All of lore.kernel.org
 help / color / mirror / Atom feed
From: Steffen Sledz <sledz@dresearch.de>
To: openembedded-devel@lists.openembedded.org
Subject: Re: confusing qemu-native problem
Date: Thu, 13 Aug 2009 09:44:20 +0200	[thread overview]
Message-ID: <4A83C454.3010001@dresearch.de> (raw)
In-Reply-To: <4A7960FC.6030608@dresearch.de>

Steffen Sledz schrieb:
> Sledz, Steffen schrieb:
>> I've seen a very confusing, indetermined behaviour of the qemu-native build in the last time (current dev branch).
>>
>> When calling the following three commands back-to-back the first and the second fail with something like "/bin/sh: .../staging/i686-linux/usr/bin/pod2man: ...bad interpreter: Permission denied" but the third call succeeds (details see below).
>>
>>  bitbake hydraip-hipox-devimage
>>  bitbake qemu-native
>>  bitbake qemu-native
>>
>> In general it looks like the success of the qemu-native is really randomly.
>>
>> Why does the one call of "bitbake qemu-native" fails and the second succeeds?
>> ...
> 
> No more ideas here?
> 
> Such indetermined behaviour is really a horror!

My google searches result in some possible causes for a behaviour like this, but none of them matches my case. :(

1. DOS line endings -> checked: none

2. noexec mount option -> checked: exec

3. acl or user_xattr effects -> checked: not used

Looking into the logs and playing with the devshell i found these interesting points:

> /home/DRESEARCH/production/METABUILD/2009-08-03-15-24-27--hydraip-build--28595/OpenEmbedded/tmp.3/staging/i686-linux/usr/bin/pod2man --help
bash: /home/DRESEARCH/production/METABUILD/2009-08-03-15-24-27--hydraip-build--28595/OpenEmbedded/tmp.3/staging/i686-linux/usr/bin/pod2man: /home/DRESEARCH/production/METABUILD/2009-08-03-15-24-27--hydraip-build--28595: bad interpreter: Permission denied

> head -n1 /home/DRESEARCH/production/METABUILD/2009-08-03-15-24-27--hydraip-build--28595/OpenEmbedded/tmp.3/staging/i686-linux/usr/bin/pod2man
#!/home/DRESEARCH/production/METABUILD/2009-08-03-15-24-27--hydraip-build--28595/OpenEmbedded/tmp.3/staging/i686-linux/usr/bin/perl

> /home/DRESEARCH/production/METABUILD/2009-08-03-15-24-27--hydraip-build--28595/OpenEmbedded/tmp.3/staging/i686-linux/usr/bin/perl --version

This is perl, v5.8.8 built for i686-linux-thread-multi
...

So the perl interpreter seems not to be the problem. Confusing for me is the original error message:

"/home/DRESEARCH/production/METABUILD/2009-08-03-15-24-27--hydraip-build--28595: bad interpreter: Permission denied"

I would expect something like:

"/home/DRESEARCH/production/METABUILD/2009-08-03-15-24-27--hydraip-build--28595/OpenEmbedded/tmp.3/staging/i686-linux/usr/bin/pod2man: bad interpreter: Permission denied"

It looks like someone cut the path to the pod2man script.

I would be really grateful for some help.

Steffen





  reply	other threads:[~2009-08-13  8:01 UTC|newest]

Thread overview: 7+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
     [not found] <AcoU52Sb18YP2SXPQq6rfJAb373LGA==>
2009-08-04  9:38 ` confusing qemu-native problem Sledz, Steffen
2009-08-04 12:29   ` Gregory Machairidis
2009-08-04 12:56     ` Steffen Sledz
2009-08-05 10:37   ` Steffen Sledz
2009-08-13  7:44     ` Steffen Sledz [this message]
2009-08-13 11:32       ` Michael Smith
2009-08-17 12:28         ` Steffen Sledz

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=4A83C454.3010001@dresearch.de \
    --to=sledz@dresearch.de \
    --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 an external index of several public inboxes,
see mirroring instructions on how to clone and mirror
all data and code used by this external index.