October 5, 201411 yr Hi, I've searched but can't find an easy answer to my wrinkle on an old issue.... I've recently changed controller from a Highpoint to the above 3Ware.... My 4TB parity drive was correctly reported before, but now the GUI is reporting it as 2.2TB. It's also reported it as 1.8TB. The 3Ware BIOS reports the disc capacity as 4TB ok, but the UNRAID syslog reports it as; md: import disk0: [8,16] (sdb) 1AMCC_E3URAJ33000000000000 size: 1759534904 The system works fine, but it doesn't sit well when it wasn't a problem with the Highpoint..... BUT! It gets even stranger..... I added two more 4TB discs (same make/model)..... preclear went ok.... However UNRAID reports one as 1.8TB and the other as 2.2TB :'( 1AMCC_ECU2YAN7000000000000 (sdm) 1759534936 1AMCC_ENPN3JFS000000000000 (sdn) 2147483647 In summary; BIOS reports correct size Smartctl reports correct size Syslog and GUI reports incorrectly Any thoughts please? syslog-20141005-050029.txt
October 5, 201411 yr While I don't have one of those controllers, it does look like there is a firmware update for them to support >2.2 TB http://www.voodooalert.de/board/index.php?page=Thread&threadID=18431
October 5, 201411 yr Author Thanks for the link I am using the latest Firmware and drivers but it did make me wonder whether UNRAID is using the latest Linux drivers. Is there a way of checking which 3Ware drivers UNRAID is using? Regards, Paul
October 5, 201411 yr The only other thing I can think of is to make sure that the HBA is in IT mode and/or its set to JBOD. Its not passing through the serial numbers of the disks correctly. It looks like its passing through virtual disks (I could be wrong, but these don't look like serial numbers from the hard drive itself. Rather they look like virtual numbers from the HBA) 1AMCC_ECU2YAN7000000000000 (sdm) 1759534936 1AMCC_ENPN3JFS000000000000 (sdn) 2147483647
October 5, 201411 yr Author Hi Squid The card is set in JBOD mode. My understanding of what I've read so far is that the 3Ware reading of disc numbers is at the crux of the problem. The first eight characters after the 1AMCC_ are part of the serial number for that disc. e.g. the first one you quote, (sdm) is actually WD-WCCECU2YAN7 The bit in bold italics is the common bit. Why it comes up in the format it does is a mystery I'm trying to solve.... The long string of numbers is the capacity of the disc that UNRAID is reporting. In the case you quote, they are both identical WD 4TB red NAS drives. UNRAID reports one as 1.8TB and the second as 2.2TB........ Go figure......
October 7, 201411 yr Author As a further update..... I booted into windows on the UNRAID machine and looked in Device Manager and Disc Management. In Device Manager there are no issues at all. In Disc Management, it was very curious. All the discs are listed as Simple;Basic; Healthy and Online. The capacities of the 4TB discs are reported correctly as 3726.02 GB but they are split into two partitions. The healthy partition of 2048GB and the rest... 1678 GB as "Unallocated." Does this help anybody to help me please? Summary; 3Ware BIOS + Web 3DM2 ultility see the 4TB capacity in full Windows sees the full 4TB but split into 2.1 TB + 1.6 TB unallocated. UNRAID only sees the 2.1 TB partition Thanks for looking, Paul
October 7, 201411 yr The issue is your card or some setting on your card. You need an adapter that reports the drive as a single object of the full size.
October 8, 201411 yr When you boot, what does the bios report the firmware version of the card as being? The drive serial numbers isn't an issue. The card is probably treating each drive as a 1-drive RAID0 (Essentially JBOD). My Areca card lists the drives based on the port they are connected to...no part of the serial number is shown. I'm still waiting for parts to build a newer server around my 9650SE. I have not tested the 3Ware with large (>2TB) drives yet.
October 10, 201411 yr Author Hi, I seem to have sorted the size issue..... but I'm not sure how.... I was getting more and more random issues in booting in the first place, so I decided to wipe the entire USB and start again from scratch. Would you believe it, but the GUI now reports the correct disc capacity. Why that has sorted it??? Not a clue. What hasn't changed though is the temperature reporting and more importantly.... performance. Especially write performance. I'm getting around 5-6 Mb/s writing files. Nothing I've done seems to make any difference. I've tried adding a cache disc and got up to the dizzy heights of 6-7 Mb/s. I'm so disillusioned with it that I fired up my old system. An old sempron powered MB with my old Highpoint 16 port PCI-X 2240 mounted in a bog standard PCI slot. It fired up no problem. The GUI reports the temps fine. The most damning thing was file transfers. Writes at 30-40 Mb/s using the same files to the same locations as previously. What on earth is going on with the 3ware 9650?? My initial, middle of the night, completely pissed off suggestion, would be to avoid the 9650SE until you can talk to; 1. preferably somebody with the same card who has got it working fine. or 2. Somebody with a 3ware card who may have some setup/config tips. Sorry to sound so negative, but it's looking like I wasted $200 and more importantly, days of my time. Regards, Paul
October 10, 201411 yr The write speeds are almost indicative of a slow network connection. Are you connected at gigabit speed or only 100mbit? I'm waiting for some (impulse) ordered parts to arrive before I start building/testing my new machine.
October 10, 201411 yr Author Yup, gigabit.... Lightning quick, pings return at 0ms..... Besides if it was a network issue my old card would be slow too...... But it's not....I can watch the same Bluray off the server at 4 different places around the house without missing a beat.... I'm convinced it's either the card setup or a comparability issue.
October 12, 201411 yr Author Hi dgaschk, Thanks for taking an interest, hopefully this is what you are asking for. Regards, Paul Tower login: root Password: Linux 3.9.11p-unRAID. root@Tower:~# ifconfig eth0 Link encap:Ethernet HWaddr 44:8a:5b:b9:51:64 inet addr:192.168.0.14 Bcast:192.168.0.255 Mask:255.255.255.0 UP BROADCAST RUNNING MULTICAST MTU:1500 Metric:1 RX packets:100026476 errors:0 dropped:124 overruns:0 frame:0 TX packets:46484109 errors:0 dropped:0 overruns:0 carrier:0 collisions:0 txqueuelen:1000 RX bytes:3673536977 (3.4 GiB) TX bytes:2635980802 (2.4 GiB) Interrupt:42 Base address:0xc000 lo Link encap:Local Loopback inet addr:127.0.0.1 Mask:255.0.0.0 UP LOOPBACK RUNNING MTU:65536 Metric:1 RX packets:95 errors:0 dropped:0 overruns:0 frame:0 TX packets:95 errors:0 dropped:0 overruns:0 carrier:0 collisions:0 txqueuelen:0 RX bytes:7860 (7.6 KiB) TX bytes:7860 (7.6 KiB) root@Tower:~# ethtool eth0 Settings for eth0: Supported ports: [ TP ] Supported link modes: 10baseT/Half 10baseT/Full 100baseT/Half 100baseT/Full 1000baseT/Full Supports auto-negotiation: Yes Advertised link modes: 10baseT/Half 10baseT/Full 100baseT/Half 100baseT/Full 1000baseT/Full Advertised pause frame use: Symmetric Receive-only Advertised auto-negotiation: Yes Speed: 1000Mb/s Duplex: Full Port: Twisted Pair PHYAD: 0 Transceiver: internal Auto-negotiation: on MDI-X: Unknown Supports Wake-on: pumbg Wake-on: g Current message level: 0x00000033 (51) Link detected: yes root@Tower:~#
October 12, 201411 yr Author Thanks, Can you suggest any other diagnostics for me please? Regards, Paul
October 13, 201411 yr Configure the array with fewer drives using only the MB SATA ports and test the speed. Move those same drives to the card and test the speed. Select new config and the rebuild parity before each set of tests that involve changing the drive set.
October 21, 201411 yr I can help with troubleshooting now. I finally have my system up and running. 9650SE-24M8, 3 different 4TB drives so far. Precleared and parity checked fine. The snippet from my system log below shows the initial boot. Latest firmware is 4.10.00.027. The unRAID supplied driver might be a little bit dated. The LSI website indicated version 2.26.08.006 but they also indicate an "upstream" driver version 2.26.02.013 which is confusing. My current issue is that I can't get the drives to spin down. Performance is fine. My drives are labelled different from yours. Oct 17 04:33:43 Storage kernel: 3ware 9000 Storage Controller device driver for Linux v2.26.02.014. Oct 17 04:33:43 Storage kernel: scsi1 : 3ware 9000 Storage Controller Oct 17 04:33:43 Storage kernel: 3w-9xxx: scsi1: Found a 3ware 9000 Storage Controller at 0xf2020000, IRQ: 16. Oct 17 04:33:43 Storage kernel: 3w-9xxx: scsi1: Firmware FE9X 4.10.00.027, BIOS BE9X 4.08.00.004, Ports: 24. Oct 17 04:33:43 Storage kernel: scsi 1:0:0:0: Direct-Access AMCC 9650SE-24M DISK 4.10 PQ: 0 ANSI: 5 Oct 17 04:33:43 Storage kernel: sd 1:0:0:0: Attached scsi generic sg1 type 0 Oct 17 04:33:43 Storage kernel: sd 1:0:0:0: [sdb] 244119552 512-byte logical blocks: (124 GB/116 GiB) Oct 17 04:33:43 Storage kernel: scsi 1:0:1:0: Direct-Access AMCC 9650SE-24M DISK 4.10 PQ: 0 ANSI: 5 Oct 17 04:33:43 Storage kernel: sd 1:0:0:0: [sdb] Write Protect is off Oct 17 04:33:43 Storage kernel: sd 1:0:0:0: [sdb] Mode Sense: 23 00 10 00 Oct 17 04:33:43 Storage kernel: sd 1:0:0:0: [sdb] Write cache: disabled, read cache: enabled, supports DPO and FUA Oct 17 04:33:43 Storage kernel: sd 1:0:1:0: [sdc] 7812478976 512-byte logical blocks: (3.99 TB/3.63 TiB) Oct 17 04:33:43 Storage kernel: sd 1:0:1:0: Attached scsi generic sg2 type 0 Oct 17 04:33:43 Storage kernel: sd 1:0:1:0: [sdc] Write Protect is off Oct 17 04:33:43 Storage kernel: sd 1:0:1:0: [sdc] Mode Sense: 23 00 10 00 Oct 17 04:33:43 Storage kernel: scsi 1:0:2:0: Direct-Access AMCC 9650SE-24M DISK 4.10 PQ: 0 ANSI: 5 Oct 17 04:33:43 Storage kernel: sd 1:0:1:0: [sdc] Write cache: disabled, read cache: enabled, supports DPO and FUA Oct 17 04:33:43 Storage kernel: sdb: unknown partition table Oct 17 04:33:43 Storage kernel: sd 1:0:2:0: [sdd] 7812478976 512-byte logical blocks: (3.99 TB/3.63 TiB) Oct 17 04:33:43 Storage kernel: sd 1:0:2:0: Attached scsi generic sg3 type 0 Oct 17 04:33:43 Storage kernel: scsi 1:0:3:0: Direct-Access AMCC 9650SE-24M DISK 4.10 PQ: 0 ANSI: 5 Oct 17 04:33:43 Storage kernel: sdc: unknown partition table Oct 17 04:33:43 Storage kernel: sd 1:0:2:0: [sdd] Write Protect is off Oct 17 04:33:43 Storage kernel: sd 1:0:2:0: [sdd] Mode Sense: 23 00 10 00 Oct 17 04:33:43 Storage kernel: sd 1:0:2:0: [sdd] Write cache: disabled, read cache: enabled, supports DPO and FUA Oct 17 04:33:43 Storage kernel: sd 1:0:0:0: [sdb] Attached SCSI disk Oct 17 04:33:43 Storage kernel: sd 1:0:3:0: [sde] 7812478976 512-byte logical blocks: (3.99 TB/3.63 TiB) Oct 17 04:33:43 Storage kernel: sd 1:0:3:0: Attached scsi generic sg4 type 0 Oct 17 04:33:43 Storage kernel: sd 1:0:1:0: [sdc] Attached SCSI disk Oct 17 04:33:43 Storage kernel: sd 1:0:3:0: [sde] Write Protect is off Oct 17 04:33:43 Storage kernel: sd 1:0:3:0: [sde] Mode Sense: 23 00 10 00 Oct 17 04:33:43 Storage kernel: sd 1:0:3:0: [sde] Write cache: disabled, read cache: enabled, supports DPO and FUA Oct 17 04:33:43 Storage kernel: sdd: unknown partition table Oct 17 04:33:43 Storage kernel: sd 1:0:2:0: [sdd] Attached SCSI disk Oct 17 04:33:43 Storage kernel: sde: unknown partition table Oct 17 04:33:43 Storage kernel: sd 1:0:3:0: [sde] Attached SCSI disk
November 2, 201411 yr Author Thanks Earthworm! Sorry for the slow reply.... been away.. Yes, I too have the latest firmware. I've now upgraded to 6beta10a, but nothing has changed with the performance. It's still crushingly slow at around 5 Mb/s. The frustrating thing is that when I attach the same discs to my old Highpoint PCI-X card (operating ina PCI slot), it rumbles along very nicely at 25-30 Mb/s. Out of sheer frustration, I've gone and bought a new Highpoint 4460 card. PCI-e and 24 ports..... what could possibly go wrong? UNRAID doesn't even see the card! Anyway that's for another thread....
Archived
This topic is now archived and is closed to further replies.