at freaks.com "The good is the enemy of the best." - Nietzsche
Better late than never now thatI have that cable debacle cleared up :) This card works great with 2008.05and I haven''t had any problems. And All I can say is Wow is it fast compared to my nvidia and other AOC card. This message posted from opensolaris.org
My Supermicro H8DA3-2''s onboard 1068E SAS chip isn''t recognized in OpenSolaris, and I''d like to keep this particular system "all Supermicro," so the L8i it is. I know there have been issues with Supermicro-branded 1068E controllers, so just wanted to verify that the stock mpt driver supports it. -- This message posted from opensolaris.org
I''d also like to know how easy it is to identify drives when you use this card? Is it easy to know which is which after you''ve had a few failure & swapped drives around? -- This message posted from opensolaris.org
On Sun, 23 Nov 2008 06:13:51 -0800 (PST) Ross <myxiplx at googlemail.com> wrote:> I''d also like to know how easy it is to identify drives when you use > this card? Is it easy to know which is which after you''ve had a few > failure & swapped drives around?Hi Ross, in general, it''s just as easy to identify drives attached to this card as it is to identify drives attached to any other card. The output from ''iostat -En'' will help, along with careful observation of the drive''s serial number and devid. You can find the devid (which, last I checked, is guaranteed to be unique) by looking for the "devid" property associated with your specific device. :name pci1000,3150 sd (block, de00000463) :compatible (de00000482TBL) | scsiclass,00.vATA.pST3320620AS.rD | | scsiclass,00.vATA.pST3320620AS | | scsa,00.bmpt | | scsiclass,00 | | scsiclass | :sas-mpt :pm-capable 0x1 :inquiry-vendor-id ST3320620AS :inquiry-revision-id 3.AAD :inquiry-device-type 0 :inquiry-product-id ST3320620AS :devid id1,sd at TATA_____ST3320620AS_________________________________________3QF0EAFP :class scsi :target 0 :lun 0 :device-blksize 0x200 :fm-ereport-capable :ddi-kernel-ioctl :ddi-failfast-supported :pm-hardware-state needs-suspend-resume :pm-components (de0000046eTBL) | NAME=spindle-motor | | 0=off | | 1=on | :lba-access-ok :inquiry-serial-no 3QF0EAFP :devfs-path /pci at 0,0/pci10de,376 at a/pci1000,3150 at 0/sd at 0,0 :driver-name sd :binding-name scsiclass,00 :bus-addr 0,0 :instance 3 :_class block :name sd hth, James -- Senior Kernel Software Engineer, Solaris Sun Microsystems http://blogs.sun.com/jmcp http://www.jmcp.homeunix.com/blog
On Sun, Nov 23, 2008 at 4:55 PM, James C. McPherson <James.McPherson at sun.com> wrote:> On Sun, 23 Nov 2008 06:13:51 -0800 (PST) > Ross <myxiplx at googlemail.com> wrote: > > > I''d also like to know how easy it is to identify drives when you use > > this card? Is it easy to know which is which after you''ve had a few > > failure & swapped drives around? > > Hi Ross, > in general, it''s just as easy to identify drives attached to > this card as it is to identify drives attached to any other > card. > > The output from ''iostat -En'' will help, along with careful > observation of the drive''s serial number and devid. You can > find the devid (which, last I checked, is guaranteed to be > unique) by looking for the "devid" property associated with > your specific device. > > > :name pci1000,3150 > sd (block, de00000463) > :compatible (de00000482TBL) > | scsiclass,00.vATA.pST3320620AS.rD | > | scsiclass,00.vATA.pST3320620AS | > | scsa,00.bmpt | > | scsiclass,00 | > | scsiclass | > :sas-mpt > :pm-capable 0x1 > :inquiry-vendor-id ST3320620AS > :inquiry-revision-id 3.AAD > :inquiry-device-type 0 > :inquiry-product-id ST3320620AS > :devid > > id1,sd at TATA_____ST3320620AS_________________________________________3QF0EAFP > :class scsi > :target 0 > :lun 0 > :device-blksize 0x200 > :fm-ereport-capable > :ddi-kernel-ioctl > :ddi-failfast-supported > :pm-hardware-state needs-suspend-resume > :pm-components (de0000046eTBL) > | NAME=spindle-motor | > | 0=off | > | 1=on | > :lba-access-ok > :inquiry-serial-no 3QF0EAFP > :devfs-path /pci at 0,0/pci10de,376 at a > /pci1000,3150 at 0/sd at 0,0 > :driver-name sd > :binding-name scsiclass,00 > :bus-addr 0,0 > :instance 3 > :_class block > :name sd > > > > hth, > James > -- >I don''t know that that necessarily makes it *EASY* to find the drive, especially if they''re in a hot-swap bay. Something like an "led_on" type command would be helpful. Whether that be through sending IO to the drive to light up an activity LED, or through other means. --Tim -------------- next part -------------- An HTML attachment was scrubbed... URL: <http://mail.opensolaris.org/pipermail/zfs-discuss/attachments/20081123/8753431c/attachment.html>
Yeah, it''s not really ''easy'', but certainly better than nothing. I would imagine it would be possible to write a script that could link all three drive identifiers, and from there it would relatively simple to create a chart adding the physical location too. I agree with Tim that we really need a way to identify drives, even a command to manually flash the fault LED would be good. However, I know that on Sun hardware, there''s a blue LED that identifies drives as soon as you "cfgadm -c unconfigure" them, so I doubt it''s going to be a priority for Sun to add this for other brands of hardware. The way I see it, the best way to manage third party SAS drives it to have a table with the following information. I don''t have a SAS controller yet though to test out the idea, does this look sensible? ZFS Drive cfgadm serial physical location (col-row) c0t0d0 sata0/1 3QF0EAFP 1-1 I''m thinking that''s an easy enough way to manage a server, and printed serial number stickers on the hot swap bays would be a good way to double check that the correct drive is about to be removed. In fact, if I created a text file linking serial numbers to physical locations, would it be possible to write one script that output all of that? Then identifying a drive would just be a case of grepping that scripts output for whatever information you know. -- This message posted from opensolaris.org
On Sat, 22 Nov 2008 10:42:51 -0800 (PST) Asa Durkee <asadurkee at gmail.com> wrote:> My Supermicro H8DA3-2''s onboard 1068E SAS chip isn''t recognized in > OpenSolaris, and I''d like to keep this particular system "all > Supermicro," so the L8i it is. I know there have been issues with > Supermicro-branded 1068E controllers, so just wanted to verify that > the stock mpt driver supports it.(a) yes, the stock mpt driver does support it (b) what issues are you referring to? In my 2008.11rc1b vbox installation, mpt was installed by default: root at osolvbox:~# pkg search /kernel/drv/mpt INDEX ACTION VALUE PACKAGE path file kernel/drv/mpt pkg:/SUNWckr at 0.5.11-0.101 Likewise with SXCE (from build 1), and root at osolvbox:~# pkg contents -r pkg:/SUNWckr at 0.5.11-0.86 |grep mpt kernel/drv/amd64/mpt kernel/drv/mpt kernel/drv/mpt.conf So I suspect your system has a vendor and device ID which isn''t included by default in /etc/driver_aliases. Try firing up a terminal before you run the installer, and run # prtconf |grep "value.*pci1000" that should show you what device aliases your card is showing up with, and you can then run a command similar to the following: # update_drv -a -i '' "pci1000,3150" '' mpt (which is valid for my system, and note the quotes). hth, James C. McPherson -- Senior Kernel Software Engineer, Solaris Sun Microsystems http://blogs.sun.com/jmcp http://www.jmcp.homeunix.com/blog
On Sun, 23 Nov 2008 17:02:57 -0600 Tim <tim at tcsac.net> wrote:> On Sun, Nov 23, 2008 at 4:55 PM, James C. McPherson > <James.McPherson at sun.com > > wrote: > > > On Sun, 23 Nov 2008 06:13:51 -0800 (PST) > > Ross <myxiplx at googlemail.com> wrote: > > > > > I''d also like to know how easy it is to identify drives when you > > > use this card? Is it easy to know which is which after you''ve > > > had a few failure & swapped drives around? > > > > Hi Ross, > > in general, it''s just as easy to identify drives attached to > > this card as it is to identify drives attached to any other > > card. > > > > The output from ''iostat -En'' will help, along with careful > > observation of the drive''s serial number and devid. You can > > find the devid (which, last I checked, is guaranteed to be > > unique) by looking for the "devid" property associated with > > your specific device....> I don''t know that that necessarily makes it *EASY* to find the drive, > especially if they''re in a hot-swap bay. Something like an "led_on" > type command would be helpful. Whether that be through sending IO to > the drive to light up an activity LED, or through other means.Right, so you need to have more information available to you from the specific hardware that you''ve got installed. We''ve got sestopo ($SRC/cmd/scsi/sestopo) which makes use of libscsi and libses to handle jbods that Sun makes, but for generic hardware it''s more difficult. As far as I can tell, there''s no non-SCSI Enclosure Services way of turning on an LED on a drive unless you want to muck around with IPMI - see ipmitool(1m). cheers, James -- Senior Kernel Software Engineer, Solaris Sun Microsystems http://blogs.sun.com/jmcp http://www.jmcp.homeunix.com/blog
Does anyone know if this card will work in a standard pci express slot? -- This message posted from opensolaris.org
On Thu, Feb 12, 2009 at 01:20, Brent <brent at xenorg.com> wrote:> Does anyone know if this card will work in a standard pci express slot?According to a post on hardforum.com, it does not work. Instead, I''d recommend the LSI SAS3442E-R (if you don''t mind that it only has 4 internal ports). It''s available on eBay for $80 right now. Will
No, I don''t believe so, I found out that it''s a port unique to Supermicro boards, and you only get one port per board, which pretty much rules this card out. However, there''s a PCI-e card from LSI using the same chipset. LSISAS3081, or something like that. A quick search for LSISAS and Solaris should bring up a few cards that will work. -- This message posted from opensolaris.org
On Wed, Feb 11, 2009 at 8:10 PM, Will Murnane <will.murnane at gmail.com>wrote:> On Thu, Feb 12, 2009 at 01:20, Brent <brent at xenorg.com> wrote: > > Does anyone know if this card will work in a standard pci express slot? > According to a post on hardforum.com, it does not work. Instead, I''d > recommend the LSI SAS3442E-R (if you don''t mind that it only has 4 > internal ports). It''s available on eBay for $80 right now. > > WillThere''s several people on this list who have already stated it does just that. As have Supermicro support. I''d write off the hardocp post (which you apparently have failed to link to) as user error. --Tim -------------- next part -------------- An HTML attachment was scrubbed... URL: <http://mail.opensolaris.org/pipermail/zfs-discuss/attachments/20090211/0b87df1d/attachment-0001.html>
Hmm... somebody needs to tell Supermicro''s sales staff then. I specifically didn''t buy their cards after they told me it wouldn''t work. -- This message posted from opensolaris.org
On Wed, Feb 11, 2009 at 8:46 PM, Ross <myxiplx at googlemail.com> wrote:> Hmm... somebody needs to tell Supermicro''s sales staff then. I > specifically didn''t buy their cards after they told me it wouldn''t work. >Looks like Brandon Wagoner was the one who got it working here. Guess we can see if he''ll enlighten us as to what sort of slot he stuck it in. Brandon? Did you stick the card into a "supermicro approved UIO slot" or just a standard PCI-E slot? --Tim -------------- next part -------------- An HTML attachment was scrubbed... URL: <http://mail.opensolaris.org/pipermail/zfs-discuss/attachments/20090211/5d0bad6b/attachment-0001.html>
Brent wrote:> Does anyone know if this card will work in a standard pci express slot?Yes. I have an AOC-USAS-L8i working in a regular PCI-E slot in my Tyan 2927 motherboard. The AOC-SAT2-MV8 also works in a regular PCI slot (although it is PCI-X card).
Dave wrote:> Brent wrote: >> Does anyone know if this card will work in a standard pci express slot? > > Yes. I have an AOC-USAS-L8i working in a regular PCI-E slot in my Tyan > 2927 motherboard. > > The AOC-SAT2-MV8 also works in a regular PCI slot (although it is PCI-X > card). >Please let the list know if you try the USASLP-L8i (the low profile version of this card). The USAS-L8i only fits in 3u rackmount chassis.
On Wed, Feb 11, 2009 at 6:10 PM, Will Murnane <will.murnane at gmail.com> wrote:> On Thu, Feb 12, 2009 at 01:20, Brent <brent at xenorg.com> wrote: >> Does anyone know if this card will work in a standard pci express slot? > According to a post on hardforum.com, it does not work. Instead, I''d > recommend the LSI SAS3442E-R (if you don''t mind that it only has 4 > internal ports). It''s available on eBay for $80 right now.There''s a post there from a guy using two of the AOC-USAS-L8i in his system here: http://hardforum.com/showthread.php?p=1033321345 As far as I know, ULI is the same as PCI-e, just "backwards". -B -- Brandon High : bhigh at freaks.com
On Thu, Feb 12, 2009 at 19:02, Brandon High <bhigh at freaks.com> wrote:> There''s a post there from a guy using two of the AOC-USAS-L8i in his > system here: > http://hardforum.com/showthread.php?p=1033321345Read again---he''s using the AOC-SAT2-MV8, which is PCI-X. That is known to work fine, even in PCI slots. He does ask in his post #7 if the AOC-USAS-L8i card works, but no response yet. Unfortunately, Hardforums have been having database problems (again, sigh) and have lost all posts made between the 5th and 11th of February. Thus, I can''t link to the post I saw about the card not working, and I don''t see it in Google cache or archive.org. Perhaps I''ll ask again and see if I can find the answer.> As far as I know, ULI is the same as PCI-e, just "backwards".I think you mean UIO. And compatible though it may be, it still leaves open the question of how to mount the card securely. Will
On Thu, Feb 12, 2009 at 1:22 PM, Will Murnane <will.murnane at gmail.com>wrote:> On Thu, Feb 12, 2009 at 19:02, Brandon High <bhigh at freaks.com> wrote: > > There''s a post there from a guy using two of the AOC-USAS-L8i in his > > system here: > > http://hardforum.com/showthread.php?p=1033321345 > Read again---he''s using the AOC-SAT2-MV8, which is PCI-X. That is > known to work fine, even in PCI slots. He does ask in his post #7 if > the AOC-USAS-L8i card works, but no response yet. > > Unfortunately, Hardforums have been having database problems (again, > sigh) and have lost all posts made between the 5th and 11th of > February. Thus, I can''t link to the post I saw about the card not > working, and I don''t see it in Google cache or archive.org. Perhaps > I''ll ask again and see if I can find the answer. > > > As far as I know, ULI is the same as PCI-e, just "backwards". > I think you mean UIO. And compatible though it may be, it still > leaves open the question of how to mount the card securely. > > Will >Are you selectively ignoring responses to this thread or something? Dave has already stated he *HAS IT WORKING TODAY*. Dave wrote: * Yes. I have an AOC-USAS-L8i working in a regular PCI-E slot in my Tyan 2927 motherboard.* The AOC-SAT2-MV8 also works in a regular PCI slot (although it is PCI-X card). -------------- next part -------------- An HTML attachment was scrubbed... URL: <http://mail.opensolaris.org/pipermail/zfs-discuss/attachments/20090212/1ab1f4ed/attachment.html>
On Thu, Feb 12, 2009 at 20:05, Tim <tim at tcsac.net> wrote:> Are you selectively ignoring responses to this thread or something? Dave > has already stated he *HAS IT WORKING TODAY*.No, I saw that post. However, I saw one unequivocal "it doesn''t work" earlier (even if I can''t show it to you), which implies to me that whether the card works or not in a particular setup is somewhat finicky. So here''s one datapoint:> Dave wrote: > > Yes. I have an AOC-USAS-L8i working in a regular PCI-E slot in my Tyan 2927 > motherboard.but the thread that Brandon linked to does not contain a datapoint. For what it''s worth, I think these are the only two datapoints I''ve seen; most threads about this card end up debating back and forth whether it will work, with nobody actually buying and testing the card. Will
Will Murnane wrote:> On Thu, Feb 12, 2009 at 20:05, Tim <tim at tcsac.net> wrote: >> Are you selectively ignoring responses to this thread or something? Dave >> has already stated he *HAS IT WORKING TODAY*. > No, I saw that post. However, I saw one unequivocal "it doesn''t work" > earlier (even if I can''t show it to you), which implies to me that > whether the card works or not in a particular setup is somewhat > finicky. So here''s one datapoint: >> Dave wrote: >> >> Yes. I have an AOC-USAS-L8i working in a regular PCI-E slot in my Tyan 2927 >> motherboard. > but the thread that Brandon linked to does not contain a datapoint. > > For what it''s worth, I think these are the only two datapoints I''ve > seen; most threads about this card end up debating back and forth > whether it will work, with nobody actually buying and testing the > card. >I can tell you that the USAS-L8i absolutely works fine with a Tyan 2927 in a Chenbro RM31616 3U rackmount chassis. In fact, I have two of the USAS-L8i in this chassis because I forgot that, unlike the 8-port AOC-SAT2-MV8, the USAS-L8i can support up to 122 drives. I have 8 drives connected to the first USAS-L8i. They are set up in a raidz-2 and I get 90-120MB/sec read and 60-75MB/sec write during my rsyncs from linux machines (this solaris box is only used to store backup data). I plan on removing the second USAS-L8i and connect all 16 drives to the first USAS-L8i when I need more storage capacity. I have no doubt that it will work as intended. I will report to the list otherwise. -- Dave
For what it''s worth, I know that at least one person is using a LSI SAS3081E card which I believe is based on exactly the same chipset: http://www.opensolaris.org/jive/message.jspa?messageID=186415 -- This message posted from opensolaris.org
Thanks for all the help guys. Based on the success reports, i''ll give it a shot in my intel s3210shlc board next week when the UIO card arrives. I''ll report back on the success or destruction that follows...now i just hope solaris 10 10/08, but it sounds like it should. Cheers, Brent -----Original Message----- From: Dave [mailto:dave-zfs at dubkat.com] Sent: Fri 2/13/2009 9:22 AM To: Will Murnane Cc: Tim; zfs-discuss at opensolaris.org; Brent Avery Subject: Re: [zfs-discuss] Supermicro AOC-USAS-L8i Will Murnane wrote:> On Thu, Feb 12, 2009 at 20:05, Tim <tim at tcsac.net> wrote: >> Are you selectively ignoring responses to this thread or something? Dave >> has already stated he *HAS IT WORKING TODAY*. > No, I saw that post. However, I saw one unequivocal "it doesn''t work" > earlier (even if I can''t show it to you), which implies to me that > whether the card works or not in a particular setup is somewhat > finicky. So here''s one datapoint: >> Dave wrote: >> >> Yes. I have an AOC-USAS-L8i working in a regular PCI-E slot in my Tyan 2927 >> motherboard. > but the thread that Brandon linked to does not contain a datapoint. > > For what it''s worth, I think these are the only two datapoints I''ve > seen; most threads about this card end up debating back and forth > whether it will work, with nobody actually buying and testing the > card. >I can tell you that the USAS-L8i absolutely works fine with a Tyan 2927 in a Chenbro RM31616 3U rackmount chassis. In fact, I have two of the USAS-L8i in this chassis because I forgot that, unlike the 8-port AOC-SAT2-MV8, the USAS-L8i can support up to 122 drives. I have 8 drives connected to the first USAS-L8i. They are set up in a raidz-2 and I get 90-120MB/sec read and 60-75MB/sec write during my rsyncs from linux machines (this solaris box is only used to store backup data). I plan on removing the second USAS-L8i and connect all 16 drives to the first USAS-L8i when I need more storage capacity. I have no doubt that it will work as intended. I will report to the list otherwise. -- Dave -------------- next part -------------- An HTML attachment was scrubbed... URL: <http://mail.opensolaris.org/pipermail/zfs-discuss/attachments/20090213/9e5aae8d/attachment.html>
Hey Tim, I''ve been happily using the AOC-USAS-L8i since we started talking about it a while ago. I have it stuck in a generic motherboard from ebay in a PCI-Express x16 slot since i wasn''t going to have a 3d card in my NAS device or anything. Using 8 sata drives across it''s two ports with mirrored vdevs in my pool. currently on 2008.11, haven''t done an image update in a bit. ~Bryan -- This message posted from opensolaris.org
hi I have a AOC-USAS-L8i working in both a Gigabyte GA-P35-DS3P and Gigabyte GA-EG45M-DS2H under OpenSolaris build 104+ (Nexenta Core 2.0 beta). the controller looks like this in lspci: 01:00.0 SCSI storage controller: LSI Logic / Symbios Logic SAS1068E PCI-Express Fusion-MPT SAS (rev 08) Subsystem: Super Micro Computer Inc Unknown device a380 Flags: bus master, fast devsel, latency 0, IRQ 15 I/O ports at a000 Memory at f1010000 (64-bit, non-prefetchable) Memory at f1000000 (64-bit, non-prefetchable) Capabilities: [50] Power Management version 2 Capabilities: [68] Express Endpoint IRQ 0 Capabilities: [98] Message Signalled Interrupts: Mask- 64bit+ Queue=0/0 Enable+ Capabilities: [b0] MSI-X: Enable- Mask- TabSize=1 with 8x 300GB (older 7200rpm disks from 2004) as raid-z1 on a 3.2Ghz Core Duo and 4GB RAM, this delivers following figures: 1. write performance (linear) ~98MBytes/s: root at marvin:/tank/storage# dd_rescue -b4M /dev/zero test Summary for /dev/zero -> test:0, errxfer: 0.0k, succxfer: 10420224.0k dd_rescue: (info): ipos: 10485760.0k, opos: 10485760.0k, xferd: 10485760.0k errs: 0, errxfer: 0.0k, succxfer: 10485760.0k +curr.rate: 106039kB/s, avg.rate: 104173kB/s, avg.load: 16.4% root at marvin:/tank/storage# dd_rescue -b4M /dev/zero test2 Summary for /dev/zero -> test2:, errxfer: 0.0k, succxfer: 4128768.0k dd_rescue: (info): ipos: 4194304.0k, opos: 4194304.0k, xferd: 4194304.0k errs: 0, errxfer: 0.0k, succxfer: 4194304.0k +curr.rate: 88486kB/s, avg.rate: 96142kB/s, avg.load: 14.1% 2. read performance (linear) ~290MBytes/s: Summary for test -> /dev/null: dd_rescue: (info): ipos: 10485760.0k, opos: 10485760.0k, xferd: 10485760.0k errs: 0, errxfer: 0.0k, succxfer: 10485760.0k +curr.rate: 0kB/s, avg.rate: 285824kB/s, avg.load: 40.4% Summary for test2 -> /dev/null: dd_rescue: (info): ipos: 4194304.0k, opos: 4194304.0k, xferd: 4194304.0k errs: 0, errxfer: 0.0k, succxfer: 4194304.0k +curr.rate: 0kB/s, avg.rate: 308484kB/s, avg.load: 39.8% regards nicola -- This message posted from opensolaris.org
On Fri, Feb 13, 2009 at 04:51, Nicola Fankhauser <nicola.fankhauser at variant.ch> wrote:> hi > > I have a AOC-USAS-L8i working in both a Gigabyte GA-P35-DS3P and Gigabyte GA-EG45M-DS2H under OpenSolaris build 104+ (Nexenta Core 2.0 beta).Very cool! It''s good to see people having success with this card. How does mounting the card work? Can one reverse the slot cover and screw it in like that, or is the card hanging free? Can you provide pictures of the card mounted in the case? Thanks! Will
> How does mounting the card work? Can one reverse the > slot cover and screw it in like that, or is the card hanging free?unfortunately, the cover does not fit in the case, so I fixed it with a tip of hot glue; the same I used to fix the intel gig-e pci-e card (which is a low-profile version). not optimal, I know, but it works. nicola -- This message posted from opensolaris.org
> I plan on removing the second USAS-L8i and connect > all 16 drives to the > first USAS-L8i when I need more storage capacity. I > have no doubt that > it will work as intended. I will report to the list > otherwise.I''m a little late to the party here. First, I''d like to thank those pioneers who came before me and found this board works fine. I have two of them on order for a server I''m putting together. I''m curious about this connecting all 16 drives bit. I read that the controller will handle up to 122 drives, but short of a lot of extra, expensive hardware, I have no idea how that is accomplished. How are you intending to connect 16 drives to the two SAS ports on the controller? On a slightly different but related topic, anyone have advice on how to connect up my drives? I''ve got room for 20 pool drives in the case. I''ll have two AOC-USAS-L8i cards along with cables to connect 16 SATA2 drives. The motherboard has 6 SATA2 connectors plus 2 SATA3 connectors. I was planning to use the SATA3 connectors for the boot drives (hopefully mirrored ZFS). Initially I''ll have three 2TB drives with the intention of using raidz1 on them. Phase 2 will be three more 2TB drives. Phase 3 will be three 1.5TB drives. (I already have most of the drives for phase 2 and 3.) I''ll probably fill the rest with a JBOD pool of assorted drives I have lying around. I''m specifically wondering if I''ll see better performance by spreading the raidz1 array devices across the controllers or if I should keep an array to a single controller as much as possible. -- This message posted from opensolaris.org
> On a slightly different but related topic, anyone have advice on how > to connect up my drives? I''ve got room for 20 pool drives in the case. > I''ll have two AOC-USAS-L8i cards along with cables to connect 16 SATA2 > drives. The motherboard has 6 SATA2 connectors plus 2 SATA3 > connectors. I was planning to use the SATA3 connectors for the boot > drives (hopefully mirrored ZFS). Initially I''ll have three 2TB drives > with the intention of using raidz1 on them. Phase 2 will be three more > 2TB drives. Phase 3 will be three 1.5TB drives. (I already have most > of the drives for phase 2 and 3.) I''ll probably fill the rest with a > JBOD pool of assorted drives I have lying around.I''d try to grab more drives of the same size to be able to do RAIDz2. It''s quite common that a drive fails and there are ''silent'' errors (non-detectable on the drive''s CRC, but detectable by ZFS) during resilver. This will give you data corruption. If you have RAIDz2, ZFS will figure that out> I''m specifically wondering if I''ll see better performance by spreading > the raidz1 array devices across the controllers or if I should keep an > array to a single controller as much as possible.This has been debated a lot. If you put a VDEV (or pool) on a single controller, if that controller dies, your pool will be unavailable until the controller is replaced. If you spread the disks over more controllers, the chances are better to allow access to the data until you replace the controller. Again, raidz2 will help out there as well Vennlige hilsener / Best regards roy -- Roy Sigurd Karlsbakk (+47) 97542685 roy at karlsbakk.net http://blogg.karlsbakk.net/ -- I all pedagogikk er det essensielt at pensum presenteres intelligibelt. Det er et element?rt imperativ for alle pedagoger ? unng? eksessiv anvendelse av idiomer med fremmed opprinnelse. I de fleste tilfeller eksisterer adekvate og relevante synonymer p? norsk.