[Linux-PowerEdge] SMBIOS bug on Dell PE Gen12 hardware with 3.18 kernel

Luke Bigum luke.bigum at lmax.com
Mon May 18 08:43:41 CDT 2015


Hi G,

Yes, a very unsupported configuration, it's built for performance :-) It's why I can't go through Dell Support. I've just a chat to our performance guys and they aren't running any patches, but the kernel is kernel-xl-3.18.8-100.el6.x86_64, not a stock 3.18 build. Also I got the hardware wrong: this is on FC630 and R730s, not Dell Gen12 like I originally said. Another spanner in the works is that we have SolarFlare cards on the PCI bus as well, and there are known issues with those cards on this hardware at the moment.

As for why I think it is in /sys/devices/platform/dcdbas/, I guessed that the four ENOMEM in this strace match the four log lines in /var/log/messages:

strace -e trace=write,open -f /usr/sbin/smbios-sys-info
...
open("/sys/devices/platform/dcdbas/smi_request", O_WRONLY|O_CREAT|O_TRUNC, 0666) = 3
open("/sys/devices/platform/dcdbas/smi_data_buf_size", O_RDWR|O_CREAT|O_TRUNC, 0666) = 4
write(4, "52\0", 3)                     = -1 ENOMEM (Cannot allocate memory)
write(3, "0", 1)                        = -1 ENODEV (No such device)
write(1, "Service Tag:            ", 24Service Tag:            ) = 24
open("/sys/devices/platform/dcdbas/smi_data_buf_phys_addr", O_RDONLY) = 4
open("/sys/devices/platform/dcdbas/smi_data", O_RDWR|O_CREAT|O_TRUNC, 0666) = 4
write(4, "1IMS\0\0\0\0001ISB\262\0\27\0\v\0\2\0\0\0\0\0\0\0\0\0\0\0\0\0"..., 52) = -1 ENOMEM (Cannot allocate memory)
write(3, "1\0", 2)                      = -1 ENODEV (No such device)
open("/sys/devices/platform/dcdbas/smi_data", O_RDONLY) = 4
write(1, "1TSFZ42\n", 81TSFZ42
)                = 8
open("/sys/devices/platform/dcdbas/smi_request", O_WRONLY|O_CREAT|O_TRUNC, 0666) = 4
open("/sys/devices/platform/dcdbas/smi_data_buf_size", O_RDWR|O_CREAT|O_TRUNC, 0666) = 5
write(5, "52\0", 3)                     = -1 ENOMEM (Cannot allocate memory)
write(4, "0", 1)                        = -1 ENODEV (No such device)
write(1, "Express Service Code:   ", 24Express Service Code:   ) = 24
open("/sys/devices/platform/dcdbas/smi_data_buf_phys_addr", O_RDONLY) = 5
open("/sys/devices/platform/dcdbas/smi_data", O_RDWR|O_CREAT|O_TRUNC, 0666) = 5
write(5, "1IMS\0\0\0\0001ISB\262\0\27\0\v\0\2\0\0\0\0\0\0\0\0\0\0\0\0\0"..., 52) = -1 ENOMEM (Cannot allocate memory)
write(4, "1\0", 2)                      = -1 ENODEV (No such device)
open("/sys/devices/platform/dcdbas/smi_data", O_RDONLY) = 5
write(1, "3978076034\n", 113978076034
)            = 11
open("/sys/devices/platform/dcdbas/smi_request", O_WRONLY|O_CREAT|O_TRUNC, 0666) = 5
open("/sys/devices/platform/dcdbas/smi_data_buf_size", O_RDWR|O_CREAT|O_TRUNC, 0666) = 6
write(6, "52\0", 3)                     = -1 ENOMEM (Cannot allocate memory)
write(5, "0", 1)                        = -1 ENODEV (No such device)
write(1, "Asset Tag:              ", 24Asset Tag:              ) = 24
open("/sys/devices/platform/dcdbas/smi_data_buf_phys_addr", O_RDONLY) = 6
open("/sys/devices/platform/dcdbas/smi_data", O_RDWR|O_CREAT|O_TRUNC, 0666) = 6
write(6, "1IMS\0\0\0\0001ISB\262\0\27\0\v\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0"..., 52) = -1 ENOMEM (Cannot allocate memory)
write(5, "1\0", 2)                      = -1 ENODEV (No such device)
open("/sys/devices/platform/dcdbas/smi_data", O_RDONLY) = 6
write(1, "\n", 1
)                       = 1
open("/sys/devices/platform/dcdbas/smi_request", O_WRONLY|O_CREAT|O_TRUNC, 0666) = 6
open("/sys/devices/platform/dcdbas/smi_data_buf_size", O_RDWR|O_CREAT|O_TRUNC, 0666) = 7
write(7, "137\0", 4)                    = -1 ENOMEM (Cannot allocate memory)
write(6, "0", 1)                        = -1 ENODEV (No such device)
write(1, "Property Ownership Tag: ", 24Property Ownership Tag: ) = 24
open("/sys/devices/platform/dcdbas/smi_data_buf_phys_addr", O_RDONLY) = 7
open("/sys/devices/platform/dcdbas/smi_data", O_RDWR|O_CREAT|O_TRUNC, 0666) = 7
write(7, "1IMS\0\0\0\0001ISB\262\0\27\0\24\0\0\0004\0\0\0\0\0\0\0\0\0\0\0"..., 137) = -1 ENOMEM (Cannot allocate memory)
write(6, "1\0", 2)                      = -1 ENODEV (No such device)
open("/sys/devices/platform/dcdbas/smi_data", O_RDONLY) = 7
write(1, "\n", 1

As far as we can tell it's not hurting anything, our rudimentary SNMP monitoring of the OMSA services is still functioning and we haven't found any of the Dell tools "breaking". Just trying to be helpful :-)

--
Luke Bigum
Senior Systems Engineer

Information Systems

----- Original Message -----
From: "Srinivas G Gowda" <Srinivas_G_Gowda at Dell.com>
To: "luke bigum" <luke.bigum at lmax.com>, linux-poweredge at lists.us.dell.com
Sent: Monday, 18 May, 2015 11:58:32 AM
Subject: RE: [Linux-PowerEdge] SMBIOS bug on Dell PE Gen12 hardware with 3.18	kernel


Hi Luke,
Since 3.18 kernel on a CentOS 6 base OS is not an officially supported/tested configuration you will most likely find issues!!!! 
However Linux-PowerEdge mailing list is a good place to get answers for queries like these.

smbios-sys-info/ smbios-sys-info-lite are utilities from libsmbios.  I tried to reproduce the issue by just running smbios-sys-info/smbios-sys-info-lite on 3.18 upstream kernel branch. But I could not see the reported failure messages !!! 
I am no OMSA expert, but ill try running OMSA 7.4 to see if I catch that failure message. 


Curios to know how you tracked the issue it to “/sys/devices/platform/dcdbas/,” ?   
also do you see the issue if you just run smbios-sys-info/smbios-sys-info-lite. ? 


Thanks,
G
________________________________________
From: linux-poweredge-bounces-Lists On Behalf Of Luke Bigum [luke.bigum at lmax.com]
Sent: Friday, May 15, 2015 4:14 PM
To: linux-poweredge-Lists
Subject: [Linux-PowerEdge] SMBIOS bug on Dell PE Gen12 hardware with 3.18       kernel

Hello,

I want to lodge a low priority bug with the Linux / PowerEdge Engineering team, but I can't find how to do that on any of Dell's Linux pages, so I'm posting here.

We're running a 3.18 kernel on a CentOS 6 base OS and OMSA 7.4 installed. I'm periodically getting this error message on our hardware running this kernel:

  kernel: Allocating domain for dcdbas failed

That error messages comes from intel-iommu.c (http://lxr.free-electrons.com/source/drivers/iommu/intel-iommu.c?v=3.18#L2944).

I've tracked down to what triggers it, it's the OMSA init scripts that call either smbios-sys-info or smbios-sys-info-lite, which are doing stuff inside /sys/devices/platform/dcdbas/, I'm not sure what specifically.

Does anyone know who'd be interested in this bug/error?

--
Luke Bigum
Senior Systems Engineer

Information Systems
---

LMAX Exchange, Yellow Building, 1A Nicholas Road, London W11 4AN
http://www.LMAX.com/

---
#1 Fastest Growing Tech Company in UK - Sunday Times Tech Track 100 (2014)

Awards
2015 Best FX Trading Venue - ECN/MTF - WSL Institutional Trading Awards
2014 Best Margin Sector Platform - Profit & Loss Readers' Choice Awards
2014 Best FX Trading Venue - ECN/MTF - WSL Institutional Trading Awards
2014 Best Infrastructure/Technology Initiative - WSL Institutional Trading Awards
2013 #15 Fastest Growing Tech Company in UK - Sunday Times Tech Track 100
2013 Best Overall Testing Project - The European Software Testing Awards
2013 Best Margin Sector Platform - Profit & Loss Readers' Choice Awards
2013 Best FX Trading Platform - ECN/MTF - WSL Institutional Trading Awards
2013 Best Executing Venue - Forex Magnates Awards
2011 Best Trading System - Financial Sector Technology Awards
2011 Innovative Programming Framework - Oracle Duke's Choice Awards
---

FX and CFDs are leveraged products that can result in
losses exceeding your deposit. They are not suitable
for everyone so please ensure you fully understand
the risks involved.

This message and its attachments are confidential,
may not be disclosed or used by any person other
than the addressee and are intended only for the
named recipient(s). This message is not intended for
any recipient(s) who based on their nationality,
place of business, domicile or for any other
reason, is/are subject to local laws or regulations
which prohibit the provision of such products and
services. This message is subject to the terms at
http://www.lmax.com/pdf/general-disclaimers.pdf
however if you cannot access these, please notify
us by replying to this email and we will send you
the terms. If you are not the intended recipient,
please notify the sender immediately and delete any
copies of this message.

LMAX Exchange is the trading name of LMAX Limited. LMAX
Limited operates a multilateral trading facility. LMAX
Limited is authorised and regulated by the Financial
Conduct Authority (firm registration number 509778)
and is a company registered in England and Wales
(number 6505809).

LMAX Hong Kong Limited is a wholly-owned subsidiary
of LMAX Limited. LMAX Hong Kong is licensed by the
Securities and Futures Commission in Hong Kong to
conduct Type 3 (leveraged foreign exchange trading)
regulated activity with CE Number BDV088.

_______________________________________________
Linux-PowerEdge mailing list
Linux-PowerEdge at dell.com
https://lists.us.dell.com/mailman/listinfo/linux-poweredge
---

LMAX Exchange, Yellow Building, 1A Nicholas Road, London W11 4AN
http://www.LMAX.com/

---
#1 Fastest Growing Tech Company in UK - Sunday Times Tech Track 100 (2014)

Awards
2015 Best FX Trading Venue - ECN/MTF - WSL Institutional Trading Awards
2014 Best Margin Sector Platform - Profit & Loss Readers' Choice Awards
2014 Best FX Trading Venue - ECN/MTF - WSL Institutional Trading Awards
2014 Best Infrastructure/Technology Initiative - WSL Institutional Trading Awards
2013 #15 Fastest Growing Tech Company in UK - Sunday Times Tech Track 100
2013 Best Overall Testing Project - The European Software Testing Awards
2013 Best Margin Sector Platform - Profit & Loss Readers' Choice Awards
2013 Best FX Trading Platform - ECN/MTF - WSL Institutional Trading Awards
2013 Best Executing Venue - Forex Magnates Awards
2011 Best Trading System - Financial Sector Technology Awards
2011 Innovative Programming Framework - Oracle Duke's Choice Awards
---

FX and CFDs are leveraged products that can result in
losses exceeding your deposit. They are not suitable
for everyone so please ensure you fully understand
the risks involved.

This message and its attachments are confidential,
may not be disclosed or used by any person other
than the addressee and are intended only for the
named recipient(s). This message is not intended for
any recipient(s) who based on their nationality,
place of business, domicile or for any other
reason, is/are subject to local laws or regulations
which prohibit the provision of such products and
services. This message is subject to the terms at
http://www.lmax.com/pdf/general-disclaimers.pdf
however if you cannot access these, please notify
us by replying to this email and we will send you
the terms. If you are not the intended recipient,
please notify the sender immediately and delete any
copies of this message.

LMAX Exchange is the trading name of LMAX Limited. LMAX
Limited operates a multilateral trading facility. LMAX
Limited is authorised and regulated by the Financial
Conduct Authority (firm registration number 509778)
and is a company registered in England and Wales
(number 6505809).

LMAX Hong Kong Limited is a wholly-owned subsidiary
of LMAX Limited. LMAX Hong Kong is licensed by the
Securities and Futures Commission in Hong Kong to
conduct Type 3 (leveraged foreign exchange trading)
regulated activity with CE Number BDV088.



More information about the Linux-PowerEdge mailing list