Openembedded Devel Discussions
 help / color / mirror / Atom feed
From: Jack Mitchell <ml@communistcode.co.uk>
To: openembedded-devel@lists.openembedded.org
Subject: Re: [meta-oe] net-snmp config help
Date: Wed, 04 Jul 2012 09:25:21 +0100	[thread overview]
Message-ID: <4FF3FDF1.5010002@communistcode.co.uk> (raw)
In-Reply-To: <4FF1B901.6000306@communistcode.co.uk>

On 02/07/12 16:06, Jack Mitchell wrote:
> On 02/07/12 15:45, Jack Mitchell wrote:
>> Ok, so I managed to get net-snmp compiling however I now have an 
>> issue where the configure stage is picking up libpci and using that 
>> to configure net-snmp to look for pci devices on startup which my 
>> beaglebone doesn't have
>>
>>    root@beaglebone:~# snmpd
>>    pcilib: Cannot open /proc/bus/pci
>>    pcilib: Cannot find any working access method.
>>    root@beaglebone:~#
>>
>> Off I went and found a solution[1] which states that I should force 
>> the PCI_* defines to off *after* the configuration stage. So I go 
>> into my work directory and alter includes/net-snmp/net-snmp-config.h:
>>
>>    /* define if you have pci_lookup_name() */
>>    #undef HAVE_PCI_LOOKUP_NAME
>>
>>    /* Define to 1 if you have the <pci/pci.h> header file. */
>>    #undef HAVE_PCI_PCI_H
>>
>> Then compile with:
>>
>>     bitbake net-snmp -c compile -f
>>
>> Then
>>
>>     bitbake core-image-minimal
>>
>> Which does net-snmp do_install and then builds the rootfs. However 
>> this makes no difference and it still fails with the same error 
>> message on start. Is my work flow for this wrong, or is it 
>> re-configuring, I'm at a bit of a loss of what to do now?
>>
>> I suppose the correct procedure would be to find out why pcilibs is 
>> being detected and configured, then fix that. Could anyone point me 
>> in the right direction?
>>
>> Regards,
>> Jack.
>>
>> [1] 
>> http://stackoverflow.com/questions/9470494/snmpd-cannot-open-proc-bus-pci 
>>
>>
>
> It seems that udev is forcing PCI down every ones throats, is this 
> necessary, can it be built without? There seems to be a lot of extra 
> configuration in there to make it happen...
>
> Onward ho!
>

As a stop gap till I manage to get to the bottom of this properly, doing:

     bitbake core-image-minimal

to get the initial image and everything built properly

     bitbake -c cleansstate pciutils

To remove pci.h from the sysroot

     bitbake -c cleansstate net-snmp

to remove the net-snmp with pci config enabled

     bitbake net-snmp

to rebuild net-snmp without pci config enabled (due to it not finding pci.h)

     bitbake core-image-minimal

pciutils gets rebuilt, net-snmp stays configured without pci and 
everybody is happy (apart from the deterministic build goddess).

Regards,

-- 

   Jack Mitchell (jack@embed.me.uk)
   Embedded Systems Engineer
   http://www.embed.me.uk

--




      reply	other threads:[~2012-07-04  8:35 UTC|newest]

Thread overview: 3+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2012-07-02 14:45 [meta-oe] net-snmp config help Jack Mitchell
2012-07-02 15:06 ` Jack Mitchell
2012-07-04  8:25   ` Jack Mitchell [this message]

Reply instructions:

You may reply publicly to this message via plain-text email
using any one of the following methods:

* Save the following mbox file, import it into your mail client,
  and reply-to-all from there: mbox

  Avoid top-posting and favor interleaved quoting:
  https://en.wikipedia.org/wiki/Posting_style#Interleaved_style

* Reply using the --to, --cc, and --in-reply-to
  switches of git-send-email(1):

  git send-email \
    --in-reply-to=4FF3FDF1.5010002@communistcode.co.uk \
    --to=ml@communistcode.co.uk \
    --cc=openembedded-devel@lists.openembedded.org \
    /path/to/YOUR_REPLY

  https://kernel.org/pub/software/scm/git/docs/git-send-email.html

* If your mail client supports setting the In-Reply-To header
  via mailto: links, try the mailto: link
Be sure your reply has a Subject: header at the top and a blank line before the message body.
This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox