Did you mean 5K3000s? I couldn't find a 3k5 model when I checked newegg.
![[H]ard|Forum](/styles/hardforum/xenforo/logo_dark.png){kind=logo}
Marvel AOC-SASLP-MV8 - 8 Sata PCI-E 4x non-raid controller card
- Thread starter Romir
- Start date
Did you mean 5K3000s? I couldn't find a 3k5 model when I checked newegg.
OmegaAvenger
2[H]4U
- Joined
- Sep 25, 2007
- Messages
- 3,240
Yeah, that was a typo. my bad.
Hi all,
I'm using this card for about half a year now under Ubuntu, unfotunate, it still doesn't work like supposed to, tried various kernel versions up to the latest 3.x.x release.
Till now 2.6.37-rc4 seems to be the most stable kernel for this card, however when performing some heavy IO, transfers keep pausing for some seconds and dmesg spits out unplug errors and drivestatus errors randomly over all drives. Don't have to explain this is quitte annoying.
Do any of you guys have got rid of the unplug notices/pauses under Linux, which kernel do you recommend?
Really appreciate your input.
I'm using this card for about half a year now under Ubuntu, unfotunate, it still doesn't work like supposed to, tried various kernel versions up to the latest 3.x.x release.
Till now 2.6.37-rc4 seems to be the most stable kernel for this card, however when performing some heavy IO, transfers keep pausing for some seconds and dmesg spits out unplug errors and drivestatus errors randomly over all drives. Don't have to explain this is quitte annoying.
Do any of you guys have got rid of the unplug notices/pauses under Linux, which kernel do you recommend?
Really appreciate your input.
hi mate,
I am using this card under ubuntu server stable and am currently using 2.6.38-8-server.
I tried 2.6.39.1, as the diff showed a small change but it caused the server to crash under load
2.6.38-8-server still produces the mvsas errors and when it does this it causes a delayed write. I currently have 7 1.5TB drives in RAID6 (software).
i'm getting sick of this and am now considering sourcing an alternative card or abandoning the whole idea
I am using this card under ubuntu server stable and am currently using 2.6.38-8-server.
I tried 2.6.39.1, as the diff showed a small change but it caused the server to crash under load
2.6.38-8-server still produces the mvsas errors and when it does this it causes a delayed write. I currently have 7 1.5TB drives in RAID6 (software).
i'm getting sick of this and am now considering sourcing an alternative card or abandoning the whole idea
Unknown-One
[H]F Junkie
- Joined
- Mar 5, 2005
- Messages
- 8,909
Heads up, it looks like this card has issues with the Samsung EcoGreen F4's as well.
I just installed two of these drives in my server, attached to the AOC-SASLP-MV8, and one of them drops every time I try to transfer data to it (the drives aren't even in an array of any kind). Plugged the drives directly into the motherboard instead, and now they work fine...
Is there a firmware update for this SAS card that might help? Failing that, is there a different SAS card known to handle "advanced format" drives like the EcoGreens?
I just installed two of these drives in my server, attached to the AOC-SASLP-MV8, and one of them drops every time I try to transfer data to it (the drives aren't even in an array of any kind). Plugged the drives directly into the motherboard instead, and now they work fine...
Is there a firmware update for this SAS card that might help? Failing that, is there a different SAS card known to handle "advanced format" drives like the EcoGreens?
There was a firmware update, mentioned previously in the thread but some people said avoid it so dunno.
I am using 1.5TB drives but still getting the dropouts / stalls.
I am using 1.5TB drives but still getting the dropouts / stalls.
Code:
[637995.020022] /build/buildd/linux-2.6.38/drivers/scsi/mvsas/mv_sas.c 1703:<7>mv_abort_task() mvi=ffff880113f80000 task=ffff880105f34c40 slot=ffff880113fa44e0 slot_idx=x0
[637995.020028] /build/buildd/linux-2.6.38/drivers/scsi/mvsas/mv_sas.c 1632:mvs_query_task:rc= 5
[637995.020053] /build/buildd/linux-2.6.38/drivers/scsi/mvsas/mv_sas.c 2083:port 3 ctrl sts=0x89800.
[637995.020055] /build/buildd/linux-2.6.38/drivers/scsi/mvsas/mv_sas.c 2085:Port 3 irq sts = 0x1001
[637995.020060] /build/buildd/linux-2.6.38/drivers/scsi/mvsas/mv_sas.c 2111:phy3 Unplug Notice
[637995.030084] /build/buildd/linux-2.6.38/drivers/scsi/mvsas/mv_sas.c 2083:port 3 ctrl sts=0x199800.
[637995.030086] /build/buildd/linux-2.6.38/drivers/scsi/mvsas/mv_sas.c 2085:Port 3 irq sts = 0x1081
[637995.042806] /build/buildd/linux-2.6.38/drivers/scsi/mvsas/mv_sas.c 2083:port 3 ctrl sts=0x199800.
[637995.042809] /build/buildd/linux-2.6.38/drivers/scsi/mvsas/mv_sas.c 2085:Port 3 irq sts = 0x10000
[637995.042812] /build/buildd/linux-2.6.38/drivers/scsi/mvsas/mv_sas.c 2138:notify plug in on phy[3]
[637995.052791] /build/buildd/linux-2.6.38/drivers/scsi/mvsas/mv_sas.c 1224:port 3 attach dev info is 0
[637995.052791] /build/buildd/linux-2.6.38/drivers/scsi/mvsas/mv_sas.c 1226:port 3 attach sas addr is 3
[637995.052791] /build/buildd/linux-2.6.38/drivers/scsi/mvsas/mv_sas.c 378:phy 3 byte dmaded.
[637997.240008] /build/buildd/linux-2.6.38/drivers/scsi/mvsas/mv_sas.c 1586:mvs_I_T_nexus_reset for device[2]:rc= 0
[637997.240016] ata9: translated ATA stat/err 0x01/04 to SCSI SK/ASC/ASCQ 0xb/00/00
[637997.240113] ata9: status=0x01 { Error }
[637997.240115] ata9: error=0x04 { DriveStatusError }treadstone
Gawd
- Joined
- Jan 25, 2010
- Messages
- 567
I have 3 AOC-SASLP-MV8 HBAs in one of my servers running WHS2011. The odd thing was that as soon as I added more than 2 drives to the second card (all ports on the first controller and about 6 ports on the third controller are in use), the system would crash with a BSOD... I reinstalled the OS a few times since I had messed around with a lot of different drivers figuring it had something to do with that, but no dice... it just kept crashing. Another odd thing was that when I plugged/connected more than two HDDs to the second controller and then booted, the windows driver (version 3.1.0.22) would actually access all the drives in sequence, however once windows finished booting, only 2 drives would show up or if they did show up, the system would crash with a BSOD as soon as I tried accessing any of the discs connected to the second controller. I had no issues accessing any of the drives connected to controller 1 and 3! I even thought that the actual controller might be faulty, but when I pulled/disconnected the drives from the first and third controller, I could add all 8 drives to the second controller and the system worked just fine, so it obviously was a driver issue. It always crashed with a IRQ_EQUAL_OR_LESS (or something similar) kind of message...
Then I stumbled onto a new driver release for the AOC-SASLP-MV8 version 4.0.0.1200... tried it and now everything works like a charm!
Also, I use a supermicro X9SCM-F motherboard for this server and in the BIOS I disabled the AOC-SASLP-MV8 BIOS boot so the windows driver actually initializes the controller. This speeds up the boot process considerably!
Then I stumbled onto a new driver release for the AOC-SASLP-MV8 version 4.0.0.1200... tried it and now everything works like a charm!
Also, I use a supermicro X9SCM-F motherboard for this server and in the BIOS I disabled the AOC-SASLP-MV8 BIOS boot so the windows driver actually initializes the controller. This speeds up the boot process considerably!
I am running this card for quite a while
OS: debian squeeze x64 (with custom kernel, see below)
setup: 5x seagate ST31500341AS 1.5TB in soft raid 6
kernels:
3.0.3
seen no real issues yet, but running it only a couple of weeks
2.6.39.1
I got a complete system freeze very rarely (~ once every 2 months), then I see also errors in my log about 1 disk so called misbehaving, not always the same disk ...
2.6.37.2
similar to the above, but more frequent errors.
for now I can live with it, but still strongly advise against buying this card!
OS: debian squeeze x64 (with custom kernel, see below)
setup: 5x seagate ST31500341AS 1.5TB in soft raid 6
kernels:
3.0.3
seen no real issues yet, but running it only a couple of weeks
2.6.39.1
I got a complete system freeze very rarely (~ once every 2 months), then I see also errors in my log about 1 disk so called misbehaving, not always the same disk ...
2.6.37.2
similar to the above, but more frequent errors.
for now I can live with it, but still strongly advise against buying this card!
with 2.6.39.1 I think so, but as said very seldom and my complete system freezes with those errors so I need to hard reset my machine (and as far as I can see the logging is also broken at that time cause I don't find it in messages back)
My idea (but without logs it's hard to say) this kind of errors occured after long idle time (not sure if for example disks go in lower power states or something, my disk are non-green or eco so I think not ...)
as far as I can see for 3.0.3 running since Aug 25:
I saw first error at the specified date like this:
Aug 31 20:08:48 fileserver kernel: ata10: hard resetting link
Aug 31 20:08:51 fileserver kernel: ata10.00: configured for UDMA/133
Aug 31 20:08:51 fileserver kernel: ata10: EH complete
and this has happened 2 or 3 times up to now
but this kind of errors are non-fatal (saw them too with 2.6.39.1 from time to time, but there is no freeze or any time-out noticeable at my samba shares for example ...)
no other errors yet on 3.0.3 which are fatal ... *Knocking on wood*
My idea (but without logs it's hard to say) this kind of errors occured after long idle time (not sure if for example disks go in lower power states or something, my disk are non-green or eco so I think not ...)
as far as I can see for 3.0.3 running since Aug 25:
I saw first error at the specified date like this:
Aug 31 20:08:48 fileserver kernel: ata10: hard resetting link
Aug 31 20:08:51 fileserver kernel: ata10.00: configured for UDMA/133
Aug 31 20:08:51 fileserver kernel: ata10: EH complete
and this has happened 2 or 3 times up to now
but this kind of errors are non-fatal (saw them too with 2.6.39.1 from time to time, but there is no freeze or any time-out noticeable at my samba shares for example ...)
no other errors yet on 3.0.3 which are fatal ... *Knocking on wood*
I got home today to find all of my drives hooked up to this card had gone missing in my WHS v1. I rebooted and they all detected fine. The Home Server Smart add-in detected two drives with potential issues. Would this cause the card to drop out completely though?
Everything was working fine until I started trying to access my video share and then the drives all dropped out again. Up until today everything for the most part has been fine. I recently swapped an older Samsung 500GB for a Hitachi 2TB 5K3000, but the drive doesn't show any errors.
Anyone have any ideas? I don't know if my hardware is going bad or if it is an issue with the two drives.
Everything was working fine until I started trying to access my video share and then the drives all dropped out again. Up until today everything for the most part has been fine. I recently swapped an older Samsung 500GB for a Hitachi 2TB 5K3000, but the drive doesn't show any errors.
Anyone have any ideas? I don't know if my hardware is going bad or if it is an issue with the two drives.
I have two AOC-SASLP-MV8 cards, bought a year or so ago when they were first available.
I've upgraded firmware on both to 3.1.0.21, but a strange thing has happened - one now thinks it's a raid card, the other the standard sata card.
Has anyone else seen this? Is there a fix?
Sorry if it's buried further back in the thread, i've searched it but not found any hints...
I've upgraded firmware on both to 3.1.0.21, but a strange thing has happened - one now thinks it's a raid card, the other the standard sata card.
Has anyone else seen this? Is there a fix?
Sorry if it's buried further back in the thread, i've searched it but not found any hints...
dogbert_2001
Limp Gawd
- Joined
- Nov 9, 2007
- Messages
- 130
My guess is the larger size of the 3.1.0.21 means there's only room for one of your cards to load the option rom and be recognized as fully functional.
I installed an AOC-SAS2LP-MV8 in my server and tried Ubuntu 13.04 (3.8 Kernel) and 13.10 (3.11 Kernel) but no drives attached to it get recognised but the adapter itself is listed when running lspci. Executing "modprobe mvsas" had no influence at all.
I tried this method:
http://hardforum.com/showpost.php?p=1035759359&postcount=439
and this method:
http://hardforum.com/showpost.php?p=1034069470&postcount=189
But I'm stuck on "make mvsas". It does not compile because of uninitialised variables and whatnot...
I'm not sure if I followed all steps correctly.
So before I put any more time and effort in it I'd like to know:
Has anyone running a Supermicro AOC-SAS2LP-MV8 with Ubuntu 12.xx or 13.xx and are there some drawbacks or problems ?
I tried this method:
http://hardforum.com/showpost.php?p=1035759359&postcount=439
and this method:
http://hardforum.com/showpost.php?p=1034069470&postcount=189
But I'm stuck on "make mvsas". It does not compile because of uninitialised variables and whatnot...
I'm not sure if I followed all steps correctly.
So before I put any more time and effort in it I'd like to know:
Has anyone running a Supermicro AOC-SAS2LP-MV8 with Ubuntu 12.xx or 13.xx and are there some drawbacks or problems ?
This patch worked for me on the AOC-SAS2LP-MV8 on Ubuntu 13.10
http://serverfault.com/questions/539696/ubuntu-12-04-server-and-supermicro-aoc-sas2lp-mv8
http://lists.debian.org/debian-kernel/2013/09/msg00184.html
Reported-by: Matt Taggart <[email protected]>
Tested-by: Matt Taggart <[email protected]>
Signed-off-by: Ben Hutchings <[email protected]>
---
drivers/scsi/mvsas/mv_init.c | 9 +++++++++
1 file changed, 9 insertions(+)
diff --git a/drivers/scsi/mvsas/mv_init.c b/drivers/scsi/mvsas/mv_init.c
index 7b7381d..83fa5f8 100644
--- a/drivers/scsi/mvsas/mv_init.c
+++ b/drivers/scsi/mvsas/mv_init.c
@@ -729,6 +729,15 @@ static struct pci_device_id mvs_pci_table[] = {
.class_mask = 0,
.driver_data = chip_9485,
},
+ {
+ .vendor = PCI_VENDOR_ID_MARVELL_EXT,
+ .device = 0x9485,
+ .subvendor = PCI_ANY_ID,
+ .subdevice = 0x9485,
+ .class = 0,
+ .class_mask = 0,
+ .driver_data = chip_9485,
+ },
{ PCI_VDEVICE(OCZ, 0x1021), chip_9485}, /* OCZ RevoDrive3 */
{ PCI_VDEVICE(OCZ, 0x1022), chip_9485}, /* OCZ RevoDrive3/zDriveR4 (exact model unknown) */
{ PCI_VDEVICE(OCZ, 0x1040), chip_9485}, /* OCZ RevoDrive3/zDriveR4 (exact model unknown) */
http://serverfault.com/questions/539696/ubuntu-12-04-server-and-supermicro-aoc-sas2lp-mv8
http://lists.debian.org/debian-kernel/2013/09/msg00184.html
Reported-by: Matt Taggart <[email protected]>
Tested-by: Matt Taggart <[email protected]>
Signed-off-by: Ben Hutchings <[email protected]>
---
drivers/scsi/mvsas/mv_init.c | 9 +++++++++
1 file changed, 9 insertions(+)
diff --git a/drivers/scsi/mvsas/mv_init.c b/drivers/scsi/mvsas/mv_init.c
index 7b7381d..83fa5f8 100644
--- a/drivers/scsi/mvsas/mv_init.c
+++ b/drivers/scsi/mvsas/mv_init.c
@@ -729,6 +729,15 @@ static struct pci_device_id mvs_pci_table[] = {
.class_mask = 0,
.driver_data = chip_9485,
},
+ {
+ .vendor = PCI_VENDOR_ID_MARVELL_EXT,
+ .device = 0x9485,
+ .subvendor = PCI_ANY_ID,
+ .subdevice = 0x9485,
+ .class = 0,
+ .class_mask = 0,
+ .driver_data = chip_9485,
+ },
{ PCI_VDEVICE(OCZ, 0x1021), chip_9485}, /* OCZ RevoDrive3 */
{ PCI_VDEVICE(OCZ, 0x1022), chip_9485}, /* OCZ RevoDrive3/zDriveR4 (exact model unknown) */
{ PCI_VDEVICE(OCZ, 0x1040), chip_9485}, /* OCZ RevoDrive3/zDriveR4 (exact model unknown) */
I realize that you asked this forever ago (sorry), but according to the reviews on NewEgg, this card does not support drives larger than 2TB.